Citation
Examining the extent of adherence and the barriers to adherence among HIV-infected children on antiretroviral therapy : a combined qualitative and quantitative approach

Material Information

Title:
Examining the extent of adherence and the barriers to adherence among HIV-infected children on antiretroviral therapy : a combined qualitative and quantitative approach
Cover title:
Examining the extent of poor adherence and the barriers to adherence among HIV-infected children on antiretroviral therapy
Creator:
Marhefka, Stephanie Lynn, 1975-
Publication Date:
Language:
English
Physical Description:
viii, 114 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Arts ( jstor )
Caregivers ( jstor )
Child psychology ( jstor )
Chronic conditions ( jstor )
Diseases ( jstor )
Dosage ( jstor )
HIV ( jstor )
Medications ( jstor )
Pediatrics ( jstor )
Pharmacies ( jstor )
Caregivers ( mesh )
HIV Infections -- drug therapy ( mesh )
Health Knowledge, Attitudes, Practice ( mesh )
Patient Compliance ( mesh )
Treatment Refusal ( mesh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 85-95).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Stephanie Lynn Marhefka.

Record Information

Source Institution:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (dissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
53949592 ( OCLC )
ocm53949592

Downloads

This item has the following downloads:


Full Text










EXAMINING THE EXTENT OF ADHERENCE AND THE BARRIERS TO ADHERENCE AMONG HIV-INFECTED CHILDREN ON ANTIRETROVIRAL THERAPY: A COMBINED QUALITATIVE AND QUANTITATIVE APPROACH




















BY

STEPHANIE LYNN MARHEFKA











A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2002













ACKNOWLEDGEMENTS

I am grateful for so many people who contributed to the successful completion of my doctoral dissertation. I would like to thank my doctoral committee members for their guidance throughout this very difficult and overwhelming process. I am particularly grateful for the help of my chair, Dr. Jim Rodrigue, who agreed to chair my dissertation even when it did not fit neatly into his silo. I also owe special thanks to Dr. Sam Sears, who chaired my master's thesis and continued to support and encourage me after I made the very difficult decision to change my area of concentration to pediatrics and subsequently change my primary mentor.

This dissertation would not have been possible without the support and

collaboration of directors and staff of the pediatric immunology clinics at the University of Florida, Gainesville, the University of Florida, Jacksonville, and the University of Maryland at Baltimore. I owe special thanks to Dr. John Sleasman, Dr. Judy Lew, Carla Duff, Dr. Mobeen Rathmore, Dr. Lauriann Sanders, Dr. Vicki Tepper, Dr. John Farley, Dr. Douglas Watson, Dr. Peter Vink, Marie Parks, and Angelo Seda, for allowing me to invade their clinics, helping me to make sense of hectic clinic environments, and teaching me about pediatric HIV clinical practice. Doctors Vicki Tepper and John Farley were especially monumental in mentoring me through the University of Maryland Institutional Review Board processes and providing me time and resources for completing my dissertation during internship and the first months of my fellowship.




ii








I also wish to extend my gratitude towards the many individuals who helped me with my literature review and data collection. Numerous research assistants retrieved articles, collected clinical data, and spent many hours conducting telephone interviews for this study. Joe Palmer was particularly gracious about working on weekends, putting in numerous hours the week before my proposal, and driving to Jacksonville weekly for data collection. Hanna Frost was my most enduring research assistant, and I am quite grateful for her continued efforts, especially during my internship year. Bill Weisner was not a research assistant per se, but he deserves my appreciation, nevertheless, as he graciously acted as a substitute research assistant when deadlines drew near.

Several individuals deserve special thanks for reviewing my manuscript at various stages of the process. Despite numerous other commitments, Dr. Vicki Tepper was kind enough to read my draft before I submitted it to my chair. Dr. Sylvie Lombardo was also extremely helpful with my result section--Merci! Jack Rusher graciously read through my discussion and helped to improve my grammar and tense consistency, and Dr. Jennifer Brown remained patient as I directed many dissertation questions her way.

Finally, I wish to thank my friends and family for their continued support and

encouragement throughout my tenure as a graduate student. I will always be grateful for the ways in which my sweetheart, Jack Rusher, encouraged me to persist, despite seemingly insurmountable hardships, and the way that he believed in me even when I struggled to believe in myself. I will also never forget that Brandy Werba, Jane Querido, and Karen Bearss stood in the bathroom with me as I cried on the day I reached my first substantial dissertation roadblock. Similarly, I will remember the way that my sister Suzanne encouraged me over the telephone to keep working. Moreover, I will not forget




111








the sanctuary that was the Gainesville swing dancing community, and how my fellow dancers kept my feet moving. I know that most of my friends and family members cannot begin to understand the processes of graduate school and dissertation completion; I am grateful to them for trying to understand, and supporting me even when they did not understand. With some confidence I can finally say that I am done with school!










































iv














TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS .............................................................. ii

A B ST R A C T ......................................................................... .. vii

IN TR O D U CTIO N ........................................................................ 1

REVIEW OF THE LITERATURE....................................................... 4

Prevalence of Poor Adherence among Children with Chronic Illness ........ 4 D efining A dherence................................................................ 5
The Implications of Poor Adherence........................................... 6
Assessment of Medication Adherence......................................... 7
Regimen Knowledge..................................................... 8
Pharm acy R efill............................................................. 12
Self-R eport ................................................................... 13
Assessment of Barriers .......................................................... 16
The Human Immunodeficiency Virus: A New Chronic Illness ........... 18
Implications of Poor Adherence to ART ......... .............. 21
Assessment of Adherence Among Children with HIV ........... 25
Regimen knowledge ........................................................ 25
Pharmacy refill and self-report ........ .................... .... 28
Barriers to ART Adherence ................................................ 30

OVERVIEW AND HYPOTHESES .................................................. 37

METHOD................................................................ 40

Participants .................................................................... .. .. 40
Procedure ....................................................................... .. 41
M easures ........................................................................ .. 44
Adherence Interview-HIV........................................................ 44
Demographic Questionnaire............................ .............. 45
Telephone Contact Form ................................................... 46
Medical Record Report Form ............................................ 46
24-Hour Recall Interview................................................... 46
Pharmacy Refill History.................................................... 48

DATA ANALYSES ..................... ........................... 49



v








R E SU L T S ..................................................................................... 52

Caregiver Knowledge of the Prescribed Regimen ............................. 52
Relationship between Knowledge and Typical Adherence Behavior ........ 53 Utility of Medication Display Card .............................................. 56
A dherence R ates ..................................................................... 56
Relationship between Adherence Measures ...................................... 57
Relationship between Adherence and Disease Severity ........................ 57
Relationship between Adherence and Knowledge ............................. 58
Relationship between Adherence and Prescribed Dosing Frequency ........ 58 Relationship between Adherence and Medication Formulation........... 60
Barriers to Adherence ........................................................... 60
Barriers Reported with the AI-HIV versus the 24RI........................ 61

DISCUSSION............................................................. 64

Caregiver Knowledge of the Prescribed Regimen .................................. 64
Adherence Rates .................................................................... 66
Barriers to Adherence ............................................................... 72
Im plications ....................................................................... 74
Study Strengths ...................................................................... 79
Study Limitations ................................................................... 80
Future D irections .................................................................... 82

R E FER EN C E S .............................................................................. 85

APPENDIX

A ELIGIBILITY CHECKLIST ..................................................... .. 96

B ADHERENCE INTERVIEW-HIV.................................................. 97

C DEMOGRAPHIC QUESTIONNAIRE ........................................... 103

D TELEPHONE CONTACT FORM .................................................. 107

E MEDICAL RECORD REPORT FORM.......................................... 108

F 24-HOUR RECALL INTERVIEW.................................................. 112

BIOGRAPHICAL SKETCH ............................................................ 114









vi













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy


EXAMINING THE EXTENT OF POOR ADHERENCE AND THE BARRIERS TO ADHERENCE AMONG HIV-INFECTED CHILDREN ON ANTIRETROVIRAL THERAPY: A COMBINED QUALITATIVE AND QUANTITATIVE APPROACH By

Stephanie L. Marhefka

December 2002

Chairperson: James R. Rodrigue, Ph.D. Major Department: Clinical and Health Psychology

This study assessed adherence to antiretroviral medications among 63 children with HIV infection. Trained interviewers administered the Adherence Interview-HIV (AI-HIV) to caregivers of infected children to assess typical medication regimen behaviors and knowledge of the target child's prescribed regimen. Next, demographic information was collected via interview and a medical record review was conducted. Two weeks following the initial interview, trained interviews began the 24-Hour Recall Interview (24RI) procedure with each caregiver; caregivers were then interviewed over the telephone two additional times within the next two weeks. Three months after the initial interview, pharmacy refill data were collected.

Results show that 33% of caregivers failed to correctly identify at least one of

their child's medication names, 31% of caregivers failed to correctly identify the dosage for at least one medication, and half of the caregivers failed to correctly identify the



vii







specific dietary requirements for at least one medication. Adherence to medicationtaking frequency varied by assessment modality. Results of the 24RI suggest that 87% of children were at least 80% adherent; however, results of the pharmacy refill history suggest that only 46% of children were at least 80% adherent. When adherence to the prescribed interval was examined with the 24RI, 47.3% of doses given were deviant from the prescribed interval by at least one hour, while 17.3% of doses given were deviant by at least 2 hours. Results of the 24RI also suggest that the average child was adherent to medication-specific dietary requirements 75% of the time. Twenty-two different barriers were reported and fall into three general categories: a) medication-specific attributes, such as the size or taste of pills; b) problems with scheduling or routine; or c) problems with the child resisting, refusing, or hiding the medication.

This study supports the assertion that a significant proportion of HIV-infected children are not receiving their medication exactly as it is prescribed. The study also suggests that a large portion of caregivers lack knowledge about their child's HIV medication regimen. Implications for adherence-related interventions are discussed.





















viii













INTRODUCTION

Estimates suggest that 18% of children in the United States have a chronic illness (Newacheck, McManus, Fox, Hung, & Halfon, 2000). Among children with chronic illness, the prevalence of poor adherence to medical regimens varies according to the study sample, the specifics of regimen requirements, the assessment of adherence, and the criteria used to classify children as adherent or poorly adherent (Rapoff, 1999). Nevertheless, estimates suggest that 21-52% of children with chronic illness may not fully adhere to their regimens (Alessandro, Vincenzo, Marco, Marcello, & Enrica, 1994; Conley & Salvatierra, 1996; Ettenger et al., 1991; Festa, Tamaroff, Chasalow, & Lanzowsky, 1992; Meyers, Thompson, & Weiland, 1996; Schoni, Horak, & Nikolaizik, 1995; Weisberg-Benchell et al., 1995). Adverse effects of poor adherence include health care costs, clinical decision-making, conclusions drawn from clinical trials, morbidity and mortality (Rapoff, 1999).

The Human Immunodeficiency Virus (HIV) is a chronic illness that affects children as well as adults. Medication regimens for HIV-infected children require multiple medications to be taken at specific times throughout the day (Scott & Sleasman, 1999). The pills are often large and difficult to swallow, and many of the liquid medications have a bad taste. For these and other reasons, families may have difficulty adhering to pediatric HIV medication regimens. This is particularly problematic, as poor adherence to anti-HIV medications is a serious public health concern.




1




2


When HIV adherence is poor, the potential consequences are severe; poor

adherence may be detrimental to both the individual and society. When children and families are nonadherent to their anti-HIV medication regimens, HIV-infected children may experience rapid physical decline with few therapeutic options (Butz, Joyner, Greenberg-Friedman, & Hutton, 1998; Fatkenhauer, Theisen, & Rockstroch, 1997). In terms of the impact on society, poor adherence may mean increased rates of transmission of HIV, and increased difficulty treating the virus (Wainberg & Friedland, 1998). This is particularly alarming, since adolescents are among the largest newly-infected HIV population (UNAIDS, 2000). Clearly, poor adherence among HIV-infected children and adults is a major public health concern.

Unfortunately, few studies have examined adherence to anti-HIV medication regimens among children. As a result, little is known about the extent and nature of adherence problems in this population. Thus, an understanding of adherence is necessary in order to develop specific, focused, and effective adherence interventions. The major purpose of this study was to assess both the extent of medication adherence problems in this population and the specific barriers that prevent consistent adherence.

This paper begins with a review of adherence to chronic illness regimens,

including the prevalence of poor adherence, the definition of adherence, implications of poor adherence, and the assessment of medication adherence and barriers to adherence. The paper then focuses on HIV as a new chronic illness affecting both children and adults. Information about the prevalence of HIV among children is presented, as well as general information about HIV and the corresponding antiretroviral therapy (ART) regimen. Next, implications of poor adherence to ART and the assessment of medication




3


adherence and barriers to adherence among children on ART are reviewed. The literature review concludes with a summary, specific aims, and hypothCeses for the following study, which examines the extent and nature of adherence and barriers to adherence among HIV-infected children and their families. The method, results, and discussion of the findings follow.













REVIEW OF THE LITERATURE

According to the American Academy of Pediatrics (AAP), childhood chronic illnesses are defined as conditions that last at least 3 months, and require medical attention and care that is above and beyond that which is normally expected for a child or adolescent of the same age (AAP, 1993). Estimates suggest that 18% of children in the United States have a chronic illness (Newacheck et al., 2000). The most common childhood chronic illnesses include rheumatoid arthritis, asthma, leukemia and other malignancies, spina bifida, seizure disorders, neuromuscular diseases, acquired immunodeficiency syndrome, and diabetes (AAP, 1993). Conditions vary widely in their onset, course, duration, and severity, and in the daily demands required for disease management.

Prevalence of Poor Adherence among Children with Chronic Illness

Among children with chronic illnesses, the prevalence of poor adherence to medical regimens varies according to the study sample, the specifics of regimen requirements, the assessment of adherence, and the criteria used to classify children as adherent or poorly adherent (Rapoff, 1999). Across studies of children with chronic illnesses, nonadherence has been estimated at 21-52% (Alessandro, Vincenzo, Marco, Marcello, & Enrica, 1994; Conley & Salvatierra, 1996; Ettenger et al., 199; Festa, Tamaroff, Chasalow, & Lanzowsky, 1992; Meyers, Thompson, & Weiland, 1996; Schoni et al., 1995; Weisberg-Benchell et al., 1995). However, these estimates were taken from a review of studies of adherence to a small set of chronic illness regimens, in which


4




5


sample sizes were often small, and the assessments of adherence varied in their ability to accurately estimate adherence.

Defining Adherence

The words "adherence" and "compliance" are often used to label behaviors

consistent with recommended medical regimens. The term "compliance" has become less popular in recent years because it is perceived as indicating an authoritarian approach to disease management (DiMatteo & DiNicola, 1982). Contemporary opinion suggests that the term "adherence" is more appropriate, as it places greater emphasis on the patient's active involvement in the treatment regimen (Cassell, 1991; Leventhal, Safer, & Panagis, 1983). For this reason, the term "adherence" is used throughout this manuscript to refer to the extent to which a person's health behavior is consistent with the medical regimen, as agreed upon by both the patient (or parent) and health care provider.

The definition of adherence is somewhat ambiguous, and is dependent upon

which aspects of the regimen are of interest. Medical regimens often include multiple components (e.g., medications, diet, and exercise) and each component may have multiple metrics that are important (e.g., frequency, amount, and duration). These multiple dimensions of adherence is important to consider, as adherence may vary from one regimen behavior to another (Johnson, 1995; Reid & Appleton, 1991) or from one metric to another (Johnson, 1995). For example, when researchers asked children and caregivers to tell them about the children's previous 24 hours, children with diabetes were reportedly mostly adherent to insulin injection, but less adherent to the prescribed diet (Johnson, 1995). Similarly, among adults with HIV, some patients have demonstrated 100% adherence with regards to some medications but poor adherence with




6


regards to other medications during the same interval (Hall et al., 1998). Unfortunately, much of the literature has examined adherence unidimensionally, failing to assess each aspect of the regimen and each metric individually.

The Implications of Poor Adherence

The impact of poor adherence in children with chronic illnesses is potentially severe and long-standing. Poor adherence adversely affects health care costs, clinical decision-making, and conclusions drawn from clinical trials, in addition to health outcomes (Rapoff, 1999). When patients are nonadherent to medical regimens, increased health care costs are incurred and health care resources are wasted. In the United States, the health care costs of poor adherence in both children and adults is estimated at $100 billion per year (Berg, Dischler, Wagner, Raja, & Palmer-Shelvin, 1993). Costs include wasted medication and unused therapeutic equipment, and increased morbidity leading to more frequent clinic appointments, emergency care, and hospitalization.

Poor adherence also adversely affects clinical decision-making. Providers may make clinical decisions based on the assumption that a patient is adherent, even when the patient is not (Rapoff, 1999). This may lead to the provision of increased time and medical resources, in the form of additional medication, therapies, or medical procedures.

In clinical trials, poor adherence may lead to an underestimation of the

effectiveness of medications, known as the compliance bias (Feinstein, 1974). If a person involved in a clinical trial fails to take medications appropriately but does not inform the investigator of her poor adherence, the investigator may assume that the medications were taken as prescribed. If several participants in a clinical trial repeat this behavior, the investigator may erroneously conclude that the drug is ineffective or that higher dosages





7


are needed to have a beneficial effect (Urquhart, 1989). Due to perceived ineffectiveness the drug may never become available, when it actually may be beneficial for those patients who take the medication as prescribed.

Finally, perhaps the most obvious implication of poor adherence is poor health

outcomes. In many chronic illnesses, and with many disease management behaviors, the ultimate consequence of long-term poor adherence is death. Before reaching that level of severity, poor adherence may lead to illness progression and complications, increased hospitalizations, and an increase in the number and frequency of disease management behaviors necessary in order to maintain or improve health, particularly for illnesses such as HIV, cystic fibrosis (CF), hypertension, and renal disease. In other illnesses, such as diabetes and asthma, poor adherence over even a short period of time may result in death (Birkhead, Attaway, Strunk, Townsend, & Teutsch, 1989; Delamater, 2000; Gerbino, 1993). With the exception of palliative treatments, poor adherence with most disease management behaviors results in increased morbidity and mortality.

Assessment of Medication Adherence

The assessment of adherence is a complex problem for which there is no easy

solution. Although several assessment methods have been used to measure adherence, no perfect method has been identified (Rapoff, 1999). In the absence of a "gold standard" for adherence assessment, researchers have suggested that multiple assessment modalities should be used in order to obtain the most comprehensive information. Adherence measurements have been characterized as "direct" and "indirect" assessments (Rapoff, 1999). Direct assessments such as biological assays, direct observations, and electronic monitoring are often difficult to obtain and can be costly (Rand & Weeks, 1998).




8


Conversely, multiple indirect assessments have been found to be effective, relatively easy to obtain, and inexpensive.

Regimen Knowledge

One important aspect of measuring adherence indirectly is determining the extent to which the persons responsible for a child's medication-taking are knowledgeable about the prescribed regimen. If these individuals do not know and understand the specifics regarding the medication regimen, complete adherence is unlikely. Among children with a variety of chronic illnesses, lack of child and caregiver regimen knowledge has been associated with poor adherence (Gudas, Koocher, & Wypij, 1991; Hanson, Henggeler, & Burghen, 1987a; levers et al., 1999; Rubin, Bauman, & Lauby, 1989). However, not all studies have found a relationship between knowledge and adherence (Beck et al., 1980). These variable findings may be explained because although knowledge of the regimen is important for adherence, it may not be necessary for children and caregivers to have complete knowledge while in the clinic. For example, knowledge of exact dosages may not relate to adherence, as that information is likely displayed on the medication containers, and is, therefore, available to children and caregivers when it is most needed. These findings may also be explained because although knowledge is important, knowledge alone is not sufficient for adherence. Thus, accurate knowledge appears to be an important component to adherence, though it should not be assumed that improving knowledge would necessarily improve adherence.

Several questionnaires have been developed to measure diabetes-related

knowledge. The Test of Diabetes Knowledge (TDK; Johnson et al., 1982) was developed to assess general knowledge of diabetes, diabetes-related problem-solving, and observed




9


skill in urine testing and self-injection. The Diabetes Knowledge and Management Skills Assessment Questionnaire (Brownlee-Duffeck et al., 1987) consists of 36 multiple-choice questions that inquire about diet, insulin injections, insulin reactions, urine testing, foot care, and general information about ketoacidosis and hyperglycemia. The medical regimen for diabetes lends itself to this type of knowledge assessment, as the techniques for self-care are generally consistent across individuals, and appropriate problem-solving is consistent across individuals and is crucial to successful self-management. This type of knowledge assessment is very specific to the regimen; however, it is difficult to imagine how this type of assessment might be useful in assessing knowledge of prescribed medication regimens for conditions such as HIV or cancer.

Similarly, the Metered Dose Inhaler Checklist (MDI) has been developed for

measuring inhaler/spacer technique among children with pulmonary disorders (Boccuti, Celano, Geller, & Phillips, 1996; Celano, Geller, Phillips, & Ziman, 1998). The measure was designed for providers to use as they observe patient use of inhalers/spacers. Six or seven of the skills assessed are considered crucial for effective administration, while four of the skills assessed are considered helpful to drug delivery; scores reflect these differences. Similar to the diabetes tools, the MDI is a very promising technique for the assessment of knowledge pertinent to pulmonary disease management, but does not apply to assessment of prescribed medication regimen knowledge.

Several other studies have reportedly assessed child and caregiver regimen knowledge, although they have failed to operationalize their knowledge assessment. Tebbi and colleagues (1986) assessed regimen knowledge among 46 children with cancer and their parents. Interview questions concerned knowledge of the medications and





10


understanding of the regimen, though the authors did not provide more specific information regarding the knowledge assessment. Adherence behavior was measured via self-report. Most who reported frequent nonadherence demonstrated poor regimen knowledge, supporting the relationship between regimen knowledge and adherence. Anthony, Paxton, Bines, and Phelan (1999) assessed maternal nutritional knowledge specific to CF. No information was provided about how nutritional knowledge was assessed, though disease-specific nutritional knowledge did positively relate to height, weight, and self-reports of dietary adherence behaviors. While these findings do support a relationship between knowledge and adherence behaviors, these studies fail to provide important methodological information about their assessments.

Beck and colleagues (1980) were among the first to assess knowledge of

prescribed medication regimens. They used a questionnaire to assess child and caregiver knowledge of the use, dose, side effects, and importance of each medication prescribed after renal transplantation. A total knowledge score was given, based on the percentage of questions answered correctly. At baseline, the mean knowledge score ranged from 7% to 89%, with a mean of 60%, suggesting that a large portion of the sample lacked significant knowledge about the medication regimen. Knowledge did not significantly correlate with adherence behavior.

Gudas, Koocher, and Wypij (1991) assessed knowledge of children's CF

regimens. They asked children ages 5-20 to state the medications that they take for CF, and the impact that the medications, chest physiotherapy, and diet have on the disease. Unfortunately, the researchers did not provide descriptive data about the degree of




11


knowledge among this group, though they did report that knowledge was widely variable and related to self-reports of adherence behavior.

DiGirolamo, Quittner, Ackerman, & Stevens (1997) were among the first to

report a comprehensive assessment of knowledge of the self-care regimen for children with CF. They developed the Treatment Adherence Questionnaire-Cystic Fibrosis (TAQ-CF), with a child and parent-version 10-item self-report measure that assesses adherence and knowledge of physician's treatment recommendations for airway clearance, aerosol treatments, and pancreatic enzyme use. The measure has been found to have good test-retest reliability (g= .62-.88).

levers and others (1999) used the TAQ-CF and found that 19.5% of mothers

incorrectly identified the prescribed frequency of airway clearance treatments, and 32% incorrectly identified the prescribed frequency of aerosol medications. Eleven and a half percent of caregivers incorrectly identified the greatest quantity of enzymes their children should take with a meal, and 29.8% incorrectly identified the greatest quantity of enzymes to be taken with a snack, though few of the enzyme identifications were grossly incorrect. Children demonstrated even poorer regimen knowledge. Both child and caregiver knowledge were predictive of self-reports of adherence.

Ricker, Delamater, and Hsu (1998) also reported the comprehensive assessment of knowledge of the self-care regimen for children with CF. They asked each child to indicate the prescribed amount of chest physiotherapy, the type of antibiotic, multivitamin, and pancreatic supplement, and corresponding frequencies, and dosing times. Regrettably, although the researchers collected these important data, they did not report any information about the knowledge-related findings.




12


Altogether, numerous studies have assessed regimen knowledge among children with chronic illnesses, though few studies have conducted comprehensive regimen knowledge assessments that are applicable to medication regimens. The studies that have used medication-relevant knowledge assessments focus on specific information, such as the names of the medications, frequencies, and dosing times. Such studies have often failed to report descriptive information about their findings and reliability coefficients, but do generally support the relationship between regimen-specific knowledge and adherence behavior.

Pharmacy Refill

Pharmacy refill history is another indirect method of assessing adherence that has gained wide acceptance, particularly in HIV research (Farley, Hines, Musk, Ferrus, & Tepper, 2002; Katko, Johnson, Fowler, & Turner, 2001; Laine et al., 2000; Monane, Gurwitz, Monane, & Avorn, 1993; Ostrop & Gill, 2000; Singh et al., 1996; Singh et al., 1999; Watson & Farley, 1999). With this method, pharmacy refill information is obtained for a specific time period, and then adherence rates are calculated. Methods of calculation may vary, as several studies have failed to report refill adherence calculation methods (Farley et al., 2002; Ostrop & Gill, 2000; Singh et al., 1996). Most commonly, the number of days for which the medication was prescribed during the interval is divided by the number of days for which the medication was dispensed (Katko, Johnson, Fowler, & Turner, 2001; Monane et al., 1994). Time periods over which refill adherence was assessed have varied from one month to one year (Farley et al., 2002; Katko, Johnson, Fowler, & Turner, 2001; Laine et al., 2000; Monane, Gurwitz, Monane, & Avorn, 1993; Singh et al., 1996; Singh et al., 1999; Watson & Farley, 1999; Farley et al., 2002).




13


Pharmacy refill histories are a practical and effective means to assess adherence. First, they necessitate minimal patient burden. Patients are asked to sign a release form authorizing their pharmacy (or pharmacies) to release refill history information, and then the researcher obtains the information from the pharmacies, without requiring additional patient effort. Second, the validity of this methodology has been supported by a strong correlation between refill history and adherence data based on the Medication Events Monitoring System (MEMS; Farley et al., 2002). Although refill histories provide only a gross estimation of adherence, which may be subject to error (Paes, Bakker, & SoeAgnue, 1998) they are useful in identifying patients who fail to refill their prescriptions, despite telling their providers that they continue to take their medication as prescribed. Self-Report

Another indirect method is self-report assessment. Self-report assessments vary in their sophistication, from single questions to extensive questionnaires and interviews. Similarly, the reliability and validity of such measures vary. Generally, physicians tend to overestimate regimen adherence (Finney, Hook, Friman, Rapoff, & Christophersen, 1993; Rand & Weeks, 1998) and parent and child reports are highly variable in their degree of accuracy, with social desirability, memory, and involvement in the regimen affecting the integrity of the data (Rand & Weeks, 1998). Self-report measures that inquire about brief, specific time periods tend to be most accurate (Kaplan & Simon, 1990; Klinnert, McQuaid, & Gavin, 1997). When the assessment is conducted accordingly, the self-report measure may be valuable in assessing degrees of adherence. Although self-reports are limited by a tendency to overestimate adherence behaviors, when self-reports do indicate poor adherence, the information is considered reliable.




14


Numerous self-report measures have been developed and used to assess adherence in children with chronic illnesses, though few measures have gained widespread popularity and use. One common method of self-report that has demonstrated good reliability is the recall interview or diary method. Specifically, the 24-hour recall interview (Freund, Johnson, Silverstein, & Thomas, 1991; Johnson, Silverstein, Rosenbloom, Carter, & Cunningham, 1986; Johnson et al., 1992) has been used successfully in the study of adherence in diabetes. The 24-hour recall interview is a diabetes-specific assessment of adherence. The assessment procedure includes three phone interviews (two during weekdays, one on weekend) within a two-week period. Patients and parents are interviewed separately and asked to recall the events of the previous 24-hours. If they fail to mention diabetes adherence behaviors, the interviewer cues them. Quantified measures correspond to each aspect of the regimen behavior (insulin injections, blood-glucose testing, diet, and exercise).

Generally, studies have supported the psychometric properties of the 24-hour

recall interview. Correlations between child and parent reports have ranged from r = .42 to r = .78 (M = .62), suggesting moderate inter-rater reliability (Johnson et al., 1986). Lack of complete concordance is not surprising, given that parents do not observe all of their children's behavior, and the likelihood of errors of memory. Agreement between child-reported adherence behaviors and observed behaviors supports the construct validity of the measure (Reynolds, Johnson, & Silverstein, 1990). Greatest concordance was found for the occurrence or nonoccurrence of behaviors regarding insulin injection, exercise, and blood-glucose testing. Poorer agreement was found for the time during which the behaviors occurred. The poorest agreement was found for specific measures of




15


dietary behaviors (e.g., grams of carbohydrate, grams of fat, and total calories). Over a three-month time period, moderate support was found for the stability of the measure, with greater stability for blood-glucose testing and dietary behaviors compared to exercise and injection behaviors (Freund et al., 1991). Altogether, the findings suggest that the 24-hour recall interview is a moderately reliable adherence measure with good construct validity.

Recall interviews have received support for their utility in the research domain; however, they lack clinical practicality. First, they require a trained interviewer who is available in the evenings and on weekends. Although it may be relatively easy to find such a person in an academic research center, it may be more difficult to find such a person in the clinical environment. Second, they require that the patients and parents are accessible by telephone in the evenings and on weekends. This may be practical for middle class families with daily routines, but may be more difficult for disadvantaged families who may be without phone service or for families in which parents work evenings and weekends. These difficulties may explain why recall interviews have not generally been integrated into most medical settings. Finally, while recall interviews can detect children who are nonadherent, as they are conducted currently, recall interviews provide little information about the barriers to adherence and potential targets for intervention. While it is helpful to assess the extent of adherence, information regarding appropriate targets for intervention would benefit both patients and providers.

While recall interviews may be impractical for clinical settings and do not identify specific barriers to adherence, the structured or semi-structured interview is an appropriate method to assess adherence and barriers to adherence in the clinical setting.




16


The clinical interview is a relatively inexpensive, unobtrusive methodology that allows for ideographic assessment of adherence behaviors and barriers. It is practical for use in both clinical and research domains. However, few studies have used the clinical interview for the assessment of adherence behaviors in children with chronic illnesses (Hanson et al., 1996; La Greca, 1995), and few known studies have used the clinical interview to assess the extent and nature of barriers to adherence.

Assessment of Barriers

Barriers to medical regimen adherence have rarely been systematically studied in the child and adult literature. Although many studies have examined barriers to treatment among children with chronic illnesses, most studies have examined barriers as individual specific constructs that may be predictive of adherence, such as perceived severity, perceived self-efficacy, and self-esteem, among others. Few studies have asked patients to identify their perceived barriers to treatment, neglecting a great resource that may help to elucidate appropriate areas for intervention.

One study assessed barriers to adherence among teenagers and adults with

diabetes (Glasgow, McCaul, & Schafer, 1986). In this study, the Barriers to Adherence Questionnaire (BAQ) was developed, based on the Behavior Analytic Model (Goldfried and D'Zurilla, 1969). Six individuals with insulin-dependent diabetes mellitus (IDDM), and two diabetes nurse-educators generated a list of barriers to adherence to various aspects of the diabetes regimen (diet, exercise, glucose testing, and insulin injections). Next, the generated items were empirically tested, and a scale with 15 items was derived. Frequency and severity ratings were provided, although the severity ratings were eliminated due to high intercorrelation with the frequency ratings. No data are available




17


regarding the reliability of the measure. Construct validity was supported by moderate correlations between the BAQ and self-reported adherence. Although the BAQ may be an effective tool in identifying barriers to adherence among individuals with diabetes, no additional research has used the BAQ to assess barriers to adherence. Perhaps the reliability of the measure was poor or the predictive validity of the measure was inadequate to warrant further use, although these are simply speculations. Nevertheless, a questionnaire that assesses regimen-specific barriers may be important in identifying targets for interventions.

Rapoff (1999) reports that he often assesses barriers to adherence during clinical interview by asking patients: "What gets in the way of you taking your medicine?" He reports that the responses can be helpful in identifying targets for intervention. However, no known published study has systematically assessed barriers in this manner.

Other literature on adherence in childhood chronic illness populations reveals common developmental and behavioral reactions to managing complex medical regimens. Several important trends have been identified. For example, the tendency for older children and adolescents to exhibit poorer adherence than younger children has been supported by this literature (Hanson, Henggeler, & Burghen, 1987b; Johnson, Silverstein, Rosenbloom, Carter, & Cunningham, 1986; Ricker, Delamater, & Hsu, 1998). This may be, in part, because parents may tend to discontinue or decrease supervision of disease management behaviors as children become older (Johnson, 1995). Moreover, adolescence, in particular, is a developmental period strongly related to decreased disease management behaviors. Adolescence is a time of striving for independence, and one way in which adolescents sometimes assert their independence is by choosing to not perform disease management behaviors.




18


Long duration of the regimen, and specific regimen attributes have also been identified as important barriers to adherence. First, long duration of the regimen may hinder adherence. It has been suggested that adherence is a greater problem with chronic versus acute regimens (Rapoff, 1999). Relatedly, with childhood chronic illnesses studies have found that adherence tends to decrease over time (Brownbridge & Fielding, 1994; Hanson, DeGuire, Schinkel, Henggeler, & Burghen, 1992; Hanson, Henggeler, Harris, Burghen, & Moore, 1989; Jacobson et al., 1990). It may be that children begin to resent medication taking and may decide to stop taking it regularly. Also, as families adjust to the medication regimen, parents may begin to rely on their children to manage their regimens independently, providing opportunities for children to skip, hide, or dump doses.

Specific attributes of the regimen may also make adherence difficult.

Behaviorists have well established the organism's tendency to avoid aversive stimuli, and this may be why complex regimen components, like chest physiotherapy, have been associated with poor adherence (Passero, Remor, & Salomon, 1981). Behaviorists have also established the organism's tendency to perform behaviors that are reinforcing, and this may explain the difficulty children with diabetes experience with abstaining from sweet and fatty foods (Glasgow, McCaul, & Schafer, 1986).

The Human Immunodeficiency Virus: A New Chronic Illness

Worldwide, more than 34 million people are infected with HIV; of those, 1.3 million are children (UNAIDS, 2000). Each day, almost 2,000 children are newly infected with the virus. Specifically, in the United States more than 12,000 children are living with HIV (UNAIDS, 2000) and more than 7,000 children are currently living with




19


AIDS, the advanced manifestation of the virus (CDC, 1996). Nationally, HIV has been ranked as the seventh leading cause of death among children ages 1 to 4 (Ventura, Peters, Martin, & Maurer, 1997). Improving mortality and morbidity among these children and adolescents is contingent upon HIV-infected children receiving appropriate treatment (Palella et al., 1998).

HIV is a retrovirus that is transmitted through exposure to infected blood and sexual fluids, typically through needle sharing, unprotected sex, or mother-to-child transmission (CDC, 1998). HIV disrupts the immune functioning of infected persons. As a result, people with HIV have difficulty fighting off certain bacteria, viruses and other microbes. Therefore, they are prone to develop opportunistic infections, which take advantage of the compromised immune system. Over 100 microorganisms have been identified as opportunistic, including: Candida allbicans, Varicella-Zoster Virus, herpes simplex virus, measles, cytomegalovirus, congenital syphyilis, and pneumonocytic carini pneumonia (PCP). If not treated successfully, such opporuntistic infections can lead to AIDS-related fatalities (Butz et al., 1998, Palella et al., 1998). Also, HIV can lead to neurological disease, neuropsychological impairment, cardiac abnormalities and hematologic problems, among others (Scott & Sleasman, 1999). With no known cure, the end result of HIV is death.

In the past, HIV was considered a debilitating illness characterized by rapid deterioration and demise. However, the development of new drug therapies for HIVinfected children and adults has led to increased life-spans for infected individuals (Martino et al., 2001, Palella et al., 1998). Martino and colleagues (2001) found that when children were prescribed at least 3 anti-retroviral medications, the risk of death




20


decreased by 70%. Now that HIV-infected children are living into adolescence and young-adulthood, the issue of adherence to their medication regimens is becoming increasingly more important.

Typically, the preferred pediatric antiretroviral medication regimen consists of combination ART, specifically, highly active antiretroviral therapy (HAART; Scott & Sleasman, 1999). With standard HAART regimens, children are prescribed at least three medications to be taken two or three times per day. Children may take their medication in pill form or in liquid doses either because they are unable to swallow pills or because some pills are not available in small doses appropriate to body weight. Many of the antiretroviral medications have unpalatable taste and gastrointestinal side effects. Side effects for antiretroviral therapy (ART) include gastrointestinal discomfort, nausea, vomiting, diarrhea, headache, peripheral neuropathy, parasthesis, rash, mouth ulcers, dry skin, and others, including organ damage (Carpenter et al., 2000). Side effects range in frequency, duration, and severity. To combat side effects, as well as opportunistic infections and other complications, healthcare providers may prescribe additional medications (Kelly, Otto-Slaj, Sikkema, Pinkerton, & Bloom, 1998), such as Zantac, Septra, and a variety of others.

With ART and other drugs to combat side effects, infections, and complications, medication regimens for HIV-infected children can be complex. Many of the medications have special instructions that are difficult to follow. For example, to ensure proper absorption, some medications should be taken with food, while others should not (Kelly et al., 1998). Others require a large amount of fluid take. Additionally, to prevent drug interactions and to maintain the appropriate level of medications in the bloodstream,




21


often the timing of medication taking is crucial (Kelly et al., 1998). With such complex instructions, it is not surprising that adherence to ART is of great concern (Friedland & Williams, 1999; Kelly et al., 1998).

Implications of Poor Adherence to ART

As with other chronic illnesses, poor adherence to ART affects healthcare costs, clinical decision-making, conclusions drawn from clinical trials, and health outcomes. The widespread use of HAART has reduced morbidity and mortality, hospitalizations, the development of opportunistic infections, and the use of home-care services and hospices (Brettle et al., 1998). In reducing morbidity and mortality, HAART has also led to reduced costs; however, when patients are nonadherent to HAART regimens, HAART may not be cost-effective, because the expensive medicines are not taken at the appropriate intervals and are likely less effective. As a result, resource utilization may be higher and costs may approach those incurred prior to the availability of ART.

Poor adherence to ART can also adversely affect clinical decision-making. For example, consider an HIV-infected patient who misses doses of ART often, but is perceived to be adherent. If viral suppression is not achieved with the current medication regimen, then a physician may conduct extensive laboratory tests in hopes of determining the reason for failure. Orders for these tests would not have been made if the physician knew the patient was poorly adherent. The physician might have instructed the patient to discontinue the medication altogether, since intermittent adherence may be more harmful than effective (Friedland & Williams, 1999). The physician may have prescribed another medication with fewer side effects or with simpler dosing, with which the patient might




22


have been more adherent. Thus, non-detection of adherence problems may result in inefficient, ineffective, and perhaps harmful medical care.

As discussed previously, poor adherence also affects conclusions drawn from clinical trials. Undetected poor adherence in clinical trials can lead researchers to erroneously conclude that drugs are not effective. With HIV, clinical trials are a critical component of the development of new medications. Since few drugs are currently available and drugs may become less effective for an individual over time (Gallant, 2000), the consequences of poor adherence in HIV clinical trials may be widespread and significant.

As antiretroviral therapy has begun to slow disease progression and increase life span, the issue of adherence to medication treatment has become more salient. Friedland and Williams (1999) stated that: "the key to success of the new regimens is the ability and willingness of HIV-positive individuals to adhere to complex antiretroviral regimens, perhaps for life." Adherence is essential to suppressing the viral load, the amount of HIV in blood. Although multiple factors may contribute to inadequate viral suppression, poor adherence to ART regimens is among the most important (Melvin, 1999).

Researchers have attempted to identify the adherence rates necessary to achieve optimal health. Among adults, one prospective study assessed adherence levels among 81 mostly male patients with MEMS for 3 to 15 months (Paterson et al., 2000). Results showed that patients with > 95% adherence to ART regimens have lower viral loads, greater increases in CD4 counts (amount of T-lymphocyte white blood cells), and lower hospitalization rates than those with poorer adherence. Another study assessed adherence among 24 HIV-infected adults on ART (Bangsberg et al., 2000). Researchers found that




23


pill count adherence significantly predicted viral load, such that a 10% difference in adherence was associated with a doubly increased viral load, suggesting that 90% adherence or greater is necessary for viral suppression. McNabb and colleagues (2001) also used the MEMS to assess adherence among 40 mostly-male HIV-infected adults. Results of this study are similar to the results of Bansberg et al (2000), in suggesting that > 90% adherence is necessary for full viral suppression. While these studies suggest that at least 90% adherence is necessary for virologic success, one study among a pediatric population suggests that lower levels may be sufficient (Farley et al., 2002). Farley and colleagues used MEMS and pharmacy refill data to assess adherence to one ART drug over a 6-month period. MEMS data alone suggested that an adherence rate of> 80% was robustly associated with virologic success, and when pharmacy refill data were combined with MEMS data, the rate of> 80% adherence was even more strongly associated with virologic success. Thus, among children, at least 80% adherence may be necessary for optimal health.

Poor adherence to the HIV regimen may also have severe health consequences such as the development of fatal opportunistic infections and drug resistance. When the virus is not completely suppressed, potentially fatal opportunistic infections may develop. If not treated successfully, opportunistic infections such as PCP lead to AIDS-related fatalities. Moreover, when medications do not fully suppress the virus, the virus rapidly mutates into strains that are resistant to drug therapies (Fatkenheuer et al., 1997). Mutations occur naturally with HIV, but are prevented when ART stops viral replication completely. When the virus is partially suppressed, as occurs with intermittent or incomplete adherence, drugs may render common strains inactive, but may not prevent





24


mutations from replicating. The result is a high presence of the mutant virus in the blood that is resistant to those drugs and similar drugs, as well. This is particularly problematic since resistance may occur most rapidly among those whose adherence falls between 50% and 90% (Bangsberg et al., 2000), a range in which many children and adults on HAART may fall (Belzer, Fuchs, Luftman, & Tucker, 1999; Kastrissios et al., 1998; Watson & Farley, 1999).

For both the individual and society, the impact of poor adherence and subsequent drug resistance may be severe. First, with the limited number of antiretroviral agents and the ability of resistance to affect entire drug classes, resistance may leave an individual with few options for additional therapy (Gallant, 2000). As was seen in the early 1980's, without therapy the virus replicates and completely suppresses the immune system. When the immune system is suppressed, the body cannot fight even a common cold. Unfortunately, without therapy, rapid physical decline is imminent. Second, research has demonstrated that drug resistance leads to increased viral replication in the blood, semen, and vaginal fluid (Gupta et al., 1997; Vernazza, Gillem, & Flepp, 1997). The potency of the virus in these fluids increases the likelihood that the virus may be transmitted to exposed individuals (Wainberg & Friedland, 1998). If the drug-resistant virus is transmitted to others, those infected will have a drug-resistant virus, as well (Friedland & Williams, 1999). Like the poorly adherent individuals with drug resistance, newly infected individuals will have little or no options for therapy. As a result, future generations of HIV-positive individuals may suffer dramatically from the poor adherence of their predecessors. Clearly, among HIV-infected individuals, the potential public health implications of poor adherence and subsequent drug resistance are grave.




25


Assessment of Adherence among Children with HIV

Despite the implications of poor adherence among children on ART, few studies have examined adherence levels in this population. Although many studies have assessed adherence as one of the inclusion criteria for clinical trials (J. W. Sleasman, personal communication, July 27, 2000), and as part of the Pediatric Aids Clinical Trials Group assessment battery, few known studies have reported systematic assessments of adherence, and no known studies have assessed knowledge of the prescribed medication regimen among HIV-infected children.

Regimen knowledge

Only one study has assessed regimen knowledge among caregivers of HIVinfected children or among the children themselves. Katko and colleagues (2001) asked 35 caregivers to name or describe their children's ART medications and corresponding dosages and dosing frequencies. Nineteen (54%) of the caregivers provided accurate medication information, and of those, 12 had pharmacy refill ratings of 90% adherence or better, while none of those who lacked knowledge of the regimen had 90% adherence ratings or higher. This study was limited by a small sample size and only one method of adherence assessment, yet the study is important, as it supports the use of knowledge assessment as an indicator of possible poor adherence.

Although only one study has included regimen knowledge as part of the adherence assessment of HIV-infected pediatric populations, several studies have assessed regimen knowledge among HIV-infected adults. In a small study, Svarstad, Chewning, Sleath, and Claesson (1999) assessed regimen knowledge among 20 HIVinfected men and women. As part of the Brief Medication Questionnaire (BMQ) they conducted a "Regimen Screen." Patients were asked to list all of the medications taken




26


within the past week and then were asked several related questions, including information about how many times per day they took each medication, and how many pills they took each time. The Regimen Screen also consisted of questions about how many doses were missed in the previous week. Four out of 20 patients reported some nonadherence or lack of knowledge on the Regimen Screen. Results showed that the Regimen Screen was associated with repeated nonadherence, measured with MEMS. However, the Regimen Screen results were not associated with sporadic nonadherence.

Bangsberg and colleagues (2001) included a brief regimen knowledge assessment in another small study. Forty-six mostly male patients on HAART were asked about their medication regimen during an interview at patients' typical places of residence. Six patients reported that they were taking either a different number of doses per day or a different number of pills per dose than the medication container instructed. Four of those patients attributed this difference to misunderstanding, while two reported that they had decided to take their medicine differently than it was prescribed.

In a larger study, Parietnti and colleagues (2001) used a pill identification test to determine regimen knowledge among 223 HIV-infected adults in France. Participants were given a board that contained pictures of 23 ART medications, with 2 similarlooking (twin pills) for each ART medication. A score was calculated based on the number of misidentifications weighted according to the degree of pill resemblance. No information was provided about the total percent of patients with inadequate knowledge, though results showed that 38% of patients with poor adherence based on an investigatorcompleted scale also had poor regimen knowledge.




27


As part of another large multi-center study, Stone and others (2001) used a structured interview to assess regimen knowledge among 289 HIV-infected women. Participants were presented with a card bearing photographs of all current FDA-approved ART medications and corresponding medication names. They were then asked whether or not they were currently taking each of the medications. Participants were then asked to report the dosing frequency and dietary requirements related to each endorsed medication. Sixty-three percent of the patients reported correct information about dosing frequency and dietary requirements for each endorsed medication. Of those with incorrect information, 8% reported incorrect information about both frequency and dietary requirements, 17% reported incorrect information about frequency only, and 12% reported incorrect information solely about the dietary requirements. The relationship between regimen knowledge and adherence behavior was not reported.

Another multi-site study used a computer-assisted structured interview (CASI) to assess regimen knowledge among HIV-infected adults (Bangsberg, Bronstone, & Hoffman, 2002). One hundred and forty-one patients were presented with a computer screen bearing names and images of ART medications. Patients were instructed to click on the images of medications currently included in their regimen. The program then prompted patients to provide information about the number of pills per dose, number of doses per day, and associated dietary requirements for each medication. Adherence was assessed in a 3-day recall format as part of the CASI. More than half of all patients made at least one error regarding their medication regimens. Fourteen percent failed to identify at least one prescribed medication, 18% incorrectly reported the number of daily doses for at least one prescribed medication, and 19% incorrectly reported the dietary




28


requirements pertinent to at least one prescribed medication. The relationship between regimen knowledge and adherence behavior was not reported.

In summary, only one known study has examined regimen knowledge among HIV-infected children, and that study supports the use of knowledge assessment as an indicator of possible poor adherence. Five known studies have assessed regimen knowledge among HIV-infected adults on ART; these studies varied in their compositions of the sample and sample size, yet all of these studies have demonstrated that some inaccurate regimen knowledge exists among adults on ART. Two of the adult studies showed that at least some poor adherers have inaccurate regimen knowledge, providing additional support for the importance of regimen knowledge assessment. Pharmacy refill and self-report

Several studies have examined medication adherence among HIV-infected

children by examining pharmacy refill histories. Watson and Farley (1999) examined adherence with pharmacy refills among 72 children ages 3 months to 12 years. Children were considered adherent if> 75% of ART medications were refilled during the first 180 days of PI therapy. Fifty-eight percent of children were considered adherent by this liberal criterion. This suggests that at least 42% families with children on ART did not claim most of their prescriptions from the pharmacy and, therefore, could not have been consistently adherent to their regimens.

Katko, Johnson, Fowler, and Turner (2001) used pharmacy refill data to examine adherence among 34 children on ART. This study considered children adherent if over a one-year period the proportion of days for which medication was dispensed to the days for which the medication was prescribed was less than or equal to 90%. Only 34% were





29


considered adherent by this more stringent criterion. Like the previous study, this study is limited by the method of adherence assessment, as refill histories are subject to error when patients have extra medication at home or when patients refill prescriptions at pharmacies that are not listed in the research databases. The approach in this study was also unidimensional, failing to account for differences in adherence based on individual medications, and adherence to food and drink intake guidelines. However, the utility of refill histories in estimating adherence has been supported by the literature (Laine et al., 2000; Monane, Gurwitz, Monane, & Avorn, 1993; Singh et al., 1999) and despite the unidimensional approach, the study is useful in suggesting that poor adherence may be a pervasive problem for families of children on ART regimens.

A recent study with children <13 years-old was the first known study to assess ART adherence multidimensionally among children (Farley et al., 2002). The study examined adherence among 26 children using MEMS, pharmacy refills, caregiver selfreports, and physician and nurse assessments. Median adherence ratings were 81% for MEMS, 79% for pharmacy refills, and 100% when caregivers were asked to report the missed doses during the 3 days prior to interview, though MEMS data demonstrated 31 missed medication events of the 126 prescribed events for the 20 respondents over the 3day period. Physician and nurse ratings were also assessed, and were significantly related to MEMS data (kappa= 74%, p< .05). Pharmacy refill ratings alone did not significantly relate to viral load data, but when MEMS and pharmacy data were combined, an adherence cut-off of 80% was a good predictor of virologic success. This study is consistent with studies of HIV-infected adults, which illustrate the utility of MEMS and pharmacy refill histories for assessing ART adherence (Katko et al., 2001;




30


McNabb et al., 2001; Paterson et al., 2000; Svarstad et al., 1999). This study also demonstrates that despite previous research suggesting that providers overestimate adherence (Rand & Weeks, 1998), some providers may be particularly astute at assessing patient adherence.

Taken together, the reviewed studies suggest that near-perfect adherence may be difficult to achieve among children and adolescents on ART regimens. If 80% adherence or greater is necessary to achieve optimal therapeutic benefit (Farley et al., 2002) the above results suggest that a large percentage of children may not be achieving maximum benefit from ART. Although theses studies are useful in estimating the prevalence of poor adherence, additional studies with larger sample sizes and multiple assessment modalities (Quittner, Espelage, levers-Landis, & Drotar, 2000b) are necessary in order to understand the extent and nature of adherence problems among families of children on ART. Only with a systemic understanding of the problem can researchers begin to develop specific, targeted interventions that are effective in promoting adherence in this population.

Barriers to ART Adherence

Understanding barriers to adherence may be essential in understanding the problem of poor adherence. Prior to developing interventions to increase adherence among HIV-infected children, researchers must identify and understand the obstacles to adherence. To date, few studies have attempted to identify these obstacles. Though barriers are assessed as part of the Pediatric AIDS Clinical Trials Group (PACTG) protocols, few known studies have examined barriers among younger HIV-infected children and their families. Furthermore, only one known study has examined barriers to




31


adherence among HIV-infected adolescents and only several studies have examined barriers to adherence among HIV-infected adults.

The PACTG adherence modules require the nurse to conduct a structured

interview to assess knowledge of the medication names, dosage, frequency of dosing, and number of missed doses in previous 3 days (Module 1), and then assess barriers by reading a list of previously identified barriers and then asking caregivers: "Over the last two weeks, have any of the following been problems for you with (drug name)?" (Module 2). When Farley and colleagues (2002) administered the modules to a group of 20 caregivers of children under age 13, 6 caregivers identified problems adhering to 17 medications. The most commonly endorsed problems were "didn't refill, ran out," "scheduling interferes with lifestyle," "child refuses," "multiple caretakers," "forgot," and "taste." All of the caregivers who identified problems had missed medication during the prior two weeks, according to MEMS data. Conversely, not all of those who had missed doses identified problems. Thus, this study identified some barriers that may be common among caregivers of HIV-infected children and suggests that problem or barrier identification may be one indicator of poor adherence. However, this study was limited by the small sample size and the close-ended format of the barrier questions.

Another study assessed barriers to adherence among adolescents with HIV.

Belzer and colleagues (1999) provided 31 teenagers with a 7-item list of reasons they may have missed their medication, and asked the teens to rate the reasons on a Likert scale that ranged from 1= "strongly disagree" to 2= "strongly agree." The items, in order from most highly endorsed to least highly endorsed, are: "too many pills," "reminds me I'm HIV-positive," "side effects," "interferes with schedule," "forgot," "disclosure




32


concerns," and "depressed." Although the study identified which of those 7 reasons was most widely endorsed by the teens, the study was limited by the pre-selected list of barriers examined and a small sample size. Further, no information was provided about how those barriers were chosen for examination. Thus, although the study examined the extent to which the 7 barriers were endorsed by the teens, the study failed to identify the variety of other barriers that may have impacted adherence.

Two qualitative studies have assessed adherence to ART regimens among adults. In a pilot study of HIV-infected adults enrolled in a clinical drug trial (Chesney, 1997), patients who admitted missing doses were asked to list as many reasons as possible for why they missed their medication. Common barriers listed were: forgetting (40%), sleeping through doses (37%), being away from home (34%), changing the therapy routine (27%), being busy (22%), being sick (13%), experiencing side effects (10%), and feeling depressed (9%). In a study of 20 HIV-infected women who were asked to "write about a future in which you only have to take one pill a day," multiple barriers to adherence were revealed. Many women reported that the regimens were difficult to follow, pointing to the dosing intervals, food guidelines, number of pills, and pill size, as particularly problematic. Side effects were also identified as significant barriers, specifically excessive perspiration, nausea and vomiting, diarrhea, and weight loss or gain associated with medication-specific food requirements. Lack of confidence in pill effectiveness was also discussed, as well as the way in which the medications serve as a regular reminder of disease status. Finally, scheduling problems and high activity were also reported as barriers to adherence. Jointly, these qualitative studies demonstrate the value of open-ended assessment for the identification of barriers to adherence.




33


Several studies have conducted quantitative assessments of barriers, based on previous qualitative studies. Mannheimer and colleagues (2002) asked over 1,000 mostly-male HIV-infected adults to complete a 10-item medication barriers chescklist. Forgetting to take the medication was the most highly endorsed barrier, followed by being away from home, experiencing side effects, and having difficulty taking the pills at specified times. The frequency of reasons was stable over 12 months.

Ammassari and others (2001) administered a 12-item barriers chescklist to 358 mostly-male HIV-infected adults. Participants were asked to rate the importance of each barrier in the missing or discontinuing of drugs on a four-point Likert scale. The most highly endorsed barriers related to the number of pills, concern about future side effects, being away from home, and concern about others gaining awareness of their medicationtaking and disease status.

Walsh and colleagues (2001) assessed barriers to adherence among 157 HIVinfected individuals ages 16 and older. Participants completed a 20-item questionnaire that listed barriers to adherence and requested them to rate on a five-point Likert scale how often each item led to missed doses. Items with the highest frequency ratings related to forgetting, oversleeping, eating at the wrong time, being busy, being in social situations, side effects, feeling ill, and not wanting to be reminded of HIV status.

Murphy, Roberts, Martin, Marelich, and Hoffman (2000) asked 39 adults on ART to complete a 23-item barriers chescklist based on the Aids Clinical Trials Group (ACTG) Baseline Adherence Questionnaire. Participants were asked to endorse an item if it prevented them from taking their medications as prescribed. The most highly endorsed items were: "slept through the dose time," "had problems taking pills with




34


special instructions," "had a change in daily routine", "did not have medications with you," "busy and did not want to stop to take medications," "simply forgot," "felt depressed or overwhelmed," "felt angry, depressed, or hopeless that you have to deal with this," "wanted to forget the whole thing," and "side effects." After completing the barriers chescklist open-ended discussions were held to discuss barriers, as well as successful strategies. During the focus groups, several themes emerged. One theme was lack of knowledge or false beliefs about the importance of taking the medicine regularly, at certain intervals, and with specific food guidelines. Participants also pointed to the complexity of the instructions, the poor taste, pill size, and side effects, as problematic. Another theme was difficulty with the scheduling of medications, and particular concerns were raised about taking medication in social situations. Finally, participants stated that animosity toward the healthcare provider and confusion about the medication-taking instructions the provider gave both negatively impact adherence.

Catz and colleagues (2000) examined regimen-specific barriers to adherence

among 72 mostly male HIV-infected adults. They developed the Barriers to Adherence Checklist (BAC), a 56-item questionnaire that measures general barriers to HIV treatment, and barriers specific to the medication regimen. Patients rate the degree to which each item reflects their own experiences. The items were consistent with barriers identified during qualitative analysis of patients on HAART (Bogart et al., 2000). The most highly endorsed barriers were: "treatment reminds me that I am HIV+," "I do not want others to know that I am HIV+," "when I have clinic appointments, I forget to ask some of my questions about treatment," "I have trouble remembering names of medicines and what they are for," and "I do not like the way my medication makes my body feel."




35


Patients who reported at least weekly missed doses reported a significantly higher number of barriers than those who were more adherent. This study is different from the other studies, as endorsement of barriers was not necessarily based on the relationship between items and missed doses, only on the extent to which items reflected participants' experiences.

Altogether, several studies have examined barriers to adherence among HIVinfected adults. Several appropriate tools have been developed for identifying barriers to ART adherence among adults, and one tool has been developed for pediatrics. Many of the barriers assessed by these measures may be relevant to pediatric HIV regimens; however, in order to determine the nature and extent of the adherence barriers for families of children on ART, an extensive examination of the barriers facing these families is necessary. Given the lack of research examining the adherence barriers to pediatric ART regimens, an in-depth assessment of medication-taking behaviors and barriers is warranted. Quantitative assessment methods are insufficient for studying this problem, as they may fail to include important barriers. Alternatively, qualitative assessment is indicated. Such a thorough approach is essential in eliciting the families' input regarding their experience of the HIV-specific disease and regimen behaviors, and in identifying those factors that may obstruct disease management. Furthermore, qualitative assessment has been implicated by the World Health Organization (O'Reilly, 1995) as an important methodology within HIV/AIDS research "to understand behaviors in their context and thereby to identify barriers and potential facilitators of behavior change" (p. 34). Thus, qualitative methods are necessary in order to understand the





36

problem more thoroughly and provide the information needed to effectively apply quantitative methods and specific, focused interventions to the problem.













OVERVIEW AND HYPOTHESES

Based on the research literature, a necessary step towards the provision of

interventions to promote adherence in this population would be an examination of the extent and nature of adherence among families of children on ART regimens. The research indicates that multiple assessment methods are needed in order to fully capture the extent and nature of adherence (Quittner et al., 2000b). For this reason, this study used a more subjective, but specific measure (structured interviews) along with a more objective, but global measure (pharmacy refill history) of adherence. Since previous studies had not asked families to identify the extent and nature of the barriers that hinder adherence to pediatric ART regimens, the full extent and nature of the barriers relevant to this population were unknown. Therefore, in this study, a qualitative assessment was conducted to comprehensively identify the barriers to adherence experienced by families of children on ART.

Based on the review of the literature, the specific aims of the current study were:

Primary Aims:

1) To document caregivers' knowledge of medication names, dosage, dosing

frequencies, dosing intervals, and medication-specific food and drink intake

guidelines.

2) To document rates of adherence to prescribed medication-taking frequencies,

intervals, and medication-specific food and drink intake guidelines.





37




38


3) To document caregivers' perceived barriers to adherence to HIV medication

regimens.

Secondary Aims:

1) To document the utility of the Adherence Interview-HIV (AI-HIV) for

obtaining information about caregiver knowledge and adherence.

2) To document the utility of a medication display card in helping caregivers

identify medication names.

3) To document the utility of the 24-hour recall interview (24RI) for assessing

adherence to pediatric HIV regimens.

4) To document the relationship between regimen knowledge and adherence to

pediatric HIV regimens.

5) To document the relationship between disease severity and adherence to

pediatric HIV regimens.

Hypotheses:

1. Caregiver knowledge of the prescribed regimen will differ significantly from

reported typical adherence behavior (AI-HIV) as some caregivers will actively decide not to give their children some medications that may have poor taste or

significant side effects.

2. Caregiver knowledge of medication names will not differ significantly for

those who used the medication display card to identify the names versus those

who did not.




39


3. All adherence measures (24RI frequency, 24RI interval, 24RI dietary, refill

ratings and nurse ratings) will correlate significantly and positively with one

another.

4. Adherence frequency measures (24RI frequency, refill ratings) will

significantly and negatively relate to disease severity (CDC severity ratings,

and viral load).

5. Adherence will significantly and positively relate to caregiver knowledge.

Specifically, caregiver medication knowledge and frequency knowledge will

relate significantly to 24RI frequency adherence and refill adherence,

caregiver knowledge of interval dosing will significantly relate to 24RI

interval adherence, and caregiver knowledge of dietary requirements will

significantly relate to 24RI dietary adherence.

6. 24RI adherence to frequency and refill adherence rates will vary between

medications, based on formulation and frequency of dosing. Due to the

complexities of thrice daily dosing, medications prescribed for thrice daily

dosing will have lower adherence, based on 24RI frequency and refill ratings, than medications prescribed for once and twice daily dosing. Also, due to the

poor taste associated with many of the liquid medications, those medications

in pill form will have higher adherence, based on 24RI frequency and refill

ratings, than those in liquid form.

7. Caregivers will report more barriers during the 24RI than during the AI-HIV,

because the daily diary format of the 24RI will help to elicit memories of

recent barriers.













METHOD

Participants

Study participants were primary caregivers of HIV-seropositive children ages 212, contacted at pediatric HIV specialty clinics in Gainesville and Jacksonville, Florida, and Baltimore, Maryland. Seventy-three primary caregivers met eligibility criteria and were asked to participate in the study. Sixty-three primary caregivers (86%) agreed to participate in the study, including 18 from the Gainesville site, 25 from the Jacksonville site, and 20 from the Baltimore site. Chi square tests and one-way ANOVAs were conducted to test for demographic differences among participants from the three sites. No significant differences were found for child or caregiver age or gender, caregiver marital status, caregiver education, caregiver employment status, household yearly income, child viral load, whether or not the children were aware of their HIV status, whether or not the caregiver was HIV-infected, and the frequency at which the caregiver reported helping the child with medication-taking. Differences were found for both child and caregiver ethnicity (X2= 15.55, < .05; X2= 12.59, p= .05, respectively), with fewer African-American participants at the Gainesville site.

Full descriptive statistics on participant demographic information can be found in Tables 1 and 2. Child participants were primarily African-American (79.4%), male (57.1%), and aware of their HIV diagnosis (50.8%). The average child participant was 8.8 years old (SD=2.97) and had been prescribed ART for an average of 6.8 years. The modal viral load was less than 50 (undetectable). Caregiver participants were primarily


40




41


African-American (66.7%), female (90.5%), and reportedly HIV uninfected (69.8%). Most caregivers were adoptive parents (28.6%), biological mothers (27%), or grandparents (23.8%).

Procedure

At each study site, the research nurse was asked to help the subject recruiter

complete the eligibility checklist (Appendix A) for each patient. Next, the recruiter asked eligible parents/guardians several preliminary questions (Appendix A also) to ensure eligibility. Multiple eligibility criteria were established for the study (Table 3).

Next, eligible participants underwent informed consent procedures in accordance with IRB approval from all three study sites. If children were accompanied to the clinic by foster parents or other caregivers, written informed consent was first obtained from the parent or legal guardian. After written informed consent was obtained from the parent or legal guardian, written informed consent was obtained from the non-parent caregiver informant.

After completing informed consent procedures, caregivers were interviewed by trained undergraduate psychology students or graduate students in clinical psychology, using the AI-HIV (Appendix B). All interviewers were Caucasian. Next, the interviewer completed the demographic questionnaire (Appendix C) and the telephone contact form (Appendix D) with the caregiver. Finally, the interviewer completed the Medical Record Report Form (Appendix E).

Following the clinic visit, parents/caregivers were contacted for telephone follow-up with the 24RI (Appendix F). Beginning two weeks following the clinic visit, parents were asked to complete the 24RI three times within a two-week period, including





42


Table 1. Participant Demographic Information


Variable Children (n=63) Caregivers (n=63)

Gender
Female 42.9% 90.5% Male 57.1% 9.5% Ethnicity
African-American 79.4% 66.7% Caucasian 12.7% 25.4% Hispanic 6.3% 4.8% Other 1.6% 3.2% HIV Status
Infected 100% 30.2% Know HIV Status 50.8% 30.2% Caregiver Relationship
Biological Parent -- 34.9% Biological Grandparent -- 23.8% Adoptive Parent -- 28.6% Foster Parent -- 1.6% Relative -- 9.5% Family Friend -- 1.6% Caregiver Marital Status
Single or Living with Partner -- 33.3% Married -- 39.7% Separated or Divorced -- 19.0% Widowed 7.9% Caregiver Education
Grade School -- 7.9% Some High School -- 25.4% Graduated High School or Earned GED -- 35% Technical/Vocational Training -- 1.6% Some College -- 23.8% Bachelor Degree or Higher -- 6.4% Caregiver Employment Status
Employed Full-time -- 36.5% Employed Part-time -- 7.9% Employed Student -- 1.6% Unemployed -- 11.1% Disabled -- 15.9% Retired -- 9.5% Homemaker -- 17.5%





43


Table 2. Additional Participant Demographic Information Variable M SD
Child Age 8.76 2.97 Years Prescribed ART 6.75 2.98 Caregiver Age 46.02 13.68 Household Yearly Income 23020.07 16751.57



Table 3. Inclusion Criteria Caregiver Must Have:

1. Considered herself/himself one of the primary people responsible for the target child's medication-taking

2. Reported that English was her/his primary language Child Must Have Been:

1. Perinatally HIV-infected

2. Age 2-12 years

3. Prescribed the same anti-retroviral medication for the past 3 months

4. Living with the parent/caregiver during the last month

5. Planning to live with the parent/caregiver for the next three months

6. Not enrolled in an adherence intervention study

7. Not a sibling of an enrolled child




44


one weekend day. Three months following the clinic visit pharmacies were contacted to obtain pharmacy recall histories for the prior three months.

Measures

Adherence Interview-HIV (AI-HIV; Appendix B)

This structured interview was designed by the author to assess typical adherence to prescribed regimens, understanding and knowledge of the prescribed regimen, and barriers to adhering to each aspect of the pediatric HIV regimen. The interview is conducted with caregivers. It was adapted from the Treatment Adherence Questionnaire, which is used to assess adherence to pediatric CF regimens (Quittner et al., 2000a). Due to the novelty of this measure, reliability and validity have not been established.

The interviewer begins by asking parents/caregivers to inform the interviewer of what the family has typically done to treat the child's illness in the last two weeks. Difficulty managing the child's regimen is normalized, in order to provide a safe, nonjudgmental context in which honest responding is encouraged (Murphy & Dillon, 1998). A display card with pictures of each medication and corresponding medication names is used to aid those who may not know the medication names. Each medication is discussed individually, in terms of how often each was taken, when the medication was taken, how much was taken, and who was involved in the medication-taking. After the informant has discussed all of the medications the child has typically taken in the last two weeks, the interviewer asks the informant to report what medications the doctor prescribed, how often they were prescribed, when the medications were prescribed to be taken, how much was prescribed to be taken at a time, and special instructions about food and drink intake. First, the actual disease management behaviors are queried in a non-





45


threatening manner, in order to promote honest responding. Then, the informant is asked to recall the recommended regimen.

For each medication in the recommended regimen, informants are asked how often, when, and within what special food and drink intake guidelines it was recommended to be taken. Also for each medication, informants are asked to identify difficulties with taking each medication, and those things that sometimes keep them from taking the medication as recommended. This allows for identification of which medications may have the most negative medication attributes, and may serve as a cue to remind informants of specific barriers. After the informants discuss all of the medications they believe to be in the child's regimen, they are asked to identify other things, not yet mentioned, that may make medication-taking hard or may keep the child from taking his/her medication as recommended.

Interviews were conducted by trained undergraduate assistants and graduate

students. Interviewers underwent training with the interview procedures and engaged in practice interviews. They were required to achieve 80% agreement with the author's coding of a practice interview before conducting interviews for the study. Demographic Questionnaire (Appendix C)

This questionnaire was designed to gather demographic information on the parent or caregiver, child, and family. The questionnaire asks the caregiver for the date of birth, gender, and ethnicity of both the child and caregiver. Questions assess the caregiver's marital status, education, work status, and household income, as well as HIV status and mode of transmission for those who are HIV positive. Questions about the child include: the age of diagnosis, mode of transmission, date of ART initiation, and whether or not the




46


child knows that he or she is HIV positive. Finally, questions about the family include who lives in the home and who helps the child take his/her medication. Telephone Contact Form (Appendix D)

This form was designed to collect the names of the caregiver and child, primary and secondary telephone numbers, and best times to call. The information was used in conducting the telephone adherence interviews. Medical Record Report Form (Appendix E)

This researcher-completed form was designed by the author to assess the clinical indications of the child's disease severity (CDC status and CD4 count), and the child's currently prescribed medications. The form also includes a nurse rating of the family's adherence to the child's regimen on a 7-point Likert scale (1 = "not at all adherent", 7 = "always adherent"); nurses are asked to rate adherence based on their clinical experiences with each patient during the prior six months. In part, the form was modeled after the Prescribed Treatment Form (PTF), which has been used to identify prescribed treatment for children with CF (Quittner et al., 2000c). Two nurses and one physician from the pediatric HIV clinic at the University of Florida reviewed the form in order to ensure appropriateness of the wording and content. 24-hour Recall Interview (24RI Appendix F)

This measure was adapted from the 24RI used to assess adherence to diabetes regimens (Freund, Johnson, Silverstein, & Thomas, 1991; Johnson, Silverstein, Rosenbloom, Carter, & Cunningham, 1986; Johnson et al., 1992). Although the 24RI for diabetes includes both parent and child report, this study includes only parent report, based on the difficulties children may have recalling details about medication-taking.




47


This measure is also different in this study based on the resulting variables. When used to assess adherence to diabetes regimens, the recall interview yields 13 variables related to insulin injection, blood-glucose checking, diet and exercise. Since the HIV regimen only requires medication-taking and related food and liquid intake, the recall interviews conducted in this study yielded three variables: medication-taking frequency, interval deviance, and dietary adherence. Caregiver reports were compared to the prescribed regimen (as defined in the medical record), except reports regarding medication-specific food and liquid intake, which were compared with instructions on published handouts that list all current ART medications (Abbott Laboratories, 2001). The resulting variables represent percents of agreement between the reported medication-related behaviors and the prescribed regimens.

The final measure modification included an assessment of barriers to adherence. At the end of the recall interview, informants were told that managing a medication schedule is difficult, and many families have difficulties following the regimen. They were asked to list the things that made it difficult to follow the physician's recommendations regarding medication taking the previous day, and the things that kept them from following the recommendations exactly the previous day. The reported barriers were coded for purposes of analyses. This method was designed to capture additional barriers that were not mentioned in the adherence interviews, as it allows families to report actual barriers they faced during a given day.

As with the Adherence Interview-HIV, the interviews were conducted by trained undergraduate assistants and graduate students. Interviewers underwent training of the interview procedures and engaged in practice interviews. They were required to achieve




48


80% agreement with the author's coding of a practice interview before conducting interviews for the study.

Pharmacy Refill History

The validity of pharmacy refills history has been supported by previous studies (Farley et al., 2002), though the reliability of pharmacy refill over time has not been well established. Therefore, like the other assessment methods, this method is somewhat exploratory in nature.

Written informed consent to contact the pharmacies, as well as the list of

pharmacies used by each family was obtained at the initial clinic visit. Pharmacies for each family were contacted and pharmacy refill histories for the three-month period were obtained. Scores were calculated by dividing the intended number of days to one refill (typically 30 days) by the actual number of days to one refill. Values greater than one were scored as 100% adherent, as the families had refilled their prescriptions prior to the date on which the medications would be needed if the family were 100% adherent. If no refill was detected within the 3 month period, the family was scored as 0% adherent.













DATA ANALYSES

Data were analyzed using the Statistical Package for Social Sciences, Version 10.0 (SPSS 10.0). First, in order to test for demographic differences across sites, Chisquare and Analyses of Variance (ANOVAs) were performed. Next, descriptive data were reviewed and the skewness and kurtosis of the outcome variables were examined in order to determine whether or not normality assumptions were met. Adherence frequency data were viewed continuously and then split at cut-offs of 80% and 90%, which were based on the research literature (Farley et al., 2002; Paterson, 2000). The first hypothesis was not tested statistically, as descriptive data were sufficient to support the null hypothesis. For the second hypothesis, a Mann-Whitney U Test was conducted in order to determine whether or not use of the display card was associated with knowledge of medication names. This was determined to be the best test for determining differences between the two groups when the test variable was nonparametric. The third hypothesis was that all adherence measures would correlate. Pearson correlations were performed, though due to the kurtosis of 24RI frequency adherence and inability to transform this variable, it was left out of the correlation matrix. Chi-square analyses were then conducted to test the fourth hypothesis. Adherence cut-offs of 80% and 90% were both included in analyses, in order to determine the best cut-off, and viral load was split to + 400 copies/ml, the highest number associated with undetectable levels of virus in the blood.




49




50


A series of analyses was then conducted to test the fifth set of hypotheses. Due to the non-normality of the knowledge variables (Table 4), knowledge of medication names and dietary requirements were split between <100% and 100% knowledge, as it was expected than any knowledge failure would relate to nonadherence categorization. Chisquare analyses were conducted to determine the relationships of 24RI frequency adherence and refill adherence to caregiver knowledge of medication names. Caregiver knowledge of frequency was eliminated from analyses due to lack of variance. Chisquare analysis was also employed to test the relationship between 24RI dietary adherence and knowledge of dietary requirements. Correlations were then preformed to test the relationship between 24RI interval deviance and interval knowledge.

For the sixth set of hypotheses, both parametric and nonparametric statistics were used. To determine the relationship between 24RI frequency adherence and dosing frequency, Kruskal-Wallis analysis was conducted due to the non-normality of the dependent variable. Next, to determine the relationship between refill adherence and dosing frequency, analysis of variance (ANOVA) was used, as the data were normally distributed. Post-hoc analysis was conducted with the Scheffe test, as it is a conservative post-hoc test. The Mann-Whitney U Test was used to determine the relationship between 24RI frequency adherence and medication formulation, as it was determined to be the best test of differences between two groups when the dependent variable is nonparametric. ANOVA was once again used when the dependent variable was changed to refill adherence.

Barriers to adherence were recorded verbatim and categorized by the author. Descriptive statistics regarding the barrier data were then examined. The Wilcoxon




50





51


Signed Ranks Test was then used to test for differences in the number of barriers reported during the AI-HIV and the 24RI.


















































51













RESULTS

Caregiver Knowledge of the Prescribed Regimen

Descriptive data regarding caregiver knowledge can be found in Table 4. Results of the AI-HIV showed that caregivers' percent knowledge of prescribed medication names ranged from 0% to 100%, with a mean of 86% (SD= .26). Thirty-eight percent consulted their own lists of the child's medication information and 27% consulted the medication display card, while only 39% responded without some aid. Sixty-seven percent of caregivers were able to correctly identify all of their children's medication names. Only one caregiver was unable to correctly identify the prescribed medicationtaking frequency for her children's known medications. When the medication names were known, the percent of medication dosages accurately reported by caregivers ranged from 0% to 100%, with a mean of 83% (SD= .29). Sixty-nine percent of caregivers correctly identified the dosages for all of their children's ART medications. Additionally, when the medication names were known and when the children were prescribed medications with specific dietary requirements, 50% of caregivers correctly identified the specific dietary requirements for their children's medications.

Caregivers were not asked directly about their knowledge of the prescribed dosing interval, as that information was assumed to be inherent in the question about frequency. Nevertheless, caregivers were asked to state the times at which the children typically receive each medication. The reported interval may be useful in understanding




52




53


caregivers' perceptions of the prescribed intervals, as caregiver reports may have been highly influenced by social desirability. For the purpose of analysis, the data were sorted into "interval buckets" indicating the amount of deviation from the prescribed interval. For example, if medication was taken less than 45 minutes before or after the prescribed interval, the interval was scored as "0;" if the medication was taken one hour before or after the prescribed interval had passed, it was scored "1." Intervals were not computed for once daily dosing, as the interview provided only one data point for dosing times of once daily medications.

Descriptive statistics regarding interval deviance are provided in Table 5. Results showed that the majority of doses were reportedly given within the prescribed intervals: 79.3% of the reported typical dosing intervals were less than 2 hours deviant from the prescribed intervals, 88.8% were less than 3 hours deviant from the prescribed intervals, and 96.3% were less than 4 hours deviant from the prescribed intervals. When only the intervals for thrice daily medications were examined, only 40.7% of the reported typical intervals were less than 2 hours deviant, 63% were less than 3 hours deviant, and 88.9% were less than 4 hours deviant.

Relationship between Knowledge and Typical Adherence Behavior

In order to determine whether or not any differences existed between the medications the caregivers believed were prescribed for the children and those the caregivers reported that the children typically received, during the AI-HIV the caregivers were asked to report the medications their children typically had taken in the past two weeks, and the corresponding dosing amounts and dosing frequencies. In all 63 interviews, concordance was 100% between the two reports. Therefore, typical adherence behavior data from the AI-HIV were not included in further analyses.





54


Table 4. Descriptive Statistics for Non-Demographic Study Variables

Variable M SD Median Mode Range Skewness Kurtosis

% Caregiver 86.46 25.18 100.00 100.00 0-100 -2.24 4.96 Knowledge of Medication Names (n= 62)

% Caregiver 84.28 28.90 100.00 100.00 0-100 -1.90 2.8 Knowledge of Medication Dosing (n= 59)

Knowledge of .73 .99 0 0 0-4 1.42 1.62 Interval Deviance (n= 51)

% 24RI Frequency 91.55 14.28 100.00 100.00 43-100 -2.01 3.5 Adherence (I= 55)

24RI Interval 1.00 1.03 .67 0 0- 4.67 1.41 2.47 Deviance (n= 51)

% 24RI Dietary 75.00 30.05 84.00 100.00 0-100 1.02 .01 Adherence (n= 42)

% Refill 62.05 35.74 70.85 100.00 0-100 -.63 -.99 Adherence (11= 59)

% Nurse-reported 5.92 1.22 6.00 7.00 2-7 -1.38 -1.58 Adherencea (n= 63)

Note. Data for percent caregiver knowledge of dosing frequency was not included in this table, as all but one caregiver had correct dosing frequency knowledge. Data for percent caregiver knowledge of dietary requirements was not included in this table because the variable is dichotomous. aScale ranged from 1 "not at all adherent" to 7 "always adherent."





55


Table 5. Typical Percent of Deviance from Prescribed Intervals
Deviance Amount % of twice daily doses % of thrice daily doses % of all doses (n= 166) (n= 27) (n= 193) Less than 1 hour 57.2 25.9 53.7 1-2 hours 25.9 14.8 25.5 2-3 hours 9.6 22.2 9.6 3-4 hours 4.8 25.9 7.4 > 4 hours 2.4 11.1 3.7




56


Utility of Medication Display Card

Twenty-seven percent of caregivers used the medication display card to identify the names of prescribed medication. A Mann-Whitney U Test was conducted in order to determine whether or not the use of the display card was associated with knowledge of the medication names. Display card use was not significantly related to knowledge of medication names.

Adherence Rates

Descriptive statistics based on adherence rates are presented in Table 4. As

measured with the 24RI, adherence to the prescribed medication-taking frequency ranged from 43% to 100% (M= .92, SD= .14), and the modal adherence to frequency was 100%. When the sample was split based on a 24RI frequency adherence cut-off of 80%, 87% of participants were considered adherent. When a cut-off of 90% was used, 75% were considered adherent. For those doses received, average deviation from the prescribed dosing intervals ranged from 0 to 4.67 hours, with a mean of 1 hour, a median of .67 hours (SD= 1.03), and a mode of zero hours. Results showed that 47.3% of the doses given were deviant from prescribed interval by at least one hour, and 17.3% were deviant by at least 2 hours. Also for those doses received and for those medications with specific dietary requirements, the average percent adherence to dietary requirements ranged from 0% to 100%, with a mean of 75% (SD= .30), median of 84%, and mode of 100%. These results varied from the results of the pharmacy refill histories, which showed that adherence to pharmacy refills ranged from 0% to 100%, with a mean of 62% (SD= .36), median of 71%, and mode of 100%. When the sample was split based on a pharmacy




57


refill adherence cut-off of 80%, 46% of participants were considered adherent. When a cut-off of 90% was used, 32% were considered adherent.

Nurses rated each child's adherence, drawing from information in the medical record and information obtained during interactions with the children and their caregivers. The mean nurse report rating was 5.92 (SD= 1.22) on a 7-point Likert scale. The median rating was 6 and the modal rating was 7. No children received a rating of 1"Not at all Adherent."

Relationship between Adherence Measures

Correlations were performed to determine relationships between adherence

measures (Table 6). 24RI frequency was not included in the correlation matrix, due to kurtosis. 24RI interval and dietary adherence ratings did not significantly correlate with pharmacy refill adherence ratings or nurse-reported adherence levels. However, pharmacy refill adherence ratings significantly and positively correlated with nursereported adherence levels (r= .26, p < .05), such that higher pharmacy refill ratings related to higher nurse report ratings of adherence.

Relationship between Adherence and Disease Severity

Chi Square analyses were performed in order to test for relationships between adherence measures and disease severity measures (viral load, CDC letter, or CDC number). When participants were categorized as "adherent (> 80% adherence rating)" and "nonadherent (< 80% adherence rating)" none of the disease severity measures was significantly related with 24RI frequency adherence or refill adherence ratings. When a more stringent criterion (90% adherence) was used, 24RI frequency was significantly and positively related to CDC number (X2= 6.78, p< .05), such that those with higher 24RI




58


frequency adherence were more likely to be classified as at one time having severe CD4 T- Lymphocyte suppression (low white blood cell count or poor immune functioning), indicating greater disease severity. Also when the more stringent criterion was used, refill adherence failed to significantly relate to viral load, CDC letter, or CDC number.

Relationship between Adherence and Knowledge

Several hypotheses were made about the relationships between adherence and

knowledge. First it was hypothesized that 24RI frequency adherence and refill adherence would significantly and positively relate to caregiver knowledge of medication names and frequencies. Frequency knowledge had little variance and, therefore, this measure was not included in analyses. Chi-square analyses were performed. Caregiver knowledge of medication names was not significantly related to 24RI frequency based on thresholds of 80% or 90%, but was significantly and positively related to refill history when a 90% adherence cut-off was used (X2= 8.369, p< .01). Results showed that caregivers with > 90% refill adherence were more likely to know the names of their child's medication. Next it was hypothesized that knowledge of dosing interval would significantly correlate with 24RI dosing interval. This hypothesis was supported (r= .42, p< .001), such that as knowledge of dosing interval deviated from the prescribed dosing interval, so did 24RI dosing interval. Finally, it was also hypothesized that 24RI dietary adherence would positively relate to knowledge of dietary requirements; however, the results of Chi-square analyses were non-significant.

Relationship between Adherence and Prescribed Dosing Frequency

Due to the non-parametric nature of 24RI frequency, Kruskal-Wallis analysis was employed to determine the relationship between 24RI frequency adherence and





59


Table 6. Correlations Between Adherence Measures
24RI interval 24RI Dietary Pharmacy Refill
24RI Interval -- -- -24RI Dietary .04 (n= 38)

Pharmacy Refill -.05 -.29 (n= 47) (n= 38) Nurse-reported -.12 -.09 .26* (n= 51) (n= 42) (n=59)


Note: *= R< .05




60


prescribed dosing frequency. The result was non-significant, suggesting that 24RI frequency adherence did not differ based on prescribed dosing frequency. Next, univariate Analysis of Variance (ANOVA) was conducted in order to determine the relationship between refill adherence and prescribed dosing frequency. Results showed that prescribed dosing frequency was significantly and positively related to refill adherence, F(184, 2)= 3.89, p < .05. Post-hoc analyses using the Scheffe test showed that participants taking medications with twice daily dosing had higher refill rates than participants taking medications with once daily dosing.

Relationship between Adherence and Medication Formulation

Mann-Whitney U analysis was employed to determine the relationship between 24RI frequency adherence and medication formulation. Results showed that formulation was significantly related to 24RI frequency adherence (p< .01), such that medications in pill form were significantly more likely to have high 24RI frequency adherence. These results contrast with results of Univariate Analysis of Variance (ANOVA), which showed that formulation was significantly and positively related to refill adherence, F (185, 2)=

4.9, p< .05. Examination of the means indicates that medications in liquid or powder formulation were significantly more likely to have high refill rates than medications in pill form.

Barriers to Adherence

During the AI-HIV, caregivers reported an average of .89 barriers to their children's adherence (SD = 1.11), with a median of one barrier, and a mode of zero barriers. Barriers were reported verbatim and then categorized. For a complete list of barriers reported, refer to Table 7.





61


When caregivers reported barriers, they were asked about the frequency of the

barrier occurrence and strength of the barrier in preventing medication-taking (e.g., "How often does this happen?" and "How often does this problem keep the child from getting the medicine?"). They were asked to answer each question using a 5-point Likert scale, with 1 indicating "not often at all" and 5 indicating "always." In response to the first question, the mean frequency of occurrence rating was 3.6 (SD= 1.6), with a median of 4 and a mode of 5. In response to the second question, the mean strength rating was 1.6, with both a median and mode of one (SD= 1.2).

Barriers Reported with the AI-HIV Versus the 24RI

The Wilcoxon Signed Ranks Test was conducted to determine whether any significant differences existed in the number of barriers reported during the AI-HIV versus the 24RI. Results showed that the number of barriers reported during the AI-HIV was not significantly different from the number reported during the 24RI. Caregivers cumulatively made 37 barrier reports during the AI-HIV while they made 32 cumulative barrier reports during the three recall interviews.




62


Table 7. Reported Barriers from AI-HIV and 24RI

Barrier Total # of Caregivers Mean Mean Who Reported Barrier Frequency Strength
(*)
Pills are too big to swallow 8 (1) 4.00 2.14 Tastes Bad 8 (3) 4.80 2.00 Child avoids, resists, prolongs, or 7 (4) 3.00 1.00 refuses the medication

Sleeping through dosing time/ 7 (6) 2.00 1.00 medicine interferes with sleep
schedule

Child or family forgets 6 (4) 2.00 1.50 Too many pills to take 6 (2) 4.50 1.50 Needs parental supervision/reminding 5 (3) 5.00 1.00 Side effects 3 (0) 2.00 1.00 Medication-taking interferes with 3 (2) -- -daily activities

Difficulty with the medicine when sick 2 (0) 1.50 1.00 Can't keep medicine down 2 (1) -- -Worry that others will find out about 2 (1) 3.00 1.00 child's HIV status

Child doesn't swallow pills 1 (0) 5.00 1.00 Child hides the medicine 1 (0) 3.00 3.00 I have difficulty traveling to get the 1 (0) 4.00 1.00 refills

I have difficulty getting the refill (no 1 (0) 1.00 1.00 reason stated)

It is too hard to keep up with the 1 (1) medicine




63



Ran out 1 (1) Pharmacy doesn't always have 1 (1) medicine in stock

Difficulty remembering to take pills 1 (1) along when going out

Out of routine 1 (1) Doesn't drink all of the medicine 1 (0) Total number of barrier reports 69 (32) Note. Numbers in parentheses represent the number of barriers reported during 24 RI. Barriers provided during the 24RI were not rated according to frequency and strength.













DISCUSSION

Caregiver Knowledge of the Prescribed Regimen

The first goal of the current study was to document caregivers' knowledge of

medication names, dosage, dosing frequencies, dosing intervals, and medication-specific dietary requirements. Although all but one caregiver correctly identified the prescribed medication-taking frequency of their child's medication, 33% of caregivers were unable to correctly identify at least one of their child's medication names, 31% of caregivers failed to correctly identify the dosage for at least one medication, and half of the caregivers failed to correctly identify the specific dietary requirements for at least one medication. These results suggest that caregivers of children on ART may have more difficulty remembering or identifying medication names and dietary requirements, and less difficulty remembering or identifying medication-taking frequencies than adults on ART (Bangsberg, Bronstone, & Hofman, 2002; Stone et al., 2001). Caregivers may have more difficulty remembering or identifying specific information about medications in their child's regimen because they are less involved in their child's regimen behaviors than adults are in their own regimen behaviors. In the future, it may be worthwhile to examine the relationship between caregiver knowledge and caregiver involvement.

While almost all caregivers exhibited accurate knowledge of dosing frequencies, results indicate that some caregivers may be unaware of the exact intervals at which the doses are to be given. When caregivers were asked to state the times of day at which the children had typically taken their medicine during the previous two weeks, 21% of the


64





65


total reported typical intervals were at least 2 hours deviant from the prescribed intervals. Reported intervals were particularly problematic for thrice daily dosing, in which 59% of intervals for medications were at least 2 hours deviant from the prescribed interval. These results suggest that caregivers may lack knowledge of appropriate dosing intervals or they may have difficulty adhering to the specified intervals when those desired intervals are known, particularly when medications are to be taken thrice daily.

It was also hypothesized that caregiver knowledge of the regimen would differ significantly from reported typical regimen behavior during the previous 2 weeks, as some caregivers would actively decide not to give their children some medications. Given the nature of the AI-HIV, which first asked about typical behavior, and then asked about knowledge of the prescribed regimen, it was thought that caregivers would feel free to disclose any such differences. Nevertheless, caregivers failed to report any differences at all between typical regimen behaviors and their knowledge of the prescribed regimen. It seems likely that even with interviewer efforts to normalize adherence problems, caregivers responded to the social desirability of the pediatric HIV clinic context. These data are consistent with other self-report clinical data from caregivers of HIV-infected children, in which a tendency to report 100% adherence was found, despite MEMS data that suggests poorer adherence (Farley et al., 2002); however, these data suggest that caregiver overestimation of their child's ART adherence may be more extensive than caregiver and child overestimation of adherence to other chronic illness regimens (Delamater, 2000; Festa et al., 1992; Passero et al., 1981; Quittner et al., 2000c; Rapoff, Lindsley, & Christopherson, 1985; Schoni et al, 1995). Perhaps the pediatric HIV clinic environments foster reporting of socially desirable behavior more so than the diabetes or




66


CF clinic evironments. Providers in HIV clinics typically expect high levels of adherence (80-95%; Farley et al., 2002; Paterson et al., 2000), and often focus on adherence extensively, because the consequences of even occasional deviance from the prescribed regimen can be severe; unfortunately, the extensive focus on adherence may reinforce the reporting of adherent behaviors, even when the reports are inaccurate. In the future, it may be worthwhile to examine patient-provider interactions to assess the extent to which providers may inadvertently reinforce overreporting of adherence behaviors. Additionally, it may be worthwhile to modify the self-report adherence assessment of the AI-HIV in order to promote more honest responding.

Another study aim was to document the utility of the medication display card for caregivers' identification of medications during the AI-HIV. Caregivers were given the option to use the display card if they had difficulty generating medication names, and 27% of caregivers exercised that option. It was hypothesized that knowledge of medication names would not differ significantly based on use of the display card, because the use of the display card would eliminate most problems in medication identification. This hypothesis was supported by the present study, and suggests that the display card is a useful component of the AI-HIV, as it helps to trigger recall, yet fails to lead the informant to a particular response.

Adherence Rates

The study also aimed to document rates of adherence to prescribed medicationtaking frequencies, intervals, and medication-specific dietary requirements. Regarding medication-taking frequencies, results vary greatly based on the mode of assessment, with higher adherence ratings resulting from the 24RI than pharmacy refill, and generally




67


high adherence ratings resulting from nurse-report. The rate from the 24RI (92% average adherence to frequency) suggests lower adherence than caregiver self-reported adherence in Farley et al. (2002), which found that all caregivers denied that their child had missed any doses within the previous 3 days; however, the present refill rates (75% average adherence) suggest higher adherence than the Farley et al. (2002) results of pharmacy refill histories and MEMS monitoring, which found average adherence ratings of 46% and 54%, respectively. Thus, the results of the present study suggest that when compared to pharmacy refill histories, the 24RI tends to overestimate adherence to ART frequency, but it may overestimate ART adherence less than other methods of self-report. When these results are compared to child adherence to diabetes and CF regimens, these results suggest that caregivers of HIV-infected children may overestimate adherence significantly more than caregivers of children with other chronic illnesses (Quittner et al., 2000b). As mentioned previously, the expectations of high adherence and extensive focus on adherence in pediatric HIV clinics may inhibit accurate self-report.

The 24RI also provided data about dosing intervals and dietary adherence. Data showed that a significant portion of the sample (17.3%) had an average dosing deviation of> 2 hours from the prescribed interval, while a much larger portion (47.3%) had an average deviation of > 1 hour. These results are consistent with the typical intervals reported during the AI-HIV in this study.

The third study hypothesis was that all adherence measures would correlate significantly and positively. The results of the study fail to support this hypothesis, as only pharmacy refill and nurse-report ratings were significantly and positively correlated. Despite the hypothesis, it is somewhat intuitive that 24RI interval and dietary adherence





68


would not strongly relate to measures of adherence to frequency, since it is one thing to examine whether or not the dose was ever received, and another thing to examine whether or not the dose was received within certain constraints. Furthermore, the lack of significant relationships between adherence measures is consistent with the results of another study that compared caregiver self-report of HAART adherence to pharmacy and MEMS data (Farley et al., 2002) and a study that compared self-report ART adherence rates to pharmacy and MEMS data among adults (Frick, Gal, Lane, & Sewell, 1998). These results are also similar to the results of studies of relationships between different adherence assessment methods among caregivers of children with other chronic illnesses (Glascow et al., 1987; Johnson et al., 1986). Finally, the significant and positive relationship between pharmacy refill data and nurse-report ratings is unsurprising, as it is likely that nurse ratings were influenced by nurses' knowledge of pharmacy refill data, as nurses have reported previously that pharmacy refill records obtained during routine clinical care were influential in nurses' ratings of patients' adherence (Farley et al., 2002).

The author also hypothesized that medication adherence would significantly and negatively relate to disease severity. Nevertheless, when an 80% adherence cut-off was used, viral load, CDC letter, and CDC number classification were not significantly related with 24RI adherence ratings or refill ratings. Ninety-percent refill adherence classification did not vary based on disease severity; however, 90% 24RI frequency adherence classification did vary based on CDC number classification, such that those who had ever had low CD4 counts (and thus, higher CDC number classifications) tended to be classified as adherent more than those with historically higher CD4 counts (and




69


thus, lower CDC number classifications). These results suggest that caregivers of children with greater CDC number classification may feel greater stimulus demand to report more adherent behaviors than caregivers of children with lower CDC number classification, even when such reporting is inconsistent with refill behavior.

Viral load was not significantly different based on adherence categorization, a finding that is inconsistent with some findings which have shown that self-reported adherence was significantly and negatively associated with viral load among adults (Duong et al., 2001; Knobel et al., 2002; Mannheimer et al., 2002; Wagner et al., 2001; Walsh, Mandalia, & Gazzard, 2002) and among children (Katko et al., 2001). Several factors may help to explain why viral load was not significantly related to adherence in this study. First, viral load is impacted by adherence over time, takes time to adjust to changes in adherence, and is also highly impacted by drug resistance, which can occur over time even when adherence is generally consistent. Also, viral load was only assessed at baseline; therefore, no follow-up data were available in order to compare adherence behavior to viral load as an outcome variable. Another possibility may be that the cutoff of 90% adherence may not be sufficient for virologic success.

The researcher also hypothesized that adherence would significantly and positively relate to knowledge. It was hypothesized that 24RI frequency and refill adherence would relate to caregiver knowledge of medication names and dosing frequencies. Although neither adherence measure was significantly associated with dosing frequency knowledge, caregivers with 100% knowledge of medication names were significantly more likely to have > 90% refill adherence than those with imperfect knowledge of medication names. This is an important finding, as it suggests that




70


clinicians may be able to gain valuable information about adherence simply by asking caregivers to state the medications that their children are currently taking. Nevertheless, it remains unclear whether caregivers are more adherent because they know the medication names, or they know the medication names because they are adherent. Further investigation is necessary in order to determine to what extent caregivers' lack of knowledge is due to lack of effective regimen-specific education, lack of regimen behaviors, and lack of caregivers' involvement in the regimen behaviors.

AI-HIV interval knowledge was hypothesized to correlate significantly and

positively with 24RI interval. The present study supports this relationship, indicating that caregiver reports of the typical intervals between dosing over a 2-week period tend to increase as caregiver reports of the typical intervals between dosing over (3) one-day periods increase. However, it is important to recognize that AI-HIV interval knowledge is a proxy measure of interval knowledge, since caregivers are not directly asked to state how many hours that they believe should pass between each dose.

It was also expected that 24RI dietary adherence would relate to knowledge of dietary requirements, though this hypothesis is not supported by the current study. This finding suggests that even when dietary requirements are known, children may not adhere to the dietary component of their medication regimens. This is consistent with findings from a review of dietary adherence among children with chronic illness (Mackner, McGrath, & Stark, 2001), which showed that dietary adherence among chronically ill children tends to hover at or below 50%, and knowledge of the dietary regimen is insufficient for dietary adherence.




71


It was also hypothesized that adherence frequency measures would significantly relate to prescribed dosing frequency and formulation. Refill adherence was significantly associated with prescribed dosing frequency, suggesting that medications prescribed for twice daily dosing are significantly more likely to be refilled than medications prescribed for once daily dosing. This result is interesting, given that once daily dosing is often preferred over twice and thrice daily dosing (Rosenbach, Allison, & Nadler, 2002). The small number of children in the study who were prescribed thrice daily dosing likely explains why thrice daily dosing was not negatively associated with adherence, though the higher refill rates among twice daily dosing versus once daily dosing are particularly surprising, as all children with once daily medications also have twice daily medications; therefore, it would seem easy for children or caregivers just to add the extra pill or pills once a day, or at least refill the once daily medications while refilling twice daily medications. One plausible explanation for this finding is that once-daily medication, at least when taken along with twice-daily medication is, in fact, easy to forget, as it is not taken as frequently as the other medication. Another explanation is related to the dietary requirements. Common once-daily medications are didanosine and efavirenz. Didanosine is to be taken on an empty stomach, at least 30 minutes before or 2 hours after eating, and efavirenz is not to be taken with a high-fat meal (Abbott Laboratories, 2001). It may be that sometimes it is too difficult to follow the dietary requirements, and therefore, an entire dose is missed (Roberts & Mann, 2000).

Both 24RI frequency and refill adherence were significantly related to medication formulation, though the results were in opposite directions. 24RI frequency adherence was generally higher for pills, though refill adherence was higher for liquid and powder




72


formulations. These findings may be due to caregivers giving greater responsibility to children taking pills, children then forgetting to take the pills or deliberately returning the pills to the container, and children then forgetting or neglecting to inform the caregiver of the need for refills. Caregivers may be more directly involved in the administration and refill of liquid medications, therefore, their reports regarding adherence to liquid medications may be more accurate.

Barriers to Adherence

Another goal of the present study was to document the perceived barriers to

adherence among caregivers of children on ART. Caregivers were asked to consider the past two weeks and report things that had made it difficult for the child to get their medication 'exactly as prescribed.' On average, caregivers reported about one barrier each, which is far lower than might be expected, given the large percentage of caregivers defined as nonadherent based on pharmacy refill rates of less than 80%. However, an examination of the self-report data serves as a reminder that barrier reporting may be strongly impacted by social desirability. Both adherent and nonadherent caregivers may be concerned that reporting hardship or problems with the regimen would lead the researcher to conclude that their child is not getting the medicine. Due to unawareness, other caregivers may be unable to articulate the barriers they experience, particularly in cases in which anger, guilt, depression, or denial prevents adherence, or in other cases, in which children lie about taking their medication. Other caregivers may experience difficulty remembering the problems that occur. It also remains possible that some families do not experience barriers to ART adherence, a phenomenon that was suggested in the present study by caregivers who reported that they experienced many problems





73


when they initially began the regimen, but that often with the help of healthcare providers, solutions were found and problems were eliminated. Finally, because children with the greatest adherence barriers may quickly develop drug resistance, and thus, frequently change medication regimens, children with the greatest adherence barriers may have been excluded from this study based on the inclusion criterion that children must have been prescribed the same medication for the previous 3 months.

Twenty-two different barriers were reported between the AI-HIV and 24RI.

Reported barriers generally related to medication-specific attributes, such as the size or taste of pills, problems with scheduling or routine, or problems with the child resisting, refusing, or hiding the medication. With the exception of the child-specific problems, the barriers reported in the current study have been reported in the adult ART literature, as well (Ammassari et al., 2001; Catz et al., 2000; Chesney, 1997; Mannheimer et al., 2002; Murphy et al., 2000; Walsh et al., 2001). Caregivers in the present study did fail to mention some barriers that have been highly endorsed by HIV-infected adults; in particular, emotional/psychological barriers such as being angry (Murphy et al., 2000) or depressed (Chesney, 1997; Murphy et al., 2000), or not wanting to be reminded of HIV (Catz et al., 2000) were not mentioned. Also, difficulty with the dietary restrictions and timing of doses (Catz et al., 2000; Mannheimer et al., 2002; Murphy et al., 2000; Roberts & Mann, 2000; Walsh et al., 2001) were not mentioned as barriers.

When caregivers reported barriers, they were asked to rate the frequency of

occurrence and the strength of barriers in preventing medication-taking. Many caregivers reported high barrier frequency ratings, and low barrier strength ratings were almost always given. These results imply that although families may experience many things




74


that make children's medication-taking difficult, ultimately, these barriers rarely prevent children from receiving doses of medication.

The present study fails to support the hypothesis that caregivers would report

more barriers during the 24RI procedure than during the AI-HIV. It was thought that the 24RI would be more effective than the AI-HIV in eliciting barrier reports because the diary format of the 24RI would help conjure memories of recent barriers. Although the 24RI did not result in a greater number of barriers reported, qualitative analysis showed that individual caregivers tended to report different barriers during the 24RI than during the AI-HIV. Therefore, it can be concluded that both the daily diary format and the clinical interview format are effective methods for obtaining information about barriers to adherence.

Implications

Overall, this study suggests that the AI-HIV, 24RI, and pharmacy refill histories are valuable means of assessing adherence to ART regimens among HIV-infected children. This study demonstrates that the AI-HIV is a useful tool for measuring knowledge of the prescribed regimen, including knowledge about medication names, dosages, dosing frequencies, dosing intervals, and dietary requirements, and for collecting information about barriers to adherence. In particular, using the AI-HIV to assess knowledge of prescribed medication names may help clinicians to detect overall adherence problems, and using the AI-HIV to assess knowledge of prescribed dosing intervals may help clinicians detect problems with dose timing. Nevertheless, the AIHIV may require some modifications in order to improve its utility. Although the format of inquiring first about the typical adherence behaviors and then about the prescribed




75


regimen is useful in assessing knowledge, the resulting information about daily adherence is limited and did not correlate with other measures of adherence in this study (24RI and pharmacy refill history). Possible improvements to the measure may include the addition of children as informants, a statement emphasizing the importance of honest responding, and the adaptation of the measure into a computerized structured interview, a format which has been found to provide more accurate responding than provider interview (Bangsberg et al., 2002). Additionally, the measure should be altered to inquire directly about knowledge of prescribed dosing intervals, rather than using a proxy measure of interval knowledge. Once these modifications are made, the AI-HIV will be a valuable brief assessment measure with which clinicians and researchers can assess both child and caregiver knowledge of the prescribed ART regimen.

Results suggest that the 24RI may grossly overestimate adherence to frequency when compared with pharmacy refill history, though the 24RI is unique in its ability to assess adherence to prescribed dietary requirements and to assess adherence barriers experienced over a short period of time. The present study suggests that the 24RI provides useful information about daily dietary adherence behaviors above and beyond an assessment of knowledge in the clinical setting. The 24RI also appears to be an effective tool for assessing barriers to adherence, as it appears to help conjure memories of barriers encountered in a given day. The apparent 24RI overestimation of adherence in this study suggests that some modifications may be necessary in order to increase the measure's clinical utility. In this study, 24RI interviewers asked caregivers to describe their child's previous day, and when medication-taking was mentioned by the caregivers, interviewers recorded information about the medication names, dosing, whether or not




76


the behavior was observed, and at what time the behavior occurred. Eliminating queries about this specific information may help to decrease the face validity and associated social desirability of the measure. Future studies should aim to test the utility of the measure once these modifications are made, and shouldexamine the reliability and validity of the 24RI using both child and caregiver reports. Once the reliability and validity of the measure are supported, clinicians may consider administering the 24RI to children who are prescribed ART and their caregivers. Clinicians may wish to administer the 24RI in the clinical setting or over the telephone at random intervals.

The present study also adds to the existing literature by confirming the value of the pharmacy refill methodology for assessing adherence to ART regimens among HIVinfected children. Pharmacy refill histories appear to provide more conservative, and seemingly more accurate adherence estimates than self-report. Thus, both clinicians and researchers should continue to consider using pharmacy refill histories as an inexpensive methodology for obtaining adherence estimates to supplement self-report data.

The present study also has implications for interventions designed to improve adherence among HIV-infected children on ART. First, by demonstrating that between 57% and 68% of children may receive suboptimal levels of ART, this study supports the widespread belief that many children are not adherent to ART regimens, and thus, supports the need for adherence interventions in this population. Next, this study points to possible targets for intervention. Results of the AI-HIV and 24RI suggest that improving knowledge about the prescribed regimen, including information about appropriate dosing intervals and dietary requirements, is one possible intervention target implicated in this study. Results of the barrier assessments suggest at least two additional




77


intervention targets. First, the barrier assessments suggest that drug manufacturers should work to modify medication attributes in order to decrease the size of pills, improve the taste of both pills and liquids, reduce the number of pills required, and generally simplify the regimen. Second, the barrier assessments suggest that healthcare providers, including physicians, nurses, pharmacists, social workers, and psychologists, should work with families to help eliminate medication scheduling problems, and educate families on the use of appropriate strategies to observe medication-taking, help remember the medication schedule, and help improve children's cooperation with medicationtaking.

Finally, results of the study imply that pediatric HIV differs significantly from other childhood chronic illnesses. The differences between pediatric HIV and other childhood chronic illnesses may be extensive, though the literature has not systematically examined them. Two such differences emerged in this study. First, results of this study imply that caregivers of HIV-infected children may be more susceptible to socially desirable adherence self-reporting than caregivers of children with CF (Quittner et al., 2000b). Caregivers of HIV-infected children may be more inclined towards overestimating adherence because of the extensive focus on adherence that is typical in HIV clinic environments, and may be more intensive than in other clinic environments. This extensive adherence focus is likely fostered by clinicians based on knowledge of the detrimental effects of even occasional lapses in adherence-consequences that are unmatched in most other childhood chronic illnesses. Caregivers of HIV-infected children also may be uncomfortable disclosing adherence problems for fear that the providers will report them to social services for medical neglect, while caregivers of





78


children with other chronic illnesses may not share that fear. Moreover, caregivers who are HIV-infected themselves may fail to report lower levels of adherence due to additional concerns that reports of poor adherence may reflect their poor adherence to their own regimens. Another difference may be that caregivers of HIV-infected children, a group that includes more African-Americans than some other chronic illesses, may be uncomfortable reporting adherence problems to Caucasian interviewers. Secondly, even the most conservative results of the present study suggest that children on ART achieve higher adherence levels than children with other chronic illness regimens (Rapoff, 1999). Children with HIV often live in economic and environment conditions that are less stable, safe, and secure than environments of children with other chronic illnesses, and some HIV-infected children live with HIV-infected caregivers who have their own medical concerns, or with foster parents or distant family members who may have other significant demands on their time (e.g., their own medical regimens and regimens of other children in the home); yet these results imply that HIV children and their families, and perhaps their providers, are actually doing a better job adhering than those who may have fewer psychosocial stressors. The HIV-infected group may do better because the regimen includes fewer components than regimens for diabetes and CF, because the patient-provider communications about the regimen are more constructive, or due to some combination of these and other factors. Regardless of the reasons for the better adherence in the HIV-infected group, this is an important observation as it demonstrates the resilience and adaptability of HIV-infected children and their families. Furthermore, taken together, the results emphasize the importance of examining adherence to





79


individual childhood chronic illnesses, and striving to understand the specific environmental contexts that may promote or inhibit adherence.

Study Strengths

This study has many strengths. First, this was the first known study to assess

knowledge of the prescribed regimen among caregivers of children on ART. Thus, this study fills a significant gap in the literature, and points clinicians and researchers to valuable information about an important first step in adherence intervention: educating children and caregivers about prescribed regimens.

Second, this study is one of few known studies to assess adherence to pediatric ART regimens, and one of only two known studies to assess adherence to pediatric ART regimens multidimensionally. Moreover, this study has successfully replicated the findings of another recent study that measured adherence to pediatric ART regimens and found over half to be less than 80% adherent to refills. Therefore, this study adds to the literature by providing empirical support for the need for effective interventions to improve children's adherence to ART.

Third, this study represents the first known qualitative assessment of barriers to children's ART adherence. The resulting list of caregivers' perceived barriers will be helpful in developing a barriers chescklist that could be used to quickly assess barriers in the clinical setting.

This study has also introduced two new tools for ART adherence assessment: the AI-HIV and the 24RI. The utility of the AI-HIV in assessing knowledge and barriers and the utility of the 24RI in assessing adherence to prescribed dosing intervals, dietary requirements, and barriers have both been supported by this study. In the future, these




80


measures may become important components of multidimensional HAART adherence assessment in both the clinical and research domains.

Finally, this study adds to the existing literature regarding childhood chronic

illnesses, as the results, which are somewhat disparate from previous findings related to adherence in other chronic illnesses, help to demonstrate the variability of disease management behaviors across different illnesses. Findings also demonstrate that despite variability in disease management behaviors, instruments that have been developed for use in measuring adherence to one chronic illness may be useful in assessing adherence to other chronic illnesses with other target behaviors; nevertheless, such instruments may require modification in order to fit with specific populations and regimens.

Study Limitations

Despite the many strengths of the current study, several weaknesses should be considered, as they may limit the interpretation of the findings. The first group of considerations pertains to selection bias. First, several groups were excluded from the study by not meeting inclusion criteria. Children concurrently enrolled in adherence intervention studies and those with unstable home environments and unstable medication regimens, foster caregivers, and caregivers for whom English was a second or other language were all excluded from the study. Therefore, the results presented may not generalize to these groups. Second, participants were recruited from 3 urban southeastern medical centers in the United States and thus, results may not generalize to other regions nationally or internationally. Also, 14% of eligible participants approached refused to participate in the study, several subjects failed to complete the 24RI, and pharmacy refill data were not retrievable for one subject. Self-selection bias may have occurred. It may





81


be that people with lower adherence were less likely to participate or less likely to complete the 24RI. If that is true, then the reported rates of adherence may overestimate adherence among the pediatric HIV population.

The next group of considerations relates to adherence measurement. Although

problems exist with all adherence measurement techniques, electronic monitoring devices have become a preferred method for measuring adherence to HAART. This study was unable to use electronic monitoring because of cost, and instead assessed adherence behaviors with clinical interview, a caregiver self-report recall interview, nurse report, and pharmacy refill histories. Historically, self-report and provider assessment of adherence tend to grossly overestimate adherence (Rand & Weeks, 1998), though pharmacy refill history has been found to strongly relate to MEMS findings (Farley et al., 2002). Thus, the pharmacy refill history appears to be a good alternative in the absence of MEMS data.

This study also neglected children as important sources of adherence information. Researchers have previously demonstrated the utility of obtaining adherence information from both children and caregivers (Johnson, 1995); therefore, the absence of child-report measures is a weakness of the present study. Child-report is particularly important as children become older and more responsible for medication-taking. Additionally, childreports of barriers is important for constructing an accurate model on which to base future interventions. This study completes an important step in the documentation of barriers to children's ART adherence, though child-report of adherence behaviors and barriers will be important in future research.





82


Finally, this study is limited by the small sample size. Ideally, at least 100

caregivers of HIV-infected children would have completed this study, thereby increasing the statistical power to support the hypotheses. Due to the relatively small numbers of children followed at each clinic (60-140) and the exclusion criteria set for the study, only 63 families were recruited. As a result of the small sample size, it was impossible compare adherence across all the different medications. Although the small sample size did limit the power of the study, this study adds important information to the literature regarding caregiver knowledge of the prescribed regimen, adherence rates, perceived barriers to adherence, and assessment of knowledge, adherence to interval and dietary requirements, and barriers to adherence among HIV-infected children and their caregivers and other children with chronic illness and their caregivers.

Future Directions

The current study has begun to close a significant gap in the literature pertaining to children's adherence to ART. In particular, this study has provided information suggesting that the AI-HIV and 24RI are useful measures for assessing caregiver knowledge, adherence behavior, and caregiver-perceived barriers to adherence among this population. Future studies should examine ways to maximize the utility and determine the reliability and validity of these instruments, and specifically, convergent validity analyses should consider both patient-specific goals of ART and patient levels of drug resistance as they relate to virologic success. Studies should also assess the utility of the AI-HIV and 24RI in assessing ART adherence based on child self-report, and caregiver and child-report combined.





83


The present study has also documented caregivers' perceived barriers to ART adherence. The next step will be to conduct a similar assessment based on child-report. The resulting lists should be combined with barrier lists used in the adult literature to devise a barriers checklist that can be used to quickly assess adherence in the clinical setting.

This study has also provided important information about significant deficits in caregiver regimen knowledge and pharmacy refill adherence, and daily barriers to regimen adherence. Efforts should now be undertaken to develop effective adherence interventions that focus on increasing regimen knowledge, increasing adherence behaviors, and decreasing barriers to adherence among HIV-infected children and their families. Researchers should also continue to work to identify other important targets for intervention. This study has begun to clarify some of the issues important to ART adherence among children, but it is only the beginning. Future studies should attempt to clarify these findings, and should conduct research into the role of child responsibility for medication-taking, and the relationship between child responsibility and adherence.

Finally, perinatally HIV-infected children on ART are quickly growing into

adolescents and young adults. The literature regarding adolescents' adherence to chronic illness regimens suggests that we must prepare for decreases in adherence during this developmental stage. The impact of HIV status on adolescents' identity and sexual experiences may further inhibit medication adherence behavior; therfore, including risk behaviors in the assessment of adherence among HIV-infected youth may add valuable information. Much research is necessary in order to inform clinicians about the most effective disease management during this tumultuous period, particularly given the severe





84


consequences of poor adherence and subsequent drug resistance. Clinicians and researchers must ask themselves and adolescents: "Given an adolescent's personal attributes, support system, and home environment, what can we expect in terms of ART adherence?"













REFERENCES

Abbott Laboratories (2001). My Medication Schedule. North Chicago, IL: Abbott Laboratories, Inc.

Alessandro, F., Vincenzo, Z.G., Marco, S., Marcello, G., & Enrica, R. (1994).
Compliance with pharmacologic prophylaxis and therapy in bronchial asthma. Annals of Allergy, 73(2), 135-40.

American Academy of Pediatrics (1993). Psychosocial risks of child health conditions in childhood and adolescence. Pediatrics, 92(6), 876-878.

Ammassari, A., Murri, R., Pezzotti, P., Trotta, M.P., Ravasio, L., DeLongis, P., LoCaputo, S., Narciso, P., Pauluzzi, S., Carosi, G., Nappa, S., Piano, P., Izzo, C.M., Lictner, M., Rezza, G., Montforte, A.A., Ippolito, G., Moroni, M., Wu, A.W., & Antinori, A. (2001). Self-reported symptoms and medication side effects influence adherence to highly active antiretroviral therapy in persons with HIV infection. Journal of Acquired Immune Deficiency Syndromes, 28, 445-449.

Anthony, H., Paxton, S., Bines, J., & Phelan, P. (1999). Psychosocial predictors of adherence to nutritional recommendations and growth outcomes in children with cystic fibrosis. Journal of Psychosomatic Research, 47(6), 623-634.

Bangsberg, D.R., Bronstone, A., & Hofman, R. (2002). A computer-based assessment detects regimen misunderstanding and nonadherence for patients on HIV antiretroviral therapy. AIDS Care, 14(1), 3-15.

Bangsberg, D.R., Hecht, F.M., Charlebois, E.D., Zolopa, A.R., Holodniy, M.,
Sheiner, L., Bamburger, J.D., Chesney, M.A., & Moss, A. (2000). Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS, 14, 357-366.

Bangsberg, D.R., Hecht, F.M., Clague, H., Charlebois, E.D., Ciccarone, D.,
Chesney, M., & Moss, A. (2001). Provider assessment of adherence to HIV antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes, 26, 435-442.

Beck, D. E., Fennell, R. S., Yost, R. L., Robinson, J. D., Geary, D. & Richards, G. A. (1980). Evaluation of an educational program on compliance with medication regimens in pediatric patients with renal transplants. The Journal of Pediatrics, 96(6), 1094-1097.



85




86


Belzer, M. E., Fuchs, D.N., Luftman, G.S., & Tucker, D.J. (1999). Antiretroviral adherence issues among HIV-positive adolescents and young adults. Journal of Adolescent Health, 25, 316-319.

Berg, J. S., Dischler, J., Wagner, D. J., Raia, J., & Palmer-Shevlin, N. (1993). Medication compliance: A health care problem. The Annals of Pharmacotherapy, 27(supplement), 2-21.

Birkhead, G., Attaway, N.J., Strunck, R.C., Townsend, M.C., & Teutsch, S. (1989). Investigation of a cluster of deaths of adolescents from asthma: Evidence implicating inadequate treatment and poor patient adherence with medications. Journal of Allergy and Clinical Immunology, 84, 484-491.

Boccuti, L., Celano, M., Geller, R.J., & Phillips, K.M. (1996). Development of a scale to measure children's metered dose inhaler and spacer technique. Annals of Allergy, Asthma, & Immunology, 77, 217-221.

Bogart, L. M., Catz, S.L., Kelly, J.A., Gray-Bernhardt, M.L., Hartmann, B.R.,
Ptto-Salaj, L.L., Hackl, K.L., & Bloom, F.R. (2000). Psychosocial issues in the era of new AIDS treatments from the perspective pf persons living with HIV. Journal of Health Psychology, 5(4), 500-516.

Brettle, R.P., Wilson, A., Povey, S., Morris, S., Morgan, R., Leen, C.L.,
Hutchinson, S., Lewis, S., & Gore, S. (1998). Combination therapy for HIV: the effect on inpatient activity, morbidity and mortality of a cohort of patients. International Journal of STD & AIDS. 9(2), 80-87.

Brownbridge, G., & Fielding, D. M. (1994). Psychosocial adjustment and adherence to daily treatments regimens. Pediatric Nephrology, 8, 744-749.

Brownlee-Duffeck, M., Peterson, L., Simonds, J.F., Goldstein, D., Kilo, C., & Hoette, S. (1987). The role of health beliefs in the regimen adherence and metabolic control of adolescents and adults with diabetes mellitus. Journal of Consulting and Clinical Psychology, 55(2), 139-144.

Butz, A. M., Joyner, M., Greenberg-Friedman, D., & Hutton, N. (1998). Primary care for children with human immunodeficiency virus infection. Journal of Pediatric Health Care, 12, 10-19.

Carpenter, C., Cooper, D.A., Fischl, M.A., Gatell, J.M., Gazzard, B.G., Hammer, S.M., Hirsch, M.S., Jacobsen, D.M., Katzenstein, D.A., Montaner, J., Richman, D., Saag, M.S., Schechter, M., Schooley, R.T., Thompson, M.A., Vella, S., Yeni, P.G., & Volberding, P.A. (2000). Antiretroviral therapy in adults. Journal of the American Medical Association. 283(3), 381-391.





87


Cassell, E.J. (1991). The Nature of Suffering and the Goals of Medicine. New York, NY: Oxford University Press.

Catz, S. L., Kelly, J.W., Bogart, L.M., Benotsch, E.G., & McAuliffe. (2000).
Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychology, 19(2), 124-133.

CDC- Centers for Disease Control (1996). AIDS among children United States, 1996. Morbidity and Mortality Weekly Report, 1005-1010.

CDC- Centers for Disease Control (1998, November). What is HIV? Retrieved August 16, 2002, from http://www.cdc.gov/hiv/pubs/faq/faql.htm

Celano, M., Geller, R.J., Phillips, K.M., & Ziman, R. (1998). Treatment
adherence among low-income children with asthma. Journal of Pediatric Psychology, 23(6), 345-349.

Chesney, M. (1997, June). New antiretroviral therapies: Adherence challenges and strategies. Presented at the Adherence in HIV Therapy: Implications for Nursing Practice Conference, San Francisco.

Conley, S.B., & Salvatierra, O. (1996). Noncompliance among adolescents: Does it impact the success of transplantation? Nephrology News & Issues, 10, 18-19.

Delamater, A.M. (2000). Critical issues in the assessment of regimen adherence in children with diabetes. In D. Drotar (Ed.), Promoting Adherence to Medical Treatment in Chronic Childhood Illness (pp. 173-195). Mahwah, NJ: Lawrence Earlbaum Associates.

DiGirolamo, A. M. Quittner, A. L., Ackerman, V, & Stevens, J. (1997).
Identification and assessment of ongoing stressors in adolescents with a chronic illnes: an application of the behavior-analytic model. Journal of Clinical Child Psychology, 26(1), 53-66.

DiMatteo, M.R., & DiNicola, D.D. (1982). Achieving Patient Compliance: The Psychology of the Medical Practitioner's Role. New York, NY: Pergamon Press.

Duong, M., Pirothe, L., Peytavin, G., Forte, F., Kohli, E., Grappin, M., Buisson, M., Chavanet, & Portier, H. (2001). Value of patient self-report and plasma human immunodeficiency virus protease inhibitor level as markers of adherence to antiretroviral therapy: Relationship to virologic response. Clinical Infectious Diseases, 33, 386-392.

Ettenger, R.B., Rosenthal, J.T., Marik, J.L., Malekzadeh, M., Forsythe, S.B., Kamile, E.S., Salusky, I.B., & Fine, R.N. (1991). Improved cadaveric renal transplant outcome in children. Pediatric Nephrology, 5, 137-142.




88


Farley, J.J., Hines, S., Musk, A., Ferrus, S., & Tepper, V. (2002). Assessment of adherence to antiviral therapy in HIV-infected children using the Medication Events Monitoring System (MEMS), pharmacy refill, provider assessment, caregiver self-report, and appointment keeping. Manuscript submitted for publication.

Fatkenhauer, G., Theisen, A., & Rockstroh, J. (1997). Virological treatment failure of protease inhibitor therapy in an unselected cohort of HIV-infected patients. AIDS, 11, F11-F116.

Feinstein, A.R. (1974). Clinical biostatistics. XXX. Biostatistical problems in 'compliance bias'. Clinical Pharmacology and Therapeutics, 16(5), 846-847.

Festa, R.S., Tamaroff, M.H., Chasalow, F., & Lanzkowsky, P. (1992).
Therapeutic adherence to oral medication regimens by adolescents with cancer: I. Laboratory assessment. Journal of Pediatrics, 120, 807-811.

Finney, J.W., Hook, R.J., Friman, P.C., Rapoff, M.A., & Christophersen, E.R. (1993). The overestimation of adherence to pediatric medical regimens. Children's Health Care, 22(4), 297-304.

Freund, A., Johnson, S.B., Silverstein, J., & Thomas, J. (1991). Assessing daily management of childhood diabetes using 24-hour recall interviews: Reliability and stability. Health Psychology, 10(3), 200-208.

Frick, P.A., Gal, P., Lane, T.W., & Sewell, P.C. (1998). Antiretroviral
medication compliance in patients with AIDS. AIDS Patient Care and STD's, 12(6), 463470.

Friedland, G.H., & Williams, A. (1999). Attaining higher goals in HIV treatment: The central importance of adherence AIDS, 13 (Supplement 1), S61-S72.

Gallant, J.E. (2000). Strategies for long-term success in the treatment of HIV infection. Journal of the American Medical Association, 283(10), 1329-1334.

Gerbino, P. (1993). Forward. Annals of Pharmacotherapy, 27, S3-S4.

Glasgow, R.E., McCaul, K.D., & Schafer, L.C. (1986). Barriers to regimen adherence among persons with insulin-dependent diabetes. Journal of Behavioral Medicine, 9(1), 65-77.

Goldfried, M. R. & D' Zurilla, T.J. (1969). A behavior-analytic model for
assessing competence. Current topics in Clinical Child Psychology (Vol. 1, pp. 151-195). New York, NY: Academic.





89


Gudas, L. J., Koocher, G.P., & Wypij, D. (1991). Perceptions of medical
compliance in children and adolescents with cystic fibrosis. Journal of Developmental and Behavioral Pediatrics, 12(4), 236-242.

Gupta, P., Mellors, J., Kingsley, L., Riddler, S., Singh, M.K., Schreiber, S., Cronin, M., & Rinaldo, C.R. (1997). High viral load in semen of human immunideficiency virus type 1-infected men at all stages of disease and its reduction by therapy with protease and nonnucleoside reverse transcriptase inhibitors. Virology, 71, 6271-6275.

Hall, D. B., Montaner, J., Reiss, P., Cooper, D., Vella, S., Dohnanyi, C., Myers, M., Lange, J., & Conway, B. (1998). Induction-maintenance antiretroviral therapy: proof of concept. AIDS, 12(7), F41-F44.

Hanson, C. L., Henggeler, S.W., & Burghen, G.A. (1987a). Model of associations between psychosocial variables and health-outcome measures of adolescents with IDDM. Diabetes Care, 10(6), 752-758.

Hanson, C. L., Henggeler, S.W., & Burghen, G.A. (1987b). Social competence and parental support as mediators of the link between stress and metabolic control in adolescents with insulin-dependent diabetes mellitus. Journal of Consulting and Clinical Psychology, 55(4), 529-533.

Hanson, C. L., Henggeler, S.W., Harris, M.A., Burghen, G.A., & Moore, M. (1989). Family system variables and the health status of adolescents with insulindependent diabetes mellitus. Health Psychology, 8(2), 239-253.

Hanson, C. L., De Guire, M.J., Schinkel, A.M., & Henggeler, S.W. (1992).
Comparing social learning and family systems correlates of adaptation in youths with IDDM. Journal of Pediatric Psychology, 17(5), 555-572.

Hanson, C. L., DeGuire, M.J., Schnikel, A.M., Kolterman, O.G., Goodman, J.P., & Buckingham, B.A. (1996). Self-care behaviors in insulin-dependent diabetes: Evaluative tools and their associations with glycemic control. Journal of Pediatric Psychology, 21(4), 467-482.

levers, C.E., Brown, R.T., Drotar, D., Caplan, D., Pishevar, B.S., & Lambert, R.G. (1999). Knowledge of physician prescriptions and adherence to treatment among children with cystic fibrosis and their mothers. Journal of Developmental & Behavioral Pediatrics. 20(5), 335-343.

Jacobson, A. M., Hauser, S.T., Lavori, P., Wolfsdorf, J. I., Herskowitz, R.D.,
Milley, J.F., Gelfand, E., Wertlieb, D., & Stein, J. (1990). Adherence among children and adolescents with insulin-dependent diabetes mellitus over a four-year longitudinal follow-up: I. The influence of patient coping and adjustment. Journal of Pediatric Psychology, 15(4), 511-526.




90



Janz, N. K. B., M.H. (1984). The health belief model: A decade later. Health Education Quarterly, 11, 1-47.

Johnson, S.B. (1995). Managing insulin-dependent diabetes mellitus in
adolescence: A developmental perspective. In L.J. Wallander & J. L. Siegel (Eds.), Adolescent Health Problems: Behavioral perspectives. (pp. 265-288). New York, NY: Guilford Press.

Johnson, S.B., Kelly, M., Henreta, J.C., Cunningham, W.R. & Tomer, A., Silverstein, J.H. (1992). A longitudinal analysis of adherence and health status in childhood diabetes. Journal of Pediatric Psychology, 17(5), 537-553.

Johnson, S.B., Pollak, R.T., Silverstein, J.H., Rosenbloom, A.L., Spillar, R., McCallum, M., & Harkavy, J. (1982). Cognitive and behavioral knowledge about insulin-dependent diabetes among children and parents. Pediatrics, 69(6), 708-713.

Johnson, S.B., Silverstein, J., Rosenbloom, A., Carter, R. & Cunningham, W. (1986). Assessing daily management in childhood diabetes. Health Psychology, 5(6), 545-564.

Kaplan, R.M., & Simon, H.J. (1990). Compliance in medical care:
Reconstruction of self-predictions. Annals of Behavioral Medicine, 12, 66-71.

Kastrissios, H., Suarez, J.R., Katzenstein, D., Girard, P., Sheiner, L.B., &
Blaschke, T.F. (1998). Characterizing patterns of drug-taking behavior with multiple drug regimen in an AIDS clinical trial. AIDS, 12(17), 2295-2303.

Katko, E., Johnson, G.M., Fowler, S.L., & Turner, R.B. (2001). Assessment of adherence with medications in human immunodeficiency virus-infected children. The Pediatric Infectious Disease Journal, 20(12), 174-176.

Kelly, J.A., Otto-Salaj, L.L., Sikkema, K.J., Pinkerton, S.D., & Bloom, F.R.
(1998). Implications of HIV treatment advances for behavioral research on AIDS: protease inhibitors and new challenges in HIV secondary prevention. Health Psychology, 17(4), 310-319.

Klinnert, M.D., McQuaid, E.L., & Gavin, L.A. (1997). Assessing the family asthma management system. Journal of Asthma, 34, 77-88.

Knobel, H., Alonso, J., Casado, J.L., Collazos, J., Gongalez, J., Ruiz, I., Kindelan, J.M., Carmona, A., Juega, J., Ocampo, A, & the GEEMA Study Group (2002). Validation of a simplified medication adherence questionnaire in a large cohort of HIVinfected patients: the GEEMA study. AIDS, 16, 605-613.




91


La Greca, A.M., & Schuman, W.B. (1995). Adherence to prescribed medical regimens. Handbook of Pediatric Psychology (2nd ed., pp. 55-140). New York, NY: Guilford Press.

Laine, C., Newschaffer, C.J., Zhang, D., Cosler, L., Hauck, W.W., & Turner, B.J. (2000). Adherence to antiretroviral therapy by pregnant women infected with human immunodeficiency virus: A pharmacy claims-based analysis. Obstetrics-Gynecology, 95(2), 167-173.

Leventhal, H., Safer, M.A., & Panagis, D.M. (1983). The impact of
communications on the self-regulation of health beliefs, decisions, and behavior. Health Education Quarterly, 10(1), 3-29.

Mackner, L.M., McGrath, A.M., & Stark, L.J. (2001). Dietary recommendations to prevent and manage chronic pediatric health conditions: Adherence, intervention, and future directions. Journal of Developmental & Behavioral Pediatrics, 22(2), 130-143.

Mannheimer, S., Friedland, G., Matts, J., Child, C., & Chesney, M. (2002). The consistency of adherence to antiretroviral therapy predicts biologic outcomes for human immunodeficiency virus-infected persons in clinical trials. Clinical Infectious Diseases, 34, 1115-1121.

Martino, M. de, Tovo, P.A., Balducci, M., Galli, L., Gabiano, C., Rezza, G., & Pezzotti P. (2001). Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. Italian Register for HIV Infection in Children and the Italian National AIDS Registry. Journal of the American Medical Association, 284(2), 190-197.

McNabb, J., Ross, J.W., Abriola, K., Turley, C, Nightingale, C.H., Nicolau, D.P. (2001). Adherence to highly active antiretroviral therapy predicts virologic outcome at an inner-city human immunodeficiency virus clinic. Clinical Infectious Diseases, 33, 700-705.

Melvin, A. J. (1999). Anti-retroviral therapy for HIV-infected children-toward maximal effectiveness. Pediatric Infectious Disease Journal, 18(8), 723-724.

Meyers, K.E., Thomson, P.D., & Weiland, H. (1996). Noncompliance in children and adolescents after renal transplantation. Transplantation, 62, 186-189.

Monane, M., Gurwitz, J. H., Monane, S., & Avorn, J. (1993). Compliance issues in medical practice. Hospital Physician, 35-39.

Monane, M., Bohn, R.L., Gurwitz, J.H., Glynn, R.J., & Avron, J. (1994).
Noncompliance with congestive heart failure therapy in the elderly. Archives of Interval Medicine, 154, 433-437.





92


Murphy, B.C. & Dillon, C. (1998). Interviewing in Action: Process and Practice. Pacific Grove, CA: Brooks/ Cole.

Murphy, D.A., Roberts, K.J., Martin, D.J., Marelich, W., & Hoffman, D. (2000). Barriers to antiretroviral adherence among HIV-infected adults. AIDS Patient Care and STDs, 14(1) 47-58.

Newacheck, P.W., McManus, M., Fox, H.B. Hung, Y.Y., & Halfon, N. (2000).
Access to health care for children with special health care needs. Pediatrics, 105(4), 760766.

O'Reilly, K. O. (1995). The role of qualitative research in the global programme on AIDS at the World Health Organization. In E. Y. Lambert, R.S. Ashery, & R.H. Needle (Eds.), Qualitative Methods of Research in Drug Abuse and HIV Research (Vol. 157, pp. 27-38). Rockville, MD: National Institute of Drug Abuse, Division of Epidemiology and Prevention Research.

Ostrop, N.J., & Gill, M.J. (2000). Antiretroviral medication adherence and
persistence with respect to adherence tool usage. AIDS Patient Care and STDs, 14(7), 351-358.

Paes, A. H., Bakker, A., & Soe-Agnie, C. J. (1998). Measurement of patient compliance. Pharmacology World Science, 20, 7-77.

Palella, F.J., Delaney, K.M., Moorman, A.C., Loveless, M.O., Fuhrer, J., Satten, G.A., Aschman, D.J., Holmberg, S.D., & The HIV Outpatient Study Investigators (1998). Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. New England Journal of Medicine, 338, 853-860.

Parienti, J.J., Verndon, R., & Bazin, C. (2001). The Pills Identification Test: A tool to assess adherence to antiretroviral therapy. Journal of the American Medical Association, 285(4), 412.

Passero, M.A., Remor, B., & Salomon, J. (1981). Patient-reported compliance with cystic fibrosis therapy. Clinical Pediatrics, 20, 264-268.

Paterson, D.L., Swindells, S., Mohr, J., Brester, M., Vergis, E., Squier, C., Wagener, M., & Singh, N. (2000). Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Annals of Internal Medicine,133 21-30.

Quittner, A.L., Drotar, D.A., levers-Landis, C. A., & Seidner, D. (2000a).
Behavioral family systems therapy (BFST) for teenagers with cystic fibrosis and their parents: Treatment and implementation manual. Unpublished Manual.




Full Text
39
3. All adherence measures (24RI frequency, 24RI interval, 24RI dietary, refill
ratings and nurse ratings) will correlate significantly and positively with one
another.
4. Adherence frequency measures (24RI frequency, refill ratings) will
significantly and negatively relate to disease severity (CDC severity ratings,
and viral load).
5. Adherence will significantly and positively relate to caregiver knowledge.
Specifically, caregiver medication knowledge and frequency knowledge will
relate significantly to 24RI frequency adherence and refill adherence,
caregiver knowledge of interval dosing will significantly relate to 24RI
interval adherence, and caregiver knowledge of dietary requirements will
significantly relate to 24RI dietary adherence.
6. 24RI adherence to frequency and refill adherence rates will vary between
medications, based on formulation and frequency of dosing. Due to the
complexities of thrice daily dosing, medications prescribed for thrice daily
dosing will have lower adherence, based on 24RI frequency and refill ratings,
than medications prescribed for once and twice daily dosing. Also, due to the
poor taste associated with many of the liquid medications, those medications
in pill form will have higher adherence, based on 24RI frequency and refill
ratings, than those in liquid form.
7. Caregivers will report more barriers during the 24RI than during the AI-HIV,
because the daily diary format of the 24RI will help to elicit memories of
recent barriers.


84
consequences of poor adherence and subsequent drug resistance. Clinicians and
researchers must ask themselves and adolescents: Given an adolescents personal
attributes, support system, and home environment, what can we expect in terms of ART
adherence?


EXAMINING THE EXTENT OF ADHERENCE AND THE BARRIERS TO
ADHERENCE AMONG HIV-INFECTED CHILDREN ON ANTIRETROVIRAL
THERAPY: A COMBINED QUALITATIVE AND QUANTITATIVE APPROACH
BY
STEPHANIE LYNN MARHEFKA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2002


113
To find out if the informant observed the child taking the medication or eating, say: Were
you with him/her when he/she took his or did you pretty much let him/her
take care of it him/herself?
If informant says that he or she was with the child, say: So did you see him/her take
his/her ? Once you have gone through the childs entire previous day, ask
the parent: Is there is anything else that you can remember about your child's day
yesterday? Is there anything else that you think would be good for me to know?
If the informant has not mentioned that the child took any medication yesterday, say:
Sometimes families have days when the children dont get any of their prescribed
medication, either because they forgot, or they ran out of medication, or for many
other reasons. Was yesterday one of those days for your family? If you ask this
question, designate that on the form by writing AQ in the top right-hand comer. Then,
if they tell you that the child did, in fact, get his/her medication yesterday, proceed with
the medication inquiry. If they say that yes, yesterday was one of those days, just listen to
whatever they have to say, and then proceed with the next question.
Then say: Was this a typical day for your child?
Then say: Many different things can make it hard for children to get the medicines
their doctors recommend. What kinds of things made it hard for your child to get
his/her medications yesterday like the doctor recommended? Record the response (s),
and ask the respondent:
What other things yesterday kept your child from getting his/her medication exactly
like the doctor suggested? Record the response, and say:
Of all the things you just listed, I would like you to tell me which one made it the
hardest for your child to get his/her medication yesterday. (Remind them of the items
on the list if necessary) Record the response and say:
Great! Now, from the other ones you said, tell me which made it the hardest for your
child to get his/her medication yesterday. (Remind them of the items on the list if
necessary) Record the response and proceed with the ranking until all the barriers are
listed in order.
When you are done with the interview, say: Thank you for sharing this information
with us. This information is very important and helps us to understand what it is like
for your family to manage your childs illness. (If appropriate) say: We will be calling
you again soon to learn more about what things are like for your family. Ill speak
with you then. Thank you. Goodbye.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Fonda D. Eyler V (J
Professor of Psychology
This dissertation was submitted to the Graduate Faculty of the College of Health
Professions and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
December 2002
Dean, College of Health Professions


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT vii
INTRODUCTION 1
REVIEW OF THE LITERATURE 4
Prevalence of Poor Adherence among Children with Chronic Illness 4
Defining Adherence 5
The Implications of Poor Adherence 6
Assessment of Medication Adherence 7
Regimen Knowledge 8
Pharmacy Refill 12
Self-Report 13
Assessment of Barriers 16
The Human Immunodeficiency Virus: A New Chronic Illness 18
Implications of Poor Adherence to ART 21
Assessment of Adherence Among Children with HIV 25
Regimen knowledge 25
Pharmacy refill and self-report 28
Barriers to ART Adherence 30
OVERVIEW AND HYPOTHESES 37
METHOD 40
Participants 40
Procedure 41
Measures 44
Adherence Interview-HIV 44
Demographic Questionnaire 45
Telephone Contact Form 46
Medical Record Report Form 46
24-Hour Recall Interview 46
Pharmacy Refill History 48
DATA ANALYSES 49
v


75
regimen is useful in assessing knowledge, the resulting information about daily adherence
is limited and did not correlate with other measures of adherence in this study (24RI and
pharmacy refill history). Possible improvements to the measure may include the addition
of children as informants, a statement emphasizing the importance of honest responding,
and the adaptation of the measure into a computerized structured interview, a format
which has been found to provide more accurate responding than provider interview
(Bangsberg et al., 2002). Additionally, the measure should be altered to inquire directly
about knowledge of prescribed dosing intervals, rather than using a proxy measure of
interval knowledge. Once these modifications are made, the AI-HIV will be a valuable
brief assessment measure with which clinicians and researchers can assess both child and
caregiver knowledge of the prescribed ART regimen.
Results suggest that the 24RI may grossly overestimate adherence to frequency
when compared with pharmacy refill history, though the 24RI is unique in its ability to
assess adherence to prescribed dietary requirements and to assess adherence barriers
experienced over a short period of time. The present study suggests that the 24RI
provides useful information about daily dietary adherence behaviors above and beyond
an assessment of knowledge in the clinical setting. The 24RI also appears to be an
effective tool for assessing barriers to adherence, as it appears to help conjure memories
of barriers encountered in a given day. The apparent 24RI overestimation of adherence
in this study suggests that some modifications may be necessary in order to increase the
measures clinical utility. In this study, 24RI interviewers asked caregivers to describe
their childs previous day, and when medication-taking was mentioned by the caregivers,
interviewers recorded information about the medication names, dosing, whether or not


Place a check in the box next to all medications prescribed for this patient. Then, write in dosing information and check one box for
frequency, and check all applicable boxes under special instructions.
Medication name
Other names
Formulation
Dosing
(in tabs, mg or
cc- indicate)
Frequency
Special Instructions
Abacavir
succinate
Ziagen, 1592U89
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Qq4 Qq6 Oq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Amprenavir
Agenerase, 141W94
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Combivir
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Delavirdine
Rescriptor
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Didanosine
ddl, Videx,
Dideoxyinosine
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Efavirenz
Sustiva, DMP-266
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 q 12
take with food
take on empty stomach
no fatty foods
other:


13
Pharmacy refill histories are a practical and effective means to assess adherence.
First, they necessitate minimal patient burden. Patients are asked to sign a release form
authorizing their pharmacy (or pharmacies) to release refill history information, and then
the researcher obtains the information from the pharmacies, without requiring additional
patient effort. Second, the validity of this methodology has been supported by a strong
correlation between refill history and adherence data based on the Medication Events
Monitoring System (MEMS; Farley et al., 2002). Although refill histories provide only a
gross estimation of adherence, which may be subject to error (Paes, Bakker, & Soe-
Agnue, 1998) they are useful in identifying patients who fail to refill their prescriptions,
despite telling their providers that they continue to take their medication as prescribed.
Self-Report
Another indirect method is self-report assessment. Self-report assessments vary
in their sophistication, from single questions to extensive questionnaires and interviews.
Similarly, the reliability and validity of such measures vary. Generally, physicians tend
to overestimate regimen adherence (Finney, Hook, Friman, Rapoff, & Christophersen,
1993; Rand & Weeks, 1998) and parent and child reports are highly variable in their
degree of accuracy, with social desirability, memory, and involvement in the regimen
affecting the integrity of the data (Rand & Weeks, 1998). Self-report measures that
inquire about brief, specific time periods tend to be most accurate (Kaplan & Simon,
1990; Klinnert, McQuaid, & Gavin, 1997). When the assessment is conducted
accordingly, the self-report measure may be valuable in assessing degrees of adherence.
Although self-reports are limited by a tendency to overestimate adherence behaviors,
when self-reports do indicate poor adherence, the information is considered reliable.


88
Farley, J.J., Hines, S., Musk, A., Ferrus, S., & Tepper, V. (2002). Assessment of
adherence to antiviral therapy in HIV-infected children using the Medication Events
Monitoring System (MEMS), pharmacy refill, provider assessment, caregiver self-report.
and appointment keeping. Manuscript submitted for publication.
Fatkenhauer, G., Theisen, A., & Rockstroh, J. (1997). Virological treatment
failure of protease inhibitor therapy in an unselected cohort of HIV-infected patients.
AIDS. 11.F11-F116.
Feinstein, A.R. (1974). Clinical biostatistics. XXX. Biostatistical problems in
'compliance bias'. Clinical Pharmacology and Therapeutics, 16(5), 846-847.
Festa, R.S., Tamaroff, M.H., Chasalow, F., & Lanzkowsky, P. (1992).
Therapeutic adherence to oral medication regimens by adolescents with cancer: I.
Laboratory assessment. Journal of Pediatrics, 120, 807-811.
Finney, J.W., Hook, R.J., Friman, P.C., Rapoff, M.A., & Christophersen, E.R.
(1993). The overestimation of adherence to pediatric medical regimens. Children's
Health Care, 22(4). 297-304.
Freund, A., Johnson, S.B., Silverstein, J., & Thomas, J. (1991). Assessing daily
management of childhood diabetes using 24-hour recall interviews: Reliability and
stability. Health Psychology, 10(31. 200-208.
Frick, P.A., Gal, P., Lane, T.W., & Sewell, P.C. (1998). Antiretroviral
medication compliance in patients with AIDS. AIDS Patient Care and STD's. 12(61. 463-
470.
Friedland, G.H., & Williams, A. (1999). Attaining higher goals in HIV treatment:
The central importance of adherence AIDS, 13 (Supplement 1), S61-S72.
Gallant, J.E. (2000). Strategies for long-term success in the treatment of HIV
infection. Journal of the American Medical Association, 28311OL 1329-1334.
Gerbino, P. (1993). Forward. Annals of Pharmacotherapy, 27. S3-S4.
Glasgow, R.E., McCaul, K.D., & Schafer, L.C. (1986). Barriers to regimen
adherence among persons with insulin-dependent diabetes. Journal of Behavioral
Medicine. 9(1). 65-77.
Goldffied, M. R. & D' Zurilla, T.J. (1969). A behavior-analytic model for
assessing competence. Current topics in Clinical Child Psychology (Vol. 1, pp. 151-195).
New York, NY: Academic.


28
requirements pertinent to at least one prescribed medication. The relationship between
regimen knowledge and adherence behavior was not reported.
In summary, only one known study has examined regimen knowledge among
HIV-infected children, and that study supports the use of knowledge assessment as an
indicator of possible poor adherence. Five known studies have assessed regimen
knowledge among HIV-infected adults on ART; these studies varied in their
compositions of the sample and sample size, yet all of these studies have demonstrated
that some inaccurate regimen knowledge exists among adults on ART. Two of the adult
studies showed that at least some poor adherers have inaccurate regimen knowledge,
providing additional support for the importance of regimen knowledge assessment.
Pharmacy refill and self-report
Several studies have examined medication adherence among HIV-infected
children by examining pharmacy refill histories. Watson and Farley (1999) examined
adherence with pharmacy refills among 72 children ages 3 months to 12 years. Children
were considered adherent if > 75% of ART medications were refilled during the first 180
days of PI therapy. Fifty-eight percent of children were considered adherent by this
liberal criterion. This suggests that at least 42% families with children on ART did not
claim most of their prescriptions from the pharmacy and, therefore, could not have been
consistently adherent to their regimens.
Katko, Johnson, Fowler, and Turner (2001) used pharmacy refill data to examine
adherence among 34 children on ART. This study considered children adherent if over a
one-year period the proportion of days for which medication was dispensed to the days
for which the medication was prescribed was less than or equal to 90%. Only 34% were


42
Table 1. Participant Demographic Information
Variable
Children (n=63)
Careeivers (ri=63)
Gender
Female
42.9%
90.5%
Male
57.1%
9.5%
Ethnicity
African-American
79.4%
66.7%
Caucasian
12.7%
25.4%
Hispanic
6.3%
4.8%
Other
1.6%
3.2%
HIV Status
Infected
100%
30.2%
Know HIV Status
50.8%
30.2%
Caregiver Relationship
Biological Parent
34.9%
Biological Grandparent

23.8%
Adoptive Parent

28.6%
Foster Parent

1.6%
Relative

9.5%
Family Friend

1.6%
Caregiver Marital Status
Single or Living with Partner
33.3%
Married

39.7%
Separated or Divorced

19.0%
Widowed
7.9%
Caregiver Education
Grade School
7.9%
Some High School

25.4%
Graduated High School or Earned GED

35%
Technical/Vocational Training

1.6%
Some College

23.8%
Bachelor Degree or Higher

6.4%
Caregiver Employment Status
Employed Full-time
36.5%
Employed Part-time

7.9%
Employed Student

1.6%
Unemployed

11.1%
Disabled

15.9%
Retired

9.5%
Homemaker

17.5%


12
Altogether, numerous studies have assessed regimen knowledge among children
with chronic illnesses, though few studies have conducted comprehensive regimen
knowledge assessments that are applicable to medication regimens. The studies that have
used medication-relevant knowledge assessments focus on specific information, such as
the names of the medications, frequencies, and dosing times. Such studies have often
failed to report descriptive information about their findings and reliability coefficients,
but do generally support the relationship between regimen-specific knowledge and
adherence behavior.
Pharmacy Refill
Pharmacy refill history is another indirect method of assessing adherence that has
gained wide acceptance, particularly in HIV research (Farley, Hines, Musk, Ferrus, &
Tepper, 2002; Katko, Johnson, Fowler, & Turner, 2001; Laine et al., 2000; Monane,
Gurwitz, Monane, & Avom, 1993; Ostrop & Gill, 2000; Singh et al., 1996; Singh et al.,
1999; Watson & Farley, 1999). With this method, pharmacy refill information is
obtained for a specific time period, and then adherence rates are calculated. Methods of
calculation may vary, as several studies have failed to report refill adherence calculation
methods (Farley et al., 2002; Ostrop & Gill, 2000; Singh et al., 1996). Most commonly,
the number of days for which the medication was prescribed during the interval is divided
by the number of days for which the medication was dispensed (Katko, Johnson, Fowler,
& Turner, 2001; Monane et al., 1994). Time periods over which refill adherence was
assessed have varied from one month to one year (Farley et al., 2002; Katko, Johnson,
Fowler, & Turner, 2001; Laine et al., 2000; Monane, Gurwitz, Monane, & Avom, 1993;
Singh et al., 1996; Singh et al., 1999; Watson & Farley, 1999; Farley et al., 2002).


5
sample sizes were often small, and the assessments of adherence varied in their ability to
accurately estimate adherence.
Defining Adherence
The words adherence and compliance are often used to label behaviors
consistent with recommended medical regimens. The term compliance has become
less popular in recent years because it is perceived as indicating an authoritarian approach
to disease management (DiMatteo & DiNicola, 1982). Contemporary opinion suggests
that the term adherence is more appropriate, as it places greater emphasis on the
patients active involvement in the treatment regimen (Cassell, 1991; Leventhal, Safer, &
Panagis, 1983). For this reason, the term adherence is used throughout this manuscript
to refer to the extent to which a persons health behavior is consistent with the medical
regimen, as agreed upon by both the patient (or parent) and health care provider.
The definition of adherence is somewhat ambiguous, and is dependent upon
which aspects of the regimen are of interest. Medical regimens often include multiple
components (e.g., medications, diet, and exercise) and each component may have
multiple metrics that are important (e.g., frequency, amount, and duration). These
multiple dimensions of adherence is important to consider, as adherence may vary from
one regimen behavior to another (Johnson 1995; Reid & Appleton, 1991) or from one
metric to another (Johnson, 1995). For example, when researchers asked children and
caregivers to tell them about the childrens previous 24 hours, children with diabetes
were reportedly mostly adherent to insulin injection, but less adherent to the prescribed
diet (Johnson, 1995). Similarly, among adults with HIV, some patients have
demonstrated 100% adherence with regards to some medications but poor adherence with


89
Gudas, L. J., Koocher, G.P., & Wypij, D. (1991). Perceptions of medical
compliance in children and adolescents with cystic fibrosis. Journal of Developmental
and Behavioral Pediatrics, 12(4), 236-242.
Gupta, P., Mellors, J., Kingsley, L., Riddler, S., Singh, M.K., Schreiber, S.,
Cronin, M., & Rinaldo, C.R. (1997). High viral load in semen of human
immunideficiency virus type 1-infected men at all stages of disease and its reduction by
therapy with protease and nonnucleoside reverse transcriptase inhibitors. Virology, 71,
6271-6275.
Hall, D. B., Montaner, J., Reiss, P., Cooper, D., Vella, S., Dohnanyi, C., Myers,
M., Lange, J., & Conway, B. (1998). Induction-maintenance antiretroviral therapy: proof
of concept. AIDS, 12(7), F41-F44.
Hanson, C. L., Henggeler, S.W., & Burghen, G.A. (1987a). Model of associations
between psychosocial variables and health-outcome measures of adolescents with IDDM.
Diabetes Care, 10(6), 752-758.
Hanson, C. L., Henggeler, S.W., & Burghen, G.A. (1987b). Social competence
and parental support as mediators of the link between stress and metabolic control in
adolescents with insulin-dependent diabetes mellitus. Journal of Consulting and Clinical
Psychology, 55(4), 529-533.
Hanson, C. L., Henggeler, S.W., Harris, M.A., Burghen, G.A., & Moore, M.
(1989). Family system variables and the health status of adolescents with insulin-
dependent diabetes mellitus. Health Psychology, 8(2), 239-253.
Hanson, C. L., De Guire, M.J., Schinkel, A.M., & Henggeler, S.W. (1992).
Comparing social learning and family systems correlates of adaptation in youths with
IDDM. Journal of Pediatric Psychology, 17(5). 555-572.
Hanson, C. L., DeGuire, M.J., Schnikel, A.M., Kolterman, O.G., Goodman, J.P.,
& Buckingham, B.A. (1996). Self-care behaviors in insulin-dependent diabetes:
Evaluative tools and their associations with glycemic control. Journal of Pediatric
Psychology, 21(4), 467-482.
levers, C.E., Brown, R.T., Drotar, D., Caplan, D., Pishevar, B.S., & Lambert,
R.G. (1999). Knowledge of physician prescriptions and adherence to treatment among
children with cystic fibrosis and their mothers. Journal of Developmental & Behavioral
Pediatrics. 20(5). 335-343. '
Jacobson, A. M., Hauser, S.T., Lavori, P., Wolfsdorf, J. I., Herskowitz, R.D.,
Milley, J.F., Gelfand, E., Wertlieb, D., & Stein, J. (1990). Adherence among children and
adolescents with insulin-dependent diabetes mellitus over a four-year longitudinal
follow-up: I. The influence of patient coping and adjustment. Journal of Pediatric
Psychology, 15(41.511-526.


OVERVIEW AND HYPOTHESES
Based on the research literature, a necessary step towards the provision of
interventions to promote adherence in this population would be an examination of the
extent and nature of adherence among families of children on ART regimens. The
research indicates that multiple assessment methods are needed in order to fully capture
the extent and nature of adherence (Quittner et al., 2000b). For this reason, this study
used a more subjective, but specific measure (structured interviews) along with a more
objective, but global measure (pharmacy refill history) of adherence. Since previous
studies had not asked families to identify the extent and nature of the barriers that hinder
adherence to pediatric ART regimens, the full extent and nature of the barriers relevant to
this population were unknown. Therefore, in this study, a qualitative assessment was
conducted to comprehensively identify the barriers to adherence experienced by families
of children on ART.
Based on the review of the literature, the specific aims of the current study were:
Primary Aims:
1) To document caregivers knowledge of medication names, dosage, dosing
frequencies, dosing intervals, and medication-specific food and drink intake
guidelines.
2) To document rates of adherence to prescribed medication-taking frequencies,
intervals, and medication-specific food and drink intake guidelines.
37


62
Table 7. Reported Barriers from AI-HIV and 24RI
Barrier
Total # of Caregivers
Who Reported Barrier
(*)
Mean
Frequency
Mean
Strength
Pills are too big to swallow
8(1)
4.00
2.14
Tastes Bad
8(3)
4.80
2.00
Child avoids, resists, prolongs, or
refuses the medication
7(4)
3.00
1.00
Sleeping through dosing time/
medicine interferes with sleep
schedule
7(6)
2.00
1.00
Child or family forgets
6(4)
2.00
1.50
Too many pills to take
6(2)
4.50
1.50
Needs parental supervision/reminding
5(3)
5.00
1.00
Side effects
3(0)
2.00
1.00
Medication-taking interferes with
daily activities
3(2)


Difficulty with the medicine when sick
2(0)
1.50
1.00
Cant keep medicine down
2(1)


Worry that others will find out about
childs HIV status
2(1)
3.00
1.00
Child doesnt swallow pills
1(0)
5.00
1.00
Child hides the medicine
1(0)
3.00
3.00
I have difficulty traveling to get the
refills
1(0)
4.00
1.00
I have difficulty getting the refill (no
reason stated)
1(0)
1.00
1.00
It is too hard to keep up with the
medicine
1(1)




58
frequency adherence were more likely to be classified as at one time having severe CD4
T- Lymphocyte suppression (low white blood cell count or poor immune functioning),
indicating greater disease severity. Also when the more stringent criterion was used,
refill adherence failed to significantly relate to viral load, CDC letter, or CDC number.
Relationship between Adherence and Knowledge
Several hypotheses were made about the relationships between adherence and
knowledge. First it was hypothesized that 24RI frequency adherence and refill adherence
would significantly and positively relate to caregiver knowledge of medication names
and frequencies. Frequency knowledge had little variance and, therefore, this measure
was not included in analyses. Chi-square analyses were performed. Caregiver
knowledge of medication names was not significantly related to 24RI frequency based on
thresholds of 80% or 90%, but was significantly and positively related to refill history
when a 90% adherence cut-off was used (X2= 8.369, p< .01). Results showed that
caregivers with > 90% refill adherence were more likely to know the names of their
childs medication. Next it was hypothesized that knowledge of dosing interval would
significantly correlate with 24RI dosing interval. This hypothesis was supported (r= .42,
2< .001), such that as knowledge of dosing interval deviated from the prescribed dosing
interval, so did 24RI dosing interval. Finally, it was also hypothesized that 24RI dietary
adherence would positively relate to knowledge of dietary requirements; however, the
results of Chi-square analyses were non-significant.
Relationship between Adherence and Prescribed Dosing Frequency
Due to the non-parametric nature of 24RI frequency, Kruskal-Wallis analysis was
employed to determine the relationship between 24RI frequency adherence and


Medication
name
Other names
Formulation
Dosing
Frequency
Special Instructions
Indinavir
Crixivan, MK-639, L-
735,524
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Qq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Lamivudine
3TC, Epivir
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Qq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Nelfinavir
Viracept, AG-1343
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Qq4 Oq6 Qq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Nevirapine
NVP, Viramune,
BIRG-587
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Qq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Ritonavir
Norvir, ABT-538
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Q q 12
take with food
take on empty stomach
no fatty foods
other:
Saquinavir
Invirase, Fortovase
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Stavudine
d4T, Zerit
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Qq8 Oql2
take with food
take on empty stomach
no fatty foods
other:



PAGE 1

(;$0,1,1* 7+( (;7(17 2) $'+(5(1&( $1' 7+( %$55,(56 72 $'+(5(1&( $021* +,9,1)(&7(' &+,/'5(1 21 $17,5(7529,5$/ 7+(5$3< $ &20%,1(' 48$/,7$7,9( $1' 48$17,7$7,9( $3352$&+ %< 67(3+$1,( /<11 0$5+().$ $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

$&.12:/('*(0(176 DP JUDWHIXO IRU VR PDQ\ SHRSOH ZKR FRQWULEXWHG WR WKH VXFFHVVIXO FRPSOHWLRQ RI P\ GRFWRUDO GLVVHUWDWLRQ ZRXOG OLNH WR WKDQN P\ GRFWRUDO FRPPLWWHH PHPEHUV IRU WKHLU JXLGDQFH WKURXJKRXW WKLV YHU\ GLIILFXOW DQG RYHUZKHOPLQJ SURFHVV DP SDUWLFXODUO\ JUDWHIXO IRU WKH KHOS RI P\ FKDLU 'U -LP 5RGULJXH ZKR DJUHHG WR FKDLU P\ GLVVHUWDWLRQ HYHQ ZKHQ LW GLG QRW ILW QHDWO\ LQWR KLV VLOR DOVR RZH VSHFLDO WKDQNV WR 'U 6DP 6HDUV ZKR FKDLUHG P\ PDVWHUf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

PAGE 3

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f§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

PAGE 4

WKH VDQFWXDU\ WKDW ZDV WKH *DLQHVYLOOH VZLQJ GDQFLQJ FRPPXQLW\ DQG KRZ P\ IHOORZ GDQFHUV NHSW P\ IHHW PRYLQJ NQRZ WKDW PRVW RI P\ IULHQGV DQG IDPLO\ PHPEHUV FDQQRW EHJLQ WR XQGHUVWDQG WKH SURFHVVHV RI JUDGXDWH VFKRRO DQG GLVVHUWDWLRQ FRPSOHWLRQ DP JUDWHIXO WR WKHP IRU WU\LQJ WR XQGHUVWDQG DQG VXSSRUWLQJ PH HYHQ ZKHQ WKH\ GLG QRW XQGHUVWDQG :LWK VRPH FRQILGHQFH FDQ ILQDOO\ VD\ WKDW DP GRQH ZLWK VFKRRO ,9

PAGE 5

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

PAGE 6

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

PAGE 7

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ (;$0,1,1* 7+( (;7(17 2) 3225 $'+(5(1&( $1' 7+( %$55,(56 72 $'+(5(1&( $021* +,9,1)(&7(' &+,/'5(1 21 $17,5(7529,5$/ 7+(5$3< $ &20%,1(' 48$/,7$7,9( $1' 48$17,7$7,9( $3352$&+ %\ 6WHSKDQLH / 0DUKHIND 'HFHPEHU &KDLUSHUVRQ -DPHV 5 5RGULJXH 3K' 0DMRU 'HSDUWPHQW &OLQLFDO DQG +HDOWK 3V\FKRORJ\ 7KLV VWXG\ DVVHVVHG DGKHUHQFH WR DQWLUHWURYLUDO PHGLFDWLRQV DPRQJ FKLOGUHQ ZLWK +,9 LQIHFWLRQ 7UDLQHG LQWHUYLHZHUV DGPLQLVWHUHG WKH $GKHUHQFH ,QWHUYLHZ+,9 $,+,9f WR FDUHJLYHUV RI LQIHFWHG FKLOGUHQ WR DVVHVV W\SLFDO PHGLFDWLRQ UHJLPHQ EHKDYLRUV DQG NQRZOHGJH RI WKH WDUJHW FKLOGfV SUHVFULEHG UHJLPHQ 1H[W GHPRJUDSKLF LQIRUPDWLRQ ZDV FROOHFWHG YLD LQWHUYLHZ DQG D PHGLFDO UHFRUG UHYLHZ ZDV FRQGXFWHG 7ZR ZHHNV IROORZLQJ WKH LQLWLDO LQWHUYLHZ WUDLQHG LQWHUYLHZV EHJDQ WKH +RXU 5HFDOO ,QWHUYLHZ 5,f SURFHGXUH ZLWK HDFK FDUHJLYHU FDUHJLYHUV ZHUH WKHQ LQWHUYLHZHG RYHU WKH WHOHSKRQH WZR DGGLWLRQDO WLPHV ZLWKLQ WKH QH[W WZR ZHHNV 7KUHH PRQWKV DIWHU WKH LQLWLDO LQWHUYLHZ SKDUPDF\ UHILOO GDWD ZHUH FROOHFWHG 5HVXOWV VKRZ WKDW b RI FDUHJLYHUV IDLOHG WR FRUUHFWO\ LGHQWLI\ DW OHDVW RQH RI WKHLU FKLOGfV PHGLFDWLRQ QDPHV b RI FDUHJLYHUV IDLOHG WR FRUUHFWO\ LGHQWLI\ WKH GRVDJH IRU DW OHDVW RQH PHGLFDWLRQ DQG KDOI RI WKH FDUHJLYHUV IDLOHG WR FRUUHFWO\ LGHQWLI\ WKH YLL

PAGE 8

VSHFLILF GLHWDU\ UHTXLUHPHQWV IRU DW OHDVW RQH PHGLFDWLRQ $GKHUHQFH WR PHGLFDWLRQn WDNLQJ IUHTXHQF\ YDULHG E\ DVVHVVPHQW PRGDOLW\ 5HVXOWV RI WKH 5, VXJJHVW WKDW b RI FKLOGUHQ ZHUH DW OHDVW b DGKHUHQW KRZHYHU UHVXOWV RI WKH SKDUPDF\ UHILOO KLVWRU\ VXJJHVW WKDW RQO\ b RI FKLOGUHQ ZHUH DW OHDVW b DGKHUHQW :KHQ DGKHUHQFH WR WKH SUHVFULEHG LQWHUYDO ZDV H[DPLQHG ZLWK WKH 5, b RI GRVHV JLYHQ ZHUH GHYLDQW IURP WKH SUHVFULEHG LQWHUYDO E\ DW OHDVW RQH KRXU ZKLOH b RI GRVHV JLYHQ ZHUH GHYLDQW E\ DW OHDVW KRXUV 5HVXOWV RI WKH 5, DOVR VXJJHVW WKDW WKH DYHUDJH FKLOG ZDV DGKHUHQW WR PHGLFDWLRQVSHFLILF GLHWDU\ UHTXLUHPHQWV b RI WKH WLPH 7ZHQW\WZR GLIIHUHQW EDUULHUV ZHUH UHSRUWHG DQG IDOO LQWR WKUHH JHQHUDO FDWHJRULHV Df PHGLFDWLRQVSHFLILF DWWULEXWHV VXFK DV WKH VL]H RU WDVWH RI SLOOV Ef SUREOHPV ZLWK VFKHGXOLQJ RU URXWLQH RU Ff SUREOHPV ZLWK WKH FKLOG UHVLVWLQJ UHIXVLQJ RU KLGLQJ WKH PHGLFDWLRQ 7KLV VWXG\ VXSSRUWV WKH DVVHUWLRQ WKDW D VLJQLILFDQW SURSRUWLRQ RI +,9LQIHFWHG FKLOGUHQ DUH QRW UHFHLYLQJ WKHLU PHGLFDWLRQ H[DFWO\ DV LW LV SUHVFULEHG 7KH VWXG\ DOVR VXJJHVWV WKDW D ODUJH SRUWLRQ RI FDUHJLYHUV ODFN NQRZOHGJH DERXW WKHLU FKLOGfV +,9 PHGLFDWLRQ UHJLPHQ ,PSOLFDWLRQV IRU DGKHUHQFHUHODWHG LQWHUYHQWLRQV DUH GLVFXVVHG YLLL

PAGE 9

,1752'8&7,21 (VWLPDWHV VXJJHVW WKDW b RI FKLOGUHQ LQ WKH 8QLWHG 6WDWHV KDYH D FKURQLF LOOQHVV 1HZDFKHFN 0F0DQXV )R[ +XQJ t +DOIRQ f $PRQJ FKLOGUHQ ZLWK FKURQLF LOOQHVV WKH SUHYDOHQFH RI SRRU DGKHUHQFH WR PHGLFDO UHJLPHQV YDULHV DFFRUGLQJ WR WKH VWXG\ VDPSOH WKH VSHFLILFV RI UHJLPHQ UHTXLUHPHQWV WKH DVVHVVPHQW RI DGKHUHQFH DQG WKH FULWHULD XVHG WR FODVVLI\ FKLOGUHQ DV DGKHUHQW RU SRRUO\ DGKHUHQW 5DSRII f 1HYHUWKHOHVV HVWLPDWHV VXJJHVW WKDW b RI FKLOGUHQ ZLWK FKURQLF LOOQHVV PD\ QRW IXOO\ DGKHUH WR WKHLU UHJLPHQV $OHVVDQGUR 9LQFHQ]R 0DUFR 0DUFHOOR t (QULFD &RQOH\ t 6DOYDWLHUUD (WWHQJHU HW DO )HVWD 7DPDURII &KDVDORZ t /DQ]RZVN\ 0H\HUV 7KRPSVRQ t :HLODQG 6FKRQL +RUDN t 1LNRODL]LN :HLVEHUJ%HQFKHOO HW DO f $GYHUVH HIIHFWV RI SRRU DGKHUHQFH LQFOXGH KHDOWK FDUH FRVWV FOLQLFDO GHFLVLRQPDNLQJ FRQFOXVLRQV GUDZQ IURP FOLQLFDO WULDOV PRUELGLW\ DQG PRUWDOLW\ 5DSRII f 7KH +XPDQ ,PPXQRGHILFLHQF\ 9LUXV +,9f LV D FKURQLF LOOQHVV WKDW DIIHFWV FKLOGUHQ DV ZHOO DV DGXOWV 0HGLFDWLRQ UHJLPHQV IRU +,9LQIHFWHG FKLOGUHQ UHTXLUH PXOWLSOH PHGLFDWLRQV WR EH WDNHQ DW VSHFLILF WLPHV WKURXJKRXW WKH GD\ 6FRWW t 6OHDVPDQ f 7KH SLOOV DUH RIWHQ ODUJH DQG GLIILFXOW WR VZDOORZ DQG PDQ\ RI WKH OLTXLG PHGLFDWLRQV KDYH D EDG WDVWH )RU WKHVH DQG RWKHU UHDVRQV IDPLOLHV PD\ KDYH GLIILFXOW\ DGKHULQJ WR SHGLDWULF +,9 PHGLFDWLRQ UHJLPHQV 7KLV LV SDUWLFXODUO\ SUREOHPDWLF DV SRRU DGKHUHQFH WR DQWL+,9 PHGLFDWLRQV LV D VHULRXV SXEOLF KHDOWK FRQFHUQ

PAGE 10

:KHQ +,9 DGKHUHQFH LV SRRU WKH SRWHQWLDO FRQVHTXHQFHV DUH VHYHUH SRRU DGKHUHQFH PD\ EH GHWULPHQWDO WR ERWK WKH LQGLYLGXDO DQG VRFLHW\ :KHQ FKLOGUHQ DQG IDPLOLHV DUH QRQDGKHUHQW WR WKHLU DQWL+,9 PHGLFDWLRQ UHJLPHQV +,9LQIHFWHG FKLOGUHQ PD\ H[SHULHQFH UDSLG SK\VLFDO GHFOLQH ZLWK IHZ WKHUDSHXWLF RSWLRQV %XW] -R\QHU *UHHQEHUJ)ULHGPDQ t +XWWRQ )DWNHQKDXHU 7KHLVHQ t 5RFNVWURFK f ,Q WHUPV RI WKH LPSDFW RQ VRFLHW\ SRRU DGKHUHQFH PD\ PHDQ LQFUHDVHG UDWHV RI WUDQVPLVVLRQ RI +,9 DQG LQFUHDVHG GLIILFXOW\ WUHDWLQJ WKH YLUXV :DLQEHUJ t )ULHGODQG f 7KLV LV SDUWLFXODUO\ DODUPLQJ VLQFH DGROHVFHQWV DUH DPRQJ WKH ODUJHVW QHZO\LQIHFWHG +,9 SRSXODWLRQ 81$,'6 f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f UHJLPHQ 1H[W LPSOLFDWLRQV RI SRRU DGKHUHQFH WR $57 DQG WKH DVVHVVPHQW RI PHGLFDWLRQ

PAGE 11

DGKHUHQFH DQG EDUULHUV WR DGKHUHQFH DPRQJ FKLOGUHQ RQ $57 DUH UHYLHZHG 7KH OLWHUDWXUH UHYLHZ FRQFOXGHV ZLWK D VXPPDU\ VSHFLILF DLPV DQG K\SRWK&HVHV IRU WKH IROORZLQJ VWXG\ ZKLFK H[DPLQHV WKH H[WHQW DQG QDWXUH RI DGKHUHQFH DQG EDUULHUV WR DGKHUHQFH DPRQJ +,9LQIHFWHG FKLOGUHQ DQG WKHLU IDPLOLHV 7KH PHWKRG UHVXOWV DQG GLVFXVVLRQ RI WKH ILQGLQJV IROORZ

PAGE 12

5(9,(: 2) 7+( /,7(5$785( $FFRUGLQJ WR WKH $PHULFDQ $FDGHP\ RI 3HGLDWULFV $$3f FKLOGKRRG FKURQLF LOOQHVVHV DUH GHILQHG DV FRQGLWLRQV WKDW ODVW DW OHDVW PRQWKV DQG UHTXLUH PHGLFDO DWWHQWLRQ DQG FDUH WKDW LV DERYH DQG EH\RQG WKDW ZKLFK LV QRUPDOO\ H[SHFWHG IRU D FKLOG RU DGROHVFHQW RI WKH VDPH DJH $$3 f (VWLPDWHV VXJJHVW WKDW b RI FKLOGUHQ LQ WKH 8QLWHG 6WDWHV KDYH D FKURQLF LOOQHVV 1HZDFKHFN HW DO f 7KH PRVW FRPPRQ FKLOGKRRG FKURQLF LOOQHVVHV LQFOXGH UKHXPDWRLG DUWKULWLV DVWKPD OHXNHPLD DQG RWKHU PDOLJQDQFLHV VSLQD ELILGD VHL]XUH GLVRUGHUV QHXURPXVFXODU GLVHDVHV DFTXLUHG LPPXQRGHILFLHQF\ V\QGURPH DQG GLDEHWHV $$3 f &RQGLWLRQV YDU\ ZLGHO\ LQ WKHLU RQVHW FRXUVH GXUDWLRQ DQG VHYHULW\ DQG LQ WKH GDLO\ GHPDQGV UHTXLUHG IRU GLVHDVH PDQDJHPHQW 3UHYDOHQFH RI 3RRU $GKHUHQFH DPRQJ &KLOGUHQ ZLWK &KURQLF ,OOQHVV $PRQJ FKLOGUHQ ZLWK FKURQLF LOOQHVVHV WKH SUHYDOHQFH RI SRRU DGKHUHQFH WR PHGLFDO UHJLPHQV YDULHV DFFRUGLQJ WR WKH VWXG\ VDPSOH WKH VSHFLILFV RI UHJLPHQ UHTXLUHPHQWV WKH DVVHVVPHQW RI DGKHUHQFH DQG WKH FULWHULD XVHG WR FODVVLI\ FKLOGUHQ DV DGKHUHQW RU SRRUO\ DGKHUHQW 5DSRII f $FURVV VWXGLHV RI FKLOGUHQ ZLWK FKURQLF LOOQHVVHV QRQDGKHUHQFH KDV EHHQ HVWLPDWHG DW b $OHVVDQGUR 9LQFHQ]R 0DUFR 0DUFHOOR t (QULFD &RQOH\ t 6DOYDWLHUUD (WWHQJHU HW DO )HVWD 7DPDURII &KDVDORZ t /DQ]RZVN\ 0H\HUV 7KRPSVRQ t :HLODQG 6FKRQL HW DO :HLVEHUJ%HQFKHOO HW DO f +RZHYHU WKHVH HVWLPDWHV ZHUH WDNHQ IURP D UHYLHZ RI VWXGLHV RI DGKHUHQFH WR D VPDOO VHW RI FKURQLF LOOQHVV UHJLPHQV LQ ZKLFK

PAGE 13

VDPSOH VL]HV ZHUH RIWHQ VPDOO DQG WKH DVVHVVPHQWV RI DGKHUHQFH YDULHG LQ WKHLU DELOLW\ WR DFFXUDWHO\ HVWLPDWH DGKHUHQFH 'HILQLQJ $GKHUHQFH 7KH ZRUGV fDGKHUHQFHf DQG fFRPSOLDQFHf DUH RIWHQ XVHG WR ODEHO EHKDYLRUV FRQVLVWHQW ZLWK UHFRPPHQGHG PHGLFDO UHJLPHQV 7KH WHUP fFRPSOLDQFHf KDV EHFRPH OHVV SRSXODU LQ UHFHQW \HDUV EHFDXVH LW LV SHUFHLYHG DV LQGLFDWLQJ DQ DXWKRULWDULDQ DSSURDFK WR GLVHDVH PDQDJHPHQW 'L0DWWHR t 'L1LFROD f &RQWHPSRUDU\ RSLQLRQ VXJJHVWV WKDW WKH WHUP fDGKHUHQFHf LV PRUH DSSURSULDWH DV LW SODFHV JUHDWHU HPSKDVLV RQ WKH SDWLHQWfV DFWLYH LQYROYHPHQW LQ WKH WUHDWPHQW UHJLPHQ &DVVHOO /HYHQWKDO 6DIHU t 3DQDJLV f )RU WKLV UHDVRQ WKH WHUP fDGKHUHQFHf LV XVHG WKURXJKRXW WKLV PDQXVFULSW WR UHIHU WR WKH H[WHQW WR ZKLFK D SHUVRQfV KHDOWK EHKDYLRU LV FRQVLVWHQW ZLWK WKH PHGLFDO UHJLPHQ DV DJUHHG XSRQ E\ ERWK WKH SDWLHQW RU SDUHQWf DQG KHDOWK FDUH SURYLGHU 7KH GHILQLWLRQ RI DGKHUHQFH LV VRPHZKDW DPELJXRXV DQG LV GHSHQGHQW XSRQ ZKLFK DVSHFWV RI WKH UHJLPHQ DUH RI LQWHUHVW 0HGLFDO UHJLPHQV RIWHQ LQFOXGH PXOWLSOH FRPSRQHQWV HJ PHGLFDWLRQV GLHW DQG H[HUFLVHf DQG HDFK FRPSRQHQW PD\ KDYH PXOWLSOH PHWULFV WKDW DUH LPSRUWDQW HJ IUHTXHQF\ DPRXQW DQG GXUDWLRQf 7KHVH PXOWLSOH GLPHQVLRQV RI DGKHUHQFH LV LPSRUWDQW WR FRQVLGHU DV DGKHUHQFH PD\ YDU\ IURP RQH UHJLPHQ EHKDYLRU WR DQRWKHU -RKQVRQ 5HLG t $SSOHWRQ f RU IURP RQH PHWULF WR DQRWKHU -RKQVRQ f )RU H[DPSOH ZKHQ UHVHDUFKHUV DVNHG FKLOGUHQ DQG FDUHJLYHUV WR WHOO WKHP DERXW WKH FKLOGUHQfV SUHYLRXV KRXUV FKLOGUHQ ZLWK GLDEHWHV ZHUH UHSRUWHGO\ PRVWO\ DGKHUHQW WR LQVXOLQ LQMHFWLRQ EXW OHVV DGKHUHQW WR WKH SUHVFULEHG GLHW -RKQVRQ f 6LPLODUO\ DPRQJ DGXOWV ZLWK +,9 VRPH SDWLHQWV KDYH GHPRQVWUDWHG b DGKHUHQFH ZLWK UHJDUGV WR VRPH PHGLFDWLRQV EXW SRRU DGKHUHQFH ZLWK

PAGE 14

UHJDUGV WR RWKHU PHGLFDWLRQV GXULQJ WKH VDPH LQWHUYDO +DOO HW DO f 8QIRUWXQDWHO\ PXFK RI WKH OLWHUDWXUH KDV H[DPLQHG DGKHUHQFH XQLGLPHQVLRQDOO\ IDLOLQJ WR DVVHVV HDFK DVSHFW RI WKH UHJLPHQ DQG HDFK PHWULF LQGLYLGXDOO\ 7KH ,PSOLFDWLRQV RI 3RRU $GKHUHQFH 7KH LPSDFW RI SRRU DGKHUHQFH LQ FKLOGUHQ ZLWK FKURQLF LOOQHVVHV LV SRWHQWLDOO\ VHYHUH DQG ORQJVWDQGLQJ 3RRU DGKHUHQFH DGYHUVHO\ DIIHFWV KHDOWK FDUH FRVWV FOLQLFDO GHFLVLRQPDNLQJ DQG FRQFOXVLRQV GUDZQ IURP FOLQLFDO WULDOV LQ DGGLWLRQ WR KHDOWK RXWFRPHV 5DSRII f :KHQ SDWLHQWV DUH QRQDGKHUHQW WR PHGLFDO UHJLPHQV LQFUHDVHG KHDOWK FDUH FRVWV DUH LQFXUUHG DQG KHDOWK FDUH UHVRXUFHV DUH ZDVWHG ,Q WKH 8QLWHG 6WDWHV WKH KHDOWK FDUH FRVWV RI SRRU DGKHUHQFH LQ ERWK FKLOGUHQ DQG DGXOWV LV HVWLPDWHG DW ELOOLRQ SHU \HDU %HUJ 'LVFKOHU :DJQHU 5DMD t 3DOPHU6KHOYLQ f &RVWV LQFOXGH ZDVWHG PHGLFDWLRQ DQG XQXVHG WKHUDSHXWLF HTXLSPHQW DQG LQFUHDVHG PRUELGLW\ OHDGLQJ WR PRUH IUHTXHQW FOLQLF DSSRLQWPHQWV HPHUJHQF\ FDUH DQG KRVSLWDOL]DWLRQ 3RRU DGKHUHQFH DOVR DGYHUVHO\ DIIHFWV FOLQLFDO GHFLVLRQPDNLQJ 3URYLGHUV PD\ PDNH FOLQLFDO GHFLVLRQV EDVHG RQ WKH DVVXPSWLRQ WKDW D SDWLHQW LV DGKHUHQW HYHQ ZKHQ WKH SDWLHQW LV QRW 5DSRII f 7KLV PD\ OHDG WR WKH SURYLVLRQ RI LQFUHDVHG WLPH DQG PHGLFDO UHVRXUFHV LQ WKH IRUP RI DGGLWLRQDO PHGLFDWLRQ WKHUDSLHV RU PHGLFDO SURFHGXUHV ,Q FOLQLFDO WULDOV SRRU DGKHUHQFH PD\ OHDG WR DQ XQGHUHVWLPDWLRQ RI WKH HIIHFWLYHQHVV RI PHGLFDWLRQV NQRZQ DV WKH FRPSOLDQFH ELDV )HLQVWHLQf ,I D SHUVRQ LQYROYHG LQ D FOLQLFDO WULDO IDLOV WR WDNH PHGLFDWLRQV DSSURSULDWHO\ EXW GRHV QRW LQIRUP WKH LQYHVWLJDWRU RI KHU SRRU DGKHUHQFH WKH LQYHVWLJDWRU PD\ DVVXPH WKDW WKH PHGLFDWLRQV ZHUH WDNHQ DV SUHVFULEHG ,I VHYHUDO SDUWLFLSDQWV LQ D FOLQLFDO WULDO UHSHDW WKLV EHKDYLRU WKH LQYHVWLJDWRU PD\ HUURQHRXVO\ FRQFOXGH WKDW WKH GUXJ LV LQHIIHFWLYH RU WKDW KLJKHU GRVDJHV

PAGE 15

DUH QHHGHG WR KDYH D EHQHILFLDO HIIHFW 8UTXKDUW f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f K\SHUWHQVLRQ DQG UHQDO GLVHDVH ,Q RWKHU LOOQHVVHV VXFK DV GLDEHWHV DQG DVWKPD SRRU DGKHUHQFH RYHU HYHQ D VKRUW SHULRG RI WLPH PD\ UHVXOW LQ GHDWK %LUNKHDG $WWDZD\ 6WUXQN 7RZQVHQG t 7HXWVFK 'HODPDWHU *HUELQR f :LWK WKH H[FHSWLRQ RI SDOOLDWLYH WUHDWPHQWV SRRU DGKHUHQFH ZLWK PRVW GLVHDVH PDQDJHPHQW EHKDYLRUV UHVXOWV LQ LQFUHDVHG PRUELGLW\ DQG PRUWDOLW\ $VVHVVPHQW RI 0HGLFDWLRQ $GKHUHQFH 7KH DVVHVVPHQW RI DGKHUHQFH LV D FRPSOH[ SUREOHP IRU ZKLFK WKHUH LV QR HDV\ VROXWLRQ $OWKRXJK VHYHUDO DVVHVVPHQW PHWKRGV KDYH EHHQ XVHG WR PHDVXUH DGKHUHQFH QR SHUIHFW PHWKRG KDV EHHQ LGHQWLILHG 5DSRII f ,Q WKH DEVHQFH RI D fJROG VWDQGDUGf IRU DGKHUHQFH DVVHVVPHQW UHVHDUFKHUV KDYH VXJJHVWHG WKDW PXOWLSOH DVVHVVPHQW PRGDOLWLHV VKRXOG EH XVHG LQ RUGHU WR REWDLQ WKH PRVW FRPSUHKHQVLYH LQIRUPDWLRQ $GKHUHQFH PHDVXUHPHQWV KDYH EHHQ FKDUDFWHUL]HG DV fGLUHFWf DQG fLQGLUHFWf DVVHVVPHQWV 5DSRII f 'LUHFW DVVHVVPHQWV VXFK DV ELRORJLFDO DVVD\V GLUHFW REVHUYDWLRQV DQG HOHFWURQLF PRQLWRULQJ DUH RIWHQ GLIILFXOW WR REWDLQ DQG FDQ EH FRVWO\ 5DQG t :HHNV f

PAGE 16

&RQYHUVHO\ PXOWLSOH LQGLUHFW DVVHVVPHQWV KDYH EHHQ IRXQG WR EH HIIHFWLYH UHODWLYHO\ HDV\ WR REWDLQ DQG LQH[SHQVLYH 5HJLPHQ .QRZOHGJH 2QH LPSRUWDQW DVSHFW RI PHDVXULQJ DGKHUHQFH LQGLUHFWO\ LV GHWHUPLQLQJ WKH H[WHQW WR ZKLFK WKH SHUVRQV UHVSRQVLEOH IRU D FKLOGfV PHGLFDWLRQWDNLQJ DUH NQRZOHGJHDEOH DERXW WKH SUHVFULEHG UHJLPHQ ,I WKHVH LQGLYLGXDOV GR QRW NQRZ DQG XQGHUVWDQG WKH VSHFLILFV UHJDUGLQJ WKH PHGLFDWLRQ UHJLPHQ FRPSOHWH DGKHUHQFH LV XQOLNHO\ $PRQJ FKLOGUHQ ZLWK D YDULHW\ RI FKURQLF LOOQHVVHV ODFN RI FKLOG DQG FDUHJLYHU UHJLPHQ NQRZOHGJH KDV EHHQ DVVRFLDWHG ZLWK SRRU DGKHUHQFH *XGDV .RRFKHU t :\SLM +DQVRQ +HQJJHOHU t %XUJKHQ D OHYHUV HW DO 5XELQ %DXPDQ t /DXE\ f +RZHYHU QRW DOO VWXGLHV KDYH IRXQG D UHODWLRQVKLS EHWZHHQ NQRZOHGJH DQG DGKHUHQFH %HFN HW DO f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f ZDV GHYHORSHG WR DVVHVV JHQHUDO NQRZOHGJH RI GLDEHWHV GLDEHWHVUHODWHG SUREOHPVROYLQJ DQG REVHUYHG

PAGE 17

VNLOO LQ XULQH WHVWLQJ DQG VHOILQMHFWLRQ 7KH 'LDEHWHV .QRZOHGJH DQG 0DQDJHPHQW 6NLOOV $VVHVVPHQW 4XHVWLRQQDLUH %URZQOHH'XIIHFN HW DO f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f KDV EHHQ GHYHORSHG IRU PHDVXULQJ LQKDOHUVSDFHU WHFKQLTXH DPRQJ FKLOGUHQ ZLWK SXOPRQDU\ GLVRUGHUV %RFFXWL &HODQR *HOOHU t 3KLOOLSV &HODQR *HOOHU 3KLOOLSV t =LPDQ f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f DVVHVVHG UHJLPHQ NQRZOHGJH DPRQJ FKLOGUHQ ZLWK FDQFHU DQG WKHLU SDUHQWV ,QWHUYLHZ TXHVWLRQV FRQFHUQHG NQRZOHGJH RI WKH PHGLFDWLRQV DQG

PAGE 18

XQGHUVWDQGLQJ RI WKH UHJLPHQ WKRXJK WKH DXWKRUV GLG QRW SURYLGH PRUH VSHFLILF LQIRUPDWLRQ UHJDUGLQJ WKH NQRZOHGJH DVVHVVPHQW $GKHUHQFH EHKDYLRU ZDV PHDVXUHG YLD VHOIUHSRUW 0RVW ZKR UHSRUWHG IUHTXHQW QRQDGKHUHQFH GHPRQVWUDWHG SRRU UHJLPHQ NQRZOHGJH VXSSRUWLQJ WKH UHODWLRQVKLS EHWZHHQ UHJLPHQ NQRZOHGJH DQG DGKHUHQFH $QWKRQ\ 3D[WRQ %LQHV DQG 3KHODQ f DVVHVVHG PDWHUQDO QXWULWLRQDO NQRZOHGJH VSHFLILF WR &) 1R LQIRUPDWLRQ ZDV SURYLGHG DERXW KRZ QXWULWLRQDO NQRZOHGJH ZDV DVVHVVHG WKRXJK GLVHDVHVSHFLILF QXWULWLRQDO NQRZOHGJH GLG SRVLWLYHO\ UHODWH WR KHLJKW ZHLJKW DQG VHOIUHSRUWV RI GLHWDU\ DGKHUHQFH EHKDYLRUV :KLOH WKHVH ILQGLQJV GR VXSSRUW D UHODWLRQVKLS EHWZHHQ NQRZOHGJH DQG DGKHUHQFH EHKDYLRUV WKHVH VWXGLHV IDLO WR SURYLGH LPSRUWDQW PHWKRGRORJLFDO LQIRUPDWLRQ DERXW WKHLU DVVHVVPHQWV %HFN DQG FROOHDJXHV f ZHUH DPRQJ WKH ILUVW WR DVVHVV NQRZOHGJH RI SUHVFULEHG PHGLFDWLRQ UHJLPHQV 7KH\ XVHG D TXHVWLRQQDLUH WR DVVHVV FKLOG DQG FDUHJLYHU NQRZOHGJH RI WKH XVH GRVH VLGH HIIHFWV DQG LPSRUWDQFH RI HDFK PHGLFDWLRQ SUHVFULEHG DIWHU UHQDO WUDQVSODQWDWLRQ $ WRWDO NQRZOHGJH VFRUH ZDV JLYHQ EDVHG RQ WKH SHUFHQWDJH RI TXHVWLRQV DQVZHUHG FRUUHFWO\ $W EDVHOLQH WKH PHDQ NQRZOHGJH VFRUH UDQJHG IURP b WR b ZLWK D PHDQ RI b VXJJHVWLQJ WKDW D ODUJH SRUWLRQ RI WKH VDPSOH ODFNHG VLJQLILFDQW NQRZOHGJH DERXW WKH PHGLFDWLRQ UHJLPHQ .QRZOHGJH GLG QRW VLJQLILFDQWO\ FRUUHODWH ZLWK DGKHUHQFH EHKDYLRU *XGDV .RRFKHU DQG :\SLM f DVVHVVHG NQRZOHGJH RI FKLOGUHQfV &) UHJLPHQV 7KH\ DVNHG FKLOGUHQ DJHV WR VWDWH WKH PHGLFDWLRQV WKDW WKH\ WDNH IRU &) DQG WKH LPSDFW WKDW WKH PHGLFDWLRQV FKHVW SK\VLRWKHUDS\ DQG GLHW KDYH RQ WKH GLVHDVH 8QIRUWXQDWHO\ WKH UHVHDUFKHUV GLG QRW SURYLGH GHVFULSWLYH GDWD DERXW WKH GHJUHH RI

PAGE 19

NQRZOHGJH DPRQJ WKLV JURXS WKRXJK WKH\ GLG UHSRUW WKDW NQRZOHGJH ZDV ZLGHO\ YDULDEOH DQG UHODWHG WR VHOIUHSRUWV RI DGKHUHQFH EHKDYLRU 'L*LURODPR 4XLWWQHU $FNHUPDQ t 6WHYHQV f ZHUH DPRQJ WKH ILUVW WR UHSRUW D FRPSUHKHQVLYH DVVHVVPHQW RI NQRZOHGJH RI WKH VHOIFDUH UHJLPHQ IRU FKLOGUHQ ZLWK &) 7KH\ GHYHORSHG WKH 7UHDWPHQW $GKHUHQFH 4XHVWLRQQDLUH&\VWLF )LEURVLV 7$4&)f ZLWK D FKLOG DQG SDUHQWYHUVLRQ LWHP VHOIUHSRUW PHDVXUH WKDW DVVHVVHV DGKHUHQFH DQG NQRZOHGJH RI SK\VLFLDQfV WUHDWPHQW UHFRPPHQGDWLRQV IRU DLUZD\ FOHDUDQFH DHURVRO WUHDWPHQWV DQG SDQFUHDWLF HQ]\PH XVH 7KH PHDVXUH KDV EHHQ IRXQG WR KDYH JRRG WHVWUHWHVW UHOLDELOLW\ D f OHYHUV DQG RWKHUV f XVHG WKH 7$4&) DQG IRXQG WKDW b RI PRWKHUV LQFRUUHFWO\ LGHQWLILHG WKH SUHVFULEHG IUHTXHQF\ RI DLUZD\ FOHDUDQFH WUHDWPHQWV DQG b LQFRUUHFWO\ LGHQWLILHG WKH SUHVFULEHG IUHTXHQF\ RI DHURVRO PHGLFDWLRQV (OHYHQ DQG D KDOI SHUFHQW RI FDUHJLYHUV LQFRUUHFWO\ LGHQWLILHG WKH JUHDWHVW TXDQWLW\ RI HQ]\PHV WKHLU FKLOGUHQ VKRXOG WDNH ZLWK D PHDO DQG b LQFRUUHFWO\ LGHQWLILHG WKH JUHDWHVW TXDQWLW\ RI HQ]\PHV WR EH WDNHQ ZLWK D VQDFN WKRXJK IHZ RI WKH HQ]\PH LGHQWLILFDWLRQV ZHUH JURVVO\ LQFRUUHFW &KLOGUHQ GHPRQVWUDWHG HYHQ SRRUHU UHJLPHQ NQRZOHGJH %RWK FKLOG DQG FDUHJLYHU NQRZOHGJH ZHUH SUHGLFWLYH RI VHOIUHSRUWV RI DGKHUHQFH 5LFNHU 'HODPDWHU DQG +VX f DOVR UHSRUWHG WKH FRPSUHKHQVLYH DVVHVVPHQW RI NQRZOHGJH RI WKH VHOIFDUH UHJLPHQ IRU FKLOGUHQ ZLWK &) 7KH\ DVNHG HDFK FKLOG WR LQGLFDWH WKH SUHVFULEHG DPRXQW RI FKHVW SK\VLRWKHUDS\ WKH W\SH RI DQWLELRWLF PXOWLYLWDPLQ DQG SDQFUHDWLF VXSSOHPHQW DQG FRUUHVSRQGLQJ IUHTXHQFLHV DQG GRVLQJ WLPHV 5HJUHWWDEO\ DOWKRXJK WKH UHVHDUFKHUV FROOHFWHG WKHVH LPSRUWDQW GDWD WKH\ GLG QRW UHSRUW DQ\ LQIRUPDWLRQ DERXW WKH NQRZOHGJHUHODWHG ILQGLQJV

PAGE 20

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t 7HSSHU .DWNR -RKQVRQ )RZOHU t 7XUQHU /DLQH HW DO 0RQDQH *XUZLW] 0RQDQH t $YRP 2VWURS t *LOO 6LQJK HW DO 6LQJK HW DO :DWVRQ t )DUOH\ f :LWK WKLV PHWKRG SKDUPDF\ UHILOO LQIRUPDWLRQ LV REWDLQHG IRU D VSHFLILF WLPH SHULRG DQG WKHQ DGKHUHQFH UDWHV DUH FDOFXODWHG 0HWKRGV RI FDOFXODWLRQ PD\ YDU\ DV VHYHUDO VWXGLHV KDYH IDLOHG WR UHSRUW UHILOO DGKHUHQFH FDOFXODWLRQ PHWKRGV )DUOH\ HW DO 2VWURS t *LOO 6LQJK HW DO f 0RVW FRPPRQO\ WKH QXPEHU RI GD\V IRU ZKLFK WKH PHGLFDWLRQ ZDV SUHVFULEHG GXULQJ WKH LQWHUYDO LV GLYLGHG E\ WKH QXPEHU RI GD\V IRU ZKLFK WKH PHGLFDWLRQ ZDV GLVSHQVHG .DWNR -RKQVRQ )RZOHU t 7XUQHU 0RQDQH HW DO f 7LPH SHULRGV RYHU ZKLFK UHILOO DGKHUHQFH ZDV DVVHVVHG KDYH YDULHG IURP RQH PRQWK WR RQH \HDU )DUOH\ HW DO .DWNR -RKQVRQ )RZOHU t 7XUQHU /DLQH HW DO 0RQDQH *XUZLW] 0RQDQH t $YRP 6LQJK HW DO 6LQJK HW DO :DWVRQ t )DUOH\ )DUOH\ HW DO f

PAGE 21

3KDUPDF\ UHILOO KLVWRULHV DUH D SUDFWLFDO DQG HIIHFWLYH PHDQV WR DVVHVV DGKHUHQFH )LUVW WKH\ QHFHVVLWDWH PLQLPDO SDWLHQW EXUGHQ 3DWLHQWV DUH DVNHG WR VLJQ D UHOHDVH IRUP DXWKRUL]LQJ WKHLU SKDUPDF\ RU SKDUPDFLHVf WR UHOHDVH UHILOO KLVWRU\ LQIRUPDWLRQ DQG WKHQ WKH UHVHDUFKHU REWDLQV WKH LQIRUPDWLRQ IURP WKH SKDUPDFLHV ZLWKRXW UHTXLULQJ DGGLWLRQDO SDWLHQW HIIRUW 6HFRQG WKH YDOLGLW\ RI WKLV PHWKRGRORJ\ KDV EHHQ VXSSRUWHG E\ D VWURQJ FRUUHODWLRQ EHWZHHQ UHILOO KLVWRU\ DQG DGKHUHQFH GDWD EDVHG RQ WKH 0HGLFDWLRQ (YHQWV 0RQLWRULQJ 6\VWHP 0(06 )DUOH\ HW DO f $OWKRXJK UHILOO KLVWRULHV SURYLGH RQO\ D JURVV HVWLPDWLRQ RI DGKHUHQFH ZKLFK PD\ EH VXEMHFW WR HUURU 3DHV %DNNHU t 6RH $JQXH f WKH\ DUH XVHIXO LQ LGHQWLI\LQJ SDWLHQWV ZKR IDLO WR UHILOO WKHLU SUHVFULSWLRQV GHVSLWH WHOOLQJ WKHLU SURYLGHUV WKDW WKH\ FRQWLQXH WR WDNH WKHLU PHGLFDWLRQ DV SUHVFULEHG 6HOI5HSRUW $QRWKHU LQGLUHFW PHWKRG LV VHOIUHSRUW DVVHVVPHQW 6HOIUHSRUW DVVHVVPHQWV YDU\ LQ WKHLU VRSKLVWLFDWLRQ IURP VLQJOH TXHVWLRQV WR H[WHQVLYH TXHVWLRQQDLUHV DQG LQWHUYLHZV 6LPLODUO\ WKH UHOLDELOLW\ DQG YDOLGLW\ RI VXFK PHDVXUHV YDU\ *HQHUDOO\ SK\VLFLDQV WHQG WR RYHUHVWLPDWH UHJLPHQ DGKHUHQFH )LQQH\ +RRN )ULPDQ 5DSRII t &KULVWRSKHUVHQ 5DQG t :HHNV f DQG SDUHQW DQG FKLOG UHSRUWV DUH KLJKO\ YDULDEOH LQ WKHLU GHJUHH RI DFFXUDF\ ZLWK VRFLDO GHVLUDELOLW\ PHPRU\ DQG LQYROYHPHQW LQ WKH UHJLPHQ DIIHFWLQJ WKH LQWHJULW\ RI WKH GDWD 5DQG t :HHNV f 6HOIUHSRUW PHDVXUHV WKDW LQTXLUH DERXW EULHI VSHFLILF WLPH SHULRGV WHQG WR EH PRVW DFFXUDWH .DSODQ t 6LPRQ .OLQQHUW 0F4XDLG t *DYLQ f :KHQ WKH DVVHVVPHQW LV FRQGXFWHG DFFRUGLQJO\ WKH VHOIUHSRUW PHDVXUH PD\ EH YDOXDEOH LQ DVVHVVLQJ GHJUHHV RI DGKHUHQFH $OWKRXJK VHOIUHSRUWV DUH OLPLWHG E\ D WHQGHQF\ WR RYHUHVWLPDWH DGKHUHQFH EHKDYLRUV ZKHQ VHOIUHSRUWV GR LQGLFDWH SRRU DGKHUHQFH WKH LQIRUPDWLRQ LV FRQVLGHUHG UHOLDEOH

PAGE 22

1XPHURXV VHOIUHSRUW PHDVXUHV KDYH EHHQ GHYHORSHG DQG XVHG WR DVVHVV DGKHUHQFH LQ FKLOGUHQ ZLWK FKURQLF LOOQHVVHV WKRXJK IHZ PHDVXUHV KDYH JDLQHG ZLGHVSUHDG SRSXODULW\ DQG XVH 2QH FRPPRQ PHWKRG RI VHOIUHSRUW WKDW KDV GHPRQVWUDWHG JRRG UHOLDELOLW\ LV WKH UHFDOO LQWHUYLHZ RU GLDU\ PHWKRG 6SHFLILFDOO\ WKH KRXU UHFDOO LQWHUYLHZ )UHXQG -RKQVRQ 6LOYHUVWHLQ t 7KRPDV -RKQVRQ 6LOYHUVWHLQ 5RVHQEORRP &DUWHU t &XQQLQJKDP -RKQVRQ HW DO f KDV EHHQ XVHG VXFFHVVIXOO\ LQ WKH VWXG\ RI DGKHUHQFH LQ GLDEHWHV 7KH KRXU UHFDOO LQWHUYLHZ LV D GLDEHWHVVSHFLILF DVVHVVPHQW RI DGKHUHQFH 7KH DVVHVVPHQW SURFHGXUH LQFOXGHV WKUHH SKRQH LQWHUYLHZV WZR GXULQJ ZHHNGD\V RQH RQ ZHHNHQGf ZLWKLQ D WZRZHHN SHULRG 3DWLHQWV DQG SDUHQWV DUH LQWHUYLHZHG VHSDUDWHO\ DQG DVNHG WR UHFDOO WKH HYHQWV RI WKH SUHYLRXV KRXUV ,I WKH\ IDLO WR PHQWLRQ GLDEHWHV DGKHUHQFH EHKDYLRUV WKH LQWHUYLHZHU FXHV WKHP 4XDQWLILHG PHDVXUHV FRUUHVSRQG WR HDFK DVSHFW RI WKH UHJLPHQ EHKDYLRU LQVXOLQ LQMHFWLRQV EORRGJOXFRVH WHVWLQJ GLHW DQG H[HUFLVHf *HQHUDOO\ VWXGLHV KDYH VXSSRUWHG WKH SV\FKRPHWULF SURSHUWLHV RI WKH KRXU UHFDOO LQWHUYLHZ &RUUHODWLRQV EHWZHHQ FKLOG DQG SDUHQW UHSRUWV KDYH UDQJHG IURP U WR U 0 f VXJJHVWLQJ PRGHUDWH LQWHUUDWHU UHOLDELOLW\ -RKQVRQ HW DO f /DFN RI FRPSOHWH FRQFRUGDQFH LV QRW VXUSULVLQJ JLYHQ WKDW SDUHQWV GR QRW REVHUYH DOO RI WKHLU FKLOGUHQfV EHKDYLRU DQG WKH OLNHOLKRRG RI HUURUV RI PHPRU\ $JUHHPHQW EHWZHHQ FKLOGUHSRUWHG DGKHUHQFH EHKDYLRUV DQG REVHUYHG EHKDYLRUV VXSSRUWV WKH FRQVWUXFW YDOLGLW\ RI WKH PHDVXUH 5H\QROGV -RKQVRQ t 6LOYHUVWHLQ f *UHDWHVW FRQFRUGDQFH ZDV IRXQG IRU WKH RFFXUUHQFH RU QRQRFFXUUHQFH RI EHKDYLRUV UHJDUGLQJ LQVXOLQ LQMHFWLRQ H[HUFLVH DQG EORRGJOXFRVH WHVWLQJ 3RRUHU DJUHHPHQW ZDV IRXQG IRU WKH WLPH GXULQJ ZKLFK WKH EHKDYLRUV RFFXUUHG 7KH SRRUHVW DJUHHPHQW ZDV IRXQG IRU VSHFLILF PHDVXUHV RI

PAGE 23

GLHWDU\ EHKDYLRUV HJ JUDPV RI FDUERK\GUDWH JUDPV RI IDW DQG WRWDO FDORULHVf 2YHU D WKUHHPRQWK WLPH SHULRG PRGHUDWH VXSSRUW ZDV IRXQG IRU WKH VWDELOLW\ RI WKH PHDVXUH ZLWK JUHDWHU VWDELOLW\ IRU EORRGJOXFRVH WHVWLQJ DQG GLHWDU\ EHKDYLRUV FRPSDUHG WR H[HUFLVH DQG LQMHFWLRQ EHKDYLRUV )UHXQG HW DK f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

PAGE 24

7KH FOLQLFDO LQWHUYLHZ LV D UHODWLYHO\ LQH[SHQVLYH XQREWUXVLYH PHWKRGRORJ\ WKDW DOORZV IRU LGHRJUDSKLF DVVHVVPHQW RI DGKHUHQFH EHKDYLRUV DQG EDUULHUV ,W LV SUDFWLFDO IRU XVH LQ ERWK FOLQLFDO DQG UHVHDUFK GRPDLQV +RZHYHU IHZ VWXGLHV KDYH XVHG WKH FOLQLFDO LQWHUYLHZ IRU WKH DVVHVVPHQW RI DGKHUHQFH EHKDYLRUV LQ FKLOGUHQ ZLWK FKURQLF LOOQHVVHV +DQVRQ HW DO /D *UHFD f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t 6FKDIHU f ,Q WKLV VWXG\ WKH %DUULHUV WR $GKHUHQFH 4XHVWLRQQDLUH %$4f ZDV GHYHORSHG EDVHG RQ WKH %HKDYLRU $QDO\WLF 0RGHO *ROGIULHG DQG 'f=XULOOD f 6L[ LQGLYLGXDOV ZLWK LQVXOLQGHSHQGHQW GLDEHWHV PHOOLWXV /''0f DQG WZR GLDEHWHV QXUVHHGXFDWRUV JHQHUDWHG D OLVW RI EDUULHUV WR DGKHUHQFH WR YDULRXV DVSHFWV RI WKH GLDEHWHV UHJLPHQ GLHW H[HUFLVH JOXFRVH WHVWLQJ DQG LQVXOLQ LQMHFWLRQVf 1H[W WKH JHQHUDWHG LWHPV ZHUH HPSLULFDOO\ WHVWHG DQG D VFDOH ZLWK LWHPV ZDV GHULYHG )UHTXHQF\ DQG VHYHULW\ UDWLQJV ZHUH SURYLGHG DOWKRXJK WKH VHYHULW\ UDWLQJV ZHUH HOLPLQDWHG GXH WR KLJK LQWHUFRUUHODWLRQ ZLWK WKH IUHTXHQF\ UDWLQJV 1R GDWD DUH DYDLODEOH

PAGE 25

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f UHSRUWV WKDW KH RIWHQ DVVHVVHV EDUULHUV WR DGKHUHQFH GXULQJ FOLQLFDO LQWHUYLHZ E\ DVNLQJ SDWLHQWV f:KDW JHWV LQ WKH ZD\ RI \RX WDNLQJ \RXU PHGLFLQH"f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t %XUJKHQ E -RKQVRQ 6LOYHUVWHLQ 5RVHQEORRP &DUWHU t &XQQLQJKDP 5LFNHU 'HODPDWHU t +VX f 7KLV PD\ EH LQ SDUW EHFDXVH SDUHQWV PD\ WHQG WR GLVFRQWLQXH RU GHFUHDVH VXSHUYLVLRQ RI GLVHDVH PDQDJHPHQW EHKDYLRUV DV FKLOGUHQ EHFRPH ROGHU -RKQVRQ f 0RUHRYHU DGROHVFHQFH LQ SDUWLFXODU LV D GHYHORSPHQWDO SHULRG VWURQJO\ UHODWHG WR GHFUHDVHG GLVHDVH PDQDJHPHQW EHKDYLRUV $GROHVFHQFH LV D WLPH RI VWULYLQJ IRU LQGHSHQGHQFH DQG RQH ZD\ LQ ZKLFK DGROHVFHQWV VRPHWLPHV DVVHUW WKHLU LQGHSHQGHQFH LV E\ FKRRVLQJ WR QRW SHUIRUP GLVHDVH PDQDJHPHQW EHKDYLRUV

PAGE 26

/RQJ GXUDWLRQ RI WKH UHJLPHQ DQG VSHFLILF UHJLPHQ DWWULEXWHV KDYH DOVR EHHQ LGHQWLILHG DV LPSRUWDQW EDUULHUV WR DGKHUHQFH )LUVW ORQJ GXUDWLRQ RI WKH UHJLPHQ PD\ KLQGHU DGKHUHQFH ,W KDV EHHQ VXJJHVWHG WKDW DGKHUHQFH LV D JUHDWHU SUREOHP ZLWK FKURQLF YHUVXV DFXWH UHJLPHQV 5DSRII f 5HODWHGO\ ZLWK FKLOGKRRG FKURQLF LOOQHVVHV VWXGLHV KDYH IRXQG WKDW DGKHUHQFH WHQGV WR GHFUHDVH RYHU WLPH %URZQEULGJH t )LHOGLQJ +DQVRQ 'H*XLUH 6FKLQNHO +HQJJHOHU t %XUJKHQ +DQVRQ +HQJJHOHU +DUULV %XUJKHQ t 0RRUH -DFREVRQ HW DO f ,W PD\ EH WKDW FKLOGUHQ EHJLQ WR UHVHQW PHGLFDWLRQ WDNLQJ DQG PD\ GHFLGH WR VWRS WDNLQJ LW UHJXODUO\ $OVR DV IDPLOLHV DGMXVW WR WKH PHGLFDWLRQ UHJLPHQ SDUHQWV PD\ EHJLQ WR UHO\ RQ WKHLU FKLOGUHQ WR PDQDJH WKHLU UHJLPHQV LQGHSHQGHQWO\ SURYLGLQJ RSSRUWXQLWLHV IRU FKLOGUHQ WR VNLS KLGH RU GXPS GRVHV 6SHFLILF DWWULEXWHV RI WKH UHJLPHQ PD\ DOVR PDNH DGKHUHQFH GLIILFXOW %HKDYLRULVWV KDYH ZHOO HVWDEOLVKHG WKH RUJDQLVPfV WHQGHQF\ WR DYRLG DYHUVLYH VWLPXOL DQG WKLV PD\ EH ZK\ FRPSOH[ UHJLPHQ FRPSRQHQWV OLNH FKHVW SK\VLRWKHUDS\ KDYH EHHQ DVVRFLDWHG ZLWK SRRU DGKHUHQFH 3DVVHUR 5HPRU t 6DORPRQ f %HKDYLRULVWV KDYH DOVR HVWDEOLVKHG WKH RUJDQLVPfV WHQGHQF\ WR SHUIRUP EHKDYLRUV WKDW DUH UHLQIRUFLQJ DQG WKLV PD\ H[SODLQ WKH GLIILFXOW\ FKLOGUHQ ZLWK GLDEHWHV H[SHULHQFH ZLWK DEVWDLQLQJ IURP VZHHW DQG IDWW\ IRRGV *ODVJRZ 0F&DXO t 6FKDIHU f 7KH +XPDQ ,PPXQRGHILFLHQF\ 9LUXV $ 1HZ &KURQLF ,OOQHVV :RUOGZLGH PRUH WKDQ PLOOLRQ SHRSOH DUH LQIHFWHG ZLWK +,9 RI WKRVH PLOOLRQ DUH FKLOGUHQ 81$,'6 f (DFK GD\ DOPRVW FKLOGUHQ DUH QHZO\ LQIHFWHG ZLWK WKH YLUXV 6SHFLILFDOO\ LQ WKH 8QLWHG 6WDWHV PRUH WKDQ FKLOGUHQ DUH OLYLQJ ZLWK +,9 81$,'6 f DQG PRUH WKDQ FKLOGUHQ DUH FXUUHQWO\ OLYLQJ ZLWK

PAGE 27

$,'6 WKH DGYDQFHG PDQLIHVWDWLRQ RI WKH YLUXV &'& f 1DWLRQDOO\ +,9 KDV EHHQ UDQNHG DV WKH VHYHQWK OHDGLQJ FDXVH RI GHDWK DPRQJ FKLOGUHQ DJHV WR 9HQWXUD 3HWHUV 0DUWLQ t 0DXUHU f ,PSURYLQJ PRUWDOLW\ DQG PRUELGLW\ DPRQJ WKHVH FKLOGUHQ DQG DGROHVFHQWV LV FRQWLQJHQW XSRQ +,9LQIHFWHG FKLOGUHQ UHFHLYLQJ DSSURSULDWH WUHDWPHQW 3DOHOOD HW DO f +,9 LV D UHWURYLUXV WKDW LV WUDQVPLWWHG WKURXJK H[SRVXUH WR LQIHFWHG EORRG DQG VH[XDO IOXLGV W\SLFDOO\ WKURXJK QHHGOH VKDULQJ XQSURWHFWHG VH[ RU PRWKHUWRFKLOG WUDQVPLVVLRQ &'& f +,9 GLVUXSWV WKH LPPXQH IXQFWLRQLQJ RI LQIHFWHG SHUVRQV $V D UHVXOW SHRSOH ZLWK +,9 KDYH GLIILFXOW\ ILJKWLQJ RII FHUWDLQ EDFWHULD YLUXVHV DQG RWKHU PLFUREHV 7KHUHIRUH WKH\ DUH SURQH WR GHYHORS RSSRUWXQLVWLF LQIHFWLRQV ZKLFK WDNH DGYDQWDJH RI WKH FRPSURPLVHG LPPXQH V\VWHP 2YHU PLFURRUJDQLVPV KDYH EHHQ LGHQWLILHG DV RSSRUWXQLVWLF LQFOXGLQJ &DQGLGD DOOELFDQV 9DULFHOOD=RVWHU 9LUXV KHUSHV VLPSOH[ YLUXV PHDVOHV F\WRPHJDORYLUXV FRQJHQLWDO V\SK\LOLV DQG SQHXPRQRF\WLF FDULQL SQHXPRQLD 3&3f ,I QRW WUHDWHG VXFFHVVIXOO\ VXFK RSSRUXQWLVWLF LQIHFWLRQV FDQ OHDG WR $,'6UHODWHG IDWDOLWLHV %XW] HW DO 3DOHOOD HW DO f $OVR +,9 FDQ OHDG WR QHXURORJLFDO GLVHDVH QHXURSV\FKRORJLFDO LPSDLUPHQW FDUGLDF DEQRUPDOLWLHV DQG KHPDWRORJLF SUREOHPV DPRQJ RWKHUV 6FRWW t 6OHDVPDQ f :LWK QR NQRZQ FXUH WKH HQG UHVXOW RI +,9 LV GHDWK ,Q WKH SDVW +,9 ZDV FRQVLGHUHG D GHELOLWDWLQJ LOOQHVV FKDUDFWHUL]HG E\ UDSLG GHWHULRUDWLRQ DQG GHPLVH +RZHYHU WKH GHYHORSPHQW RI QHZ GUXJ WKHUDSLHV IRU +,9 LQIHFWHG FKLOGUHQ DQG DGXOWV KDV OHG WR LQFUHDVHG OLIHVSDQV IRU LQIHFWHG LQGLYLGXDOV 0DUWLQR HW DO 3DOHOOD HW DO f 0DUWLQR DQG FROOHDJXHV f IRXQG WKDW ZKHQ FKLOGUHQ ZHUH SUHVFULEHG DW OHDVW DQWLUHWURYLUDO PHGLFDWLRQV WKH ULVN RI GHDWK

PAGE 28

GHFUHDVHG E\ b 1RZ WKDW +,9LQIHFWHG FKLOGUHQ DUH OLYLQJ LQWR DGROHVFHQFH DQG \RXQJDGXOWKRRG WKH LVVXH RI DGKHUHQFH WR WKHLU PHGLFDWLRQ UHJLPHQV LV EHFRPLQJ LQFUHDVLQJO\ PRUH LPSRUWDQW 7\SLFDOO\ WKH SUHIHUUHG SHGLDWULF DQWLUHWURYLUDO PHGLFDWLRQ UHJLPHQ FRQVLVWV RI FRPELQDWLRQ $57 VSHFLILFDOO\ KLJKO\ DFWLYH DQWLUHWURYLUDO WKHUDS\ +$$57 6FRWW t 6OHDVPDQ f :LWK VWDQGDUG +$$57 UHJLPHQV FKLOGUHQ DUH SUHVFULEHG DW OHDVW WKUHH PHGLFDWLRQV WR EH WDNHQ WZR RU WKUHH WLPHV SHU GD\ &KLOGUHQ PD\ WDNH WKHLU PHGLFDWLRQ LQ SLOO IRUP RU LQ OLTXLG GRVHV HLWKHU EHFDXVH WKH\ DUH XQDEOH WR VZDOORZ SLOOV RU EHFDXVH VRPH SLOOV DUH QRW DYDLODEOH LQ VPDOO GRVHV DSSURSULDWH WR ERG\ ZHLJKW 0DQ\ RI WKH DQWLUHWURYLUDO PHGLFDWLRQV KDYH XQSDODWDEOH WDVWH DQG JDVWURLQWHVWLQDO VLGH HIIHFWV 6LGH HIIHFWV IRU DQWLUHWURYLUDO WKHUDS\ $57f LQFOXGH JDVWURLQWHVWLQDO GLVFRPIRUW QDXVHD YRPLWLQJ GLDUUKHD KHDGDFKH SHULSKHUDO QHXURSDWK\ SDUDVWKHVLV UDVK PRXWK XOFHUV GU\ VNLQ DQG RWKHUV LQFOXGLQJ RUJDQ GDPDJH &DUSHQWHU HW DO f 6LGH HIIHFWV UDQJH LQ IUHTXHQF\ GXUDWLRQ DQG VHYHULW\ 7R FRPEDW VLGH HIIHFWV DV ZHOO DV RSSRUWXQLVWLF LQIHFWLRQV DQG RWKHU FRPSOLFDWLRQV KHDOWKFDUH SURYLGHUV PD\ SUHVFULEH DGGLWLRQDO PHGLFDWLRQV .HOO\ 2WWR6ODM 6LNNHPD 3LQNHUWRQ t %ORRP f VXFK DV =DQWDF 6HSWUD DQG D YDULHW\ RI RWKHUV :LWK $57 DQG RWKHU GUXJV WR FRPEDW VLGH HIIHFWV LQIHFWLRQV DQG FRPSOLFDWLRQV PHGLFDWLRQ UHJLPHQV IRU +,9LQIHFWHG FKLOGUHQ FDQ EH FRPSOH[ 0DQ\ RI WKH PHGLFDWLRQV KDYH VSHFLDO LQVWUXFWLRQV WKDW DUH GLIILFXOW WR IROORZ )RU H[DPSOH WR HQVXUH SURSHU DEVRUSWLRQ VRPH PHGLFDWLRQV VKRXOG EH WDNHQ ZLWK IRRG ZKLOH RWKHUV VKRXOG QRW .HOO\ HW DK f 2WKHUV UHTXLUH D ODUJH DPRXQW RI IOXLG WDNH $GGLWLRQDOO\ WR SUHYHQW GUXJ LQWHUDFWLRQV DQG WR PDLQWDLQ WKH DSSURSULDWH OHYHO RI PHGLFDWLRQV LQ WKH EORRGVWUHDP

PAGE 29

RIWHQ WKH WLPLQJ RI PHGLFDWLRQ WDNLQJ LV FUXFLDO .HOO\ HW DO f :LWK VXFK FRPSOH[ LQVWUXFWLRQV LW LV QRW VXUSULVLQJ WKDW DGKHUHQFH WR $57 LV RI JUHDW FRQFHUQ )ULHGODQG t :LOOLDPV .HOO\ HW DO f ,PSOLFDWLRQV RI 3RRU $GKHUHQFH WR $57 $V ZLWK RWKHU FKURQLF LOOQHVVHV SRRU DGKHUHQFH WR $57 DIIHFWV KHDOWKFDUH FRVWV FOLQLFDO GHFLVLRQPDNLQJ FRQFOXVLRQV GUDZQ IURP FOLQLFDO WULDOV DQG KHDOWK RXWFRPHV 7KH ZLGHVSUHDG XVH RI +$$57 KDV UHGXFHG PRUELGLW\ DQG PRUWDOLW\ KRVSLWDOL]DWLRQV WKH GHYHORSPHQW RI RSSRUWXQLVWLF LQIHFWLRQV DQG WKH XVH RI KRPHFDUH VHUYLFHV DQG KRVSLFHV %UHWWOH HW DO f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t :LOOLDPV f 7KH SK\VLFLDQ PD\ KDYH SUHVFULEHG DQRWKHU PHGLFDWLRQ ZLWK IHZHU VLGH HIIHFWV RU ZLWK VLPSOHU GRVLQJ ZLWK ZKLFK WKH SDWLHQW PLJKW

PAGE 30

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f WKH FRQVHTXHQFHV RI SRRU DGKHUHQFH LQ +,9 FOLQLFDO WULDOV PD\ EH ZLGHVSUHDG DQG VLJQLILFDQW $V DQWLUHWURYLUDO WKHUDS\ KDV EHJXQ WR VORZ GLVHDVH SURJUHVVLRQ DQG LQFUHDVH OLIH VSDQ WKH LVVXH RI DGKHUHQFH WR PHGLFDWLRQ WUHDWPHQW KDV EHFRPH PRUH VDOLHQW )ULHGODQG DQG :LOOLDPV f VWDWHG WKDW fWKH NH\ WR VXFFHVV RI WKH QHZ UHJLPHQV LV WKH DELOLW\ DQG ZLOOLQJQHVV RI +,9SRVLWLYH LQGLYLGXDOV WR DGKHUH WR FRPSOH[ DQWLUHWURYLUDO UHJLPHQV SHUKDSV IRU OLIHf $GKHUHQFH LV HVVHQWLDO WR VXSSUHVVLQJ WKH YLUDO ORDG WKH DPRXQW RI +,9 LQ EORRG $OWKRXJK PXOWLSOH IDFWRUV PD\ FRQWULEXWH WR LQDGHTXDWH YLUDO VXSSUHVVLRQ SRRU DGKHUHQFH WR $57 UHJLPHQV LV DPRQJ WKH PRVW LPSRUWDQW 0HOYLQ f 5HVHDUFKHUV KDYH DWWHPSWHG WR LGHQWLI\ WKH DGKHUHQFH UDWHV QHFHVVDU\ WR DFKLHYH RSWLPDO KHDOWK $PRQJ DGXOWV RQH SURVSHFWLYH VWXG\ DVVHVVHG DGKHUHQFH OHYHOV DPRQJ PRVWO\ PDOH SDWLHQWV ZLWK 0(06 IRU WR PRQWKV 3DWHUVRQ HW DO f 5HVXOWV VKRZHG WKDW SDWLHQWV ZLWK b DGKHUHQFH WR $57 UHJLPHQV KDYH ORZHU YLUDO ORDGV JUHDWHU LQFUHDVHV LQ &' FRXQWV DPRXQW RI 7O\PSKRF\WH ZKLWH EORRG FHOOVf DQG ORZHU KRVSLWDOL]DWLRQ UDWHV WKDQ WKRVH ZLWK SRRUHU DGKHUHQFH $QRWKHU VWXG\ DVVHVVHG DGKHUHQFH DPRQJ +,9LQIHFWHG DGXOWV RQ $57 %DQJVEHUJ HW DK f 5HVHDUFKHUV IRXQG WKDW

PAGE 31

SLOO FRXQW DGKHUHQFH VLJQLILFDQWO\ SUHGLFWHG YLUDO ORDG VXFK WKDW D b GLIIHUHQFH LQ DGKHUHQFH ZDV DVVRFLDWHG ZLWK D GRXEO\ LQFUHDVHG YLUDO ORDG VXJJHVWLQJ WKDW b DGKHUHQFH RU JUHDWHU LV QHFHVVDU\ IRU YLUDO VXSSUHVVLRQ 0F1DEE DQG FROOHDJXHV f DOVR XVHG WKH 0(06 WR DVVHVV DGKHUHQFH DPRQJ PRVWO\PDOH +,9LQIHFWHG DGXOWV 5HVXOWV RI WKLV VWXG\ DUH VLPLODU WR WKH UHVXOWV RI %DQVEHUJ HW DO f LQ VXJJHVWLQJ WKDW b DGKHUHQFH LV QHFHVVDU\ IRU IXOO YLUDO VXSSUHVVLRQ :KLOH WKHVH VWXGLHV VXJJHVW WKDW DW OHDVW b DGKHUHQFH LV QHFHVVDU\ IRU YLURORJLF VXFFHVV RQH VWXG\ DPRQJ D SHGLDWULF SRSXODWLRQ VXJJHVWV WKDW ORZHU OHYHOV PD\ EH VXIILFLHQW )DUOH\ HW DO f )DUOH\ DQG FROOHDJXHV XVHG 0(06 DQG SKDUPDF\ UHILOO GDWD WR DVVHVV DGKHUHQFH WR RQH $57 GUXJ RYHU D PRQWK SHULRG 0(06 GDWD DORQH VXJJHVWHG WKDW DQ DGKHUHQFH UDWH RI b ZDV UREXVWO\ DVVRFLDWHG ZLWK YLURORJLF VXFFHVV DQG ZKHQ SKDUPDF\ UHILOO GDWD ZHUH FRPELQHG ZLWK 0(06 GDWD WKH UDWH RI b DGKHUHQFH ZDV HYHQ PRUH VWURQJO\ DVVRFLDWHG ZLWK YLURORJLF VXFFHVV 7KXV DPRQJ FKLOGUHQ DW OHDVW b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f 0XWDWLRQV RFFXU QDWXUDOO\ ZLWK +,9 EXW DUH SUHYHQWHG ZKHQ $57 VWRSV YLUDO UHSOLFDWLRQ FRPSOHWHO\ :KHQ WKH YLUXV LV SDUWLDOO\ VXSSUHVVHG DV RFFXUV ZLWK LQWHUPLWWHQW RU LQFRPSOHWH DGKHUHQFH GUXJV PD\ UHQGHU FRPPRQ VWUDLQV LQDFWLYH EXW PD\ QRW SUHYHQW

PAGE 32

PXWDWLRQV IURP UHSOLFDWLQJ 7KH UHVXOW LV D KLJK SUHVHQFH RI WKH PXWDQW YLUXV LQ WKH EORRG WKDW LV UHVLVWDQW WR WKRVH GUXJV DQG VLPLODU GUXJV DV ZHOO 7KLV LV SDUWLFXODUO\ SUREOHPDWLF VLQFH UHVLVWDQFH PD\ RFFXU PRVW UDSLGO\ DPRQJ WKRVH ZKRVH DGKHUHQFH IDOOV EHWZHHQ b DQG b %DQJVEHUJ HW DO f D UDQJH LQ ZKLFK PDQ\ FKLOGUHQ DQG DGXOWV RQ +$$57 PD\ IDOO %HO]HU )XFKV /XIWPDQ t 7XFNHU .DVWULVVLRV HW DO :DWVRQ t )DUOH\ f )RU ERWK WKH LQGLYLGXDO DQG VRFLHW\ WKH LPSDFW RI SRRU DGKHUHQFH DQG VXEVHTXHQW GUXJ UHVLVWDQFH PD\ EH VHYHUH )LUVW ZLWK WKH OLPLWHG QXPEHU RI DQWLUHWURYLUDO DJHQWV DQG WKH DELOLW\ RI UHVLVWDQFH WR DIIHFW HQWLUH GUXJ FODVVHV UHVLVWDQFH PD\ OHDYH DQ LQGLYLGXDO ZLWK IHZ RSWLRQV IRU DGGLWLRQDO WKHUDS\ *DOODQW f $V ZDV VHHQ LQ WKH HDUO\ fV ZLWKRXW WKHUDS\ WKH YLUXV UHSOLFDWHV DQG FRPSOHWHO\ VXSSUHVVHV WKH LPPXQH V\VWHP :KHQ WKH LPPXQH V\VWHP LV VXSSUHVVHG WKH ERG\ FDQQRW ILJKW HYHQ D FRPPRQ FROG 8QIRUWXQDWHO\ ZLWKRXW WKHUDS\ UDSLG SK\VLFDO GHFOLQH LV LPPLQHQW 6HFRQG UHVHDUFK KDV GHPRQVWUDWHG WKDW GUXJ UHVLVWDQFH OHDGV WR LQFUHDVHG YLUDO UHSOLFDWLRQ LQ WKH EORRG VHPHQ DQG YDJLQDO IOXLG *XSWD HW DO 9HPD]]D *LOOHP t )OHSS f 7KH SRWHQF\ RI WKH YLUXV LQ WKHVH IOXLGV LQFUHDVHV WKH OLNHOLKRRG WKDW WKH YLUXV PD\ EH WUDQVPLWWHG WR H[SRVHG LQGLYLGXDOV :DLQEHUJ t )ULHGODQG f ,I WKH GUXJUHVLVWDQW YLUXV LV WUDQVPLWWHG WR RWKHUV WKRVH LQIHFWHG ZLOO KDYH D GUXJUHVLVWDQW YLUXV DV ZHOO )ULHGODQG t :LOOLDPV f /LNH WKH SRRUO\ DGKHUHQW LQGLYLGXDOV ZLWK GUXJ UHVLVWDQFH QHZO\ LQIHFWHG LQGLYLGXDOV ZLOO KDYH OLWWOH RU QR RSWLRQV IRU WKHUDS\ $V D UHVXOW IXWXUH JHQHUDWLRQV RI +,9SRVLWLYH LQGLYLGXDOV PD\ VXIIHU GUDPDWLFDOO\ IURP WKH SRRU DGKHUHQFH RI WKHLU SUHGHFHVVRUV &OHDUO\ DPRQJ +,9LQIHFWHG LQGLYLGXDOV WKH SRWHQWLDO SXEOLF KHDOWK LPSOLFDWLRQV RI SRRU DGKHUHQFH DQG VXEVHTXHQW GUXJ UHVLVWDQFH DUH JUDYH

PAGE 33

$VVHVVPHQW RI $GKHUHQFH DPRQJ &KLOGUHQ ZLWK +,9 'HVSLWH WKH LPSOLFDWLRQV RI SRRU DGKHUHQFH DPRQJ FKLOGUHQ RQ $57 IHZ VWXGLHV KDYH H[DPLQHG DGKHUHQFH OHYHOV LQ WKLV SRSXODWLRQ $OWKRXJK PDQ\ VWXGLHV KDYH DVVHVVHG DGKHUHQFH DV RQH RI WKH LQFOXVLRQ FULWHULD IRU FOLQLFDO WULDOV : 6OHDVPDQ SHUVRQDO FRPPXQLFDWLRQ -XO\ f DQG DV SDUW RI WKH 3HGLDWULF $LGV &OLQLFDO 7ULDOV *URXS DVVHVVPHQW EDWWHU\ IHZ NQRZQ VWXGLHV KDYH UHSRUWHG V\VWHPDWLF DVVHVVPHQWV RI DGKHUHQFH DQG QR NQRZQ VWXGLHV KDYH DVVHVVHG NQRZOHGJH RI WKH SUHVFULEHG PHGLFDWLRQ UHJLPHQ DPRQJ +,9LQIHFWHG FKLOGUHQ 5HJLPHQ NQRZOHGJH 2QO\ RQH VWXG\ KDV DVVHVVHG UHJLPHQ NQRZOHGJH DPRQJ FDUHJLYHUV RI +,9 LQIHFWHG FKLOGUHQ RU DPRQJ WKH FKLOGUHQ WKHPVHOYHV .DWNR DQG FROOHDJXHV f DVNHG FDUHJLYHUV WR QDPH RU GHVFULEH WKHLU FKLOGUHQfV $57 PHGLFDWLRQV DQG FRUUHVSRQGLQJ GRVDJHV DQG GRVLQJ IUHTXHQFLHV 1LQHWHHQ bf RI WKH FDUHJLYHUV SURYLGHG DFFXUDWH PHGLFDWLRQ LQIRUPDWLRQ DQG RI WKRVH KDG SKDUPDF\ UHILOO UDWLQJV RI b DGKHUHQFH RU EHWWHU ZKLOH QRQH RI WKRVH ZKR ODFNHG NQRZOHGJH RI WKH UHJLPHQ KDG b DGKHUHQFH UDWLQJV RU KLJKHU 7KLV VWXG\ ZDV OLPLWHG E\ D VPDOO VDPSOH VL]H DQG RQO\ RQH PHWKRG RI DGKHUHQFH DVVHVVPHQW \HW WKH VWXG\ LV LPSRUWDQW DV LW VXSSRUWV WKH XVH RI NQRZOHGJH DVVHVVPHQW DV DQ LQGLFDWRU RI SRVVLEOH SRRU DGKHUHQFH $OWKRXJK RQO\ RQH VWXG\ KDV LQFOXGHG UHJLPHQ NQRZOHGJH DV SDUW RI WKH DGKHUHQFH DVVHVVPHQW RI +,9LQIHFWHG SHGLDWULF SRSXODWLRQV VHYHUDO VWXGLHV KDYH DVVHVVHG UHJLPHQ NQRZOHGJH DPRQJ +,9LQIHFWHG DGXOWV ,Q D VPDOO VWXG\ 6YDUVWDG &KHZQLQJ 6OHDWK DQG &ODHVVRQ f DVVHVVHG UHJLPHQ NQRZOHGJH DPRQJ +,9 LQIHFWHG PHQ DQG ZRPHQ $V SDUW RI WKH %ULHI 0HGLFDWLRQ 4XHVWLRQQDLUH %04f WKH\ FRQGXFWHG D f5HJLPHQ 6FUHHQf 3DWLHQWV ZHUH DVNHG WR OLVW DOO RI WKH PHGLFDWLRQV WDNHQ

PAGE 34

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f LQFOXGHG D EULHI UHJLPHQ NQRZOHGJH DVVHVVPHQW LQ DQRWKHU VPDOO VWXG\ )RUW\VL[ PRVWO\ PDOH SDWLHQWV RQ +$$57 ZHUH DVNHG DERXW WKHLU PHGLFDWLRQ UHJLPHQ GXULQJ DQ LQWHUYLHZ DW SDWLHQWVf W\SLFDO SODFHV RI UHVLGHQFH 6L[ SDWLHQWV UHSRUWHG WKDW WKH\ ZHUH WDNLQJ HLWKHU D GLIIHUHQW QXPEHU RI GRVHV SHU GD\ RU D GLIIHUHQW QXPEHU RI SLOOV SHU GRVH WKDQ WKH PHGLFDWLRQ FRQWDLQHU LQVWUXFWHG )RXU RI WKRVH SDWLHQWV DWWULEXWHG WKLV GLIIHUHQFH WR PLVXQGHUVWDQGLQJ ZKLOH WZR UHSRUWHG WKDW WKH\ KDG GHFLGHG WR WDNH WKHLU PHGLFLQH GLIIHUHQWO\ WKDQ LW ZDV SUHVFULEHG ,Q D ODUJHU VWXG\ 3DULHWQWL DQG FROOHDJXHV f XVHG D SLOO LGHQWLILFDWLRQ WHVW WR GHWHUPLQH UHJLPHQ NQRZOHGJH DPRQJ +,9LQIHFWHG DGXOWV LQ )UDQFH 3DUWLFLSDQWV ZHUH JLYHQ D ERDUG WKDW FRQWDLQHG SLFWXUHV RI $57 PHGLFDWLRQV ZLWK VLPLODUn ORRNLQJ WZLQ SLOOVf IRU HDFK $57 PHGLFDWLRQ $ VFRUH ZDV FDOFXODWHG EDVHG RQ WKH QXPEHU RI PLVLGHQWLILFDWLRQV ZHLJKWHG DFFRUGLQJ WR WKH GHJUHH RI SLOO UHVHPEODQFH 1R LQIRUPDWLRQ ZDV SURYLGHG DERXW WKH WRWDO SHUFHQW RI SDWLHQWV ZLWK LQDGHTXDWH NQRZOHGJH WKRXJK UHVXOWV VKRZHG WKDW b RI SDWLHQWV ZLWK SRRU DGKHUHQFH EDVHG RQ DQ LQYHVWLJDWRU FRPSOHWHG VFDOH DOVR KDG SRRU UHJLPHQ NQRZOHGJH

PAGE 35

$V SDUW RI DQRWKHU ODUJH PXOWLFHQWHU VWXG\ 6WRQH DQG RWKHUV f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b UHSRUWHG LQFRUUHFW LQIRUPDWLRQ DERXW ERWK IUHTXHQF\ DQG GLHWDU\ UHTXLUHPHQWV b UHSRUWHG LQFRUUHFW LQIRUPDWLRQ DERXW IUHTXHQF\ RQO\ DQG b UHSRUWHG LQFRUUHFW LQIRUPDWLRQ VROHO\ DERXW WKH GLHWDU\ UHTXLUHPHQWV 7KH UHODWLRQVKLS EHWZHHQ UHJLPHQ NQRZOHGJH DQG DGKHUHQFH EHKDYLRU ZDV QRW UHSRUWHG $QRWKHU PXOWLVLWH VWXG\ XVHG D FRPSXWHUDVVLVWHG VWUXFWXUHG LQWHUYLHZ &$6,f WR DVVHVV UHJLPHQ NQRZOHGJH DPRQJ +,9LQIHFWHG DGXOWV %DQJVEHUJ %URQVWRQH t +RIIPDQ f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b LQFRUUHFWO\ UHSRUWHG WKH QXPEHU RI GDLO\ GRVHV IRU DW OHDVW RQH SUHVFULEHG PHGLFDWLRQ DQG b LQFRUUHFWO\ UHSRUWHG WKH GLHWDU\

PAGE 36

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f H[DPLQHG DGKHUHQFH ZLWK SKDUPDF\ UHILOOV DPRQJ FKLOGUHQ DJHV PRQWKV WR \HDUV &KLOGUHQ ZHUH FRQVLGHUHG DGKHUHQW LI b RI $57 PHGLFDWLRQV ZHUH UHILOOHG GXULQJ WKH ILUVW GD\V RI 3, WKHUDS\ )LIW\HLJKW SHUFHQW RI FKLOGUHQ ZHUH FRQVLGHUHG DGKHUHQW E\ WKLV OLEHUDO FULWHULRQ 7KLV VXJJHVWV WKDW DW OHDVW b IDPLOLHV ZLWK FKLOGUHQ RQ $57 GLG QRW FODLP PRVW RI WKHLU SUHVFULSWLRQV IURP WKH SKDUPDF\ DQG WKHUHIRUH FRXOG QRW KDYH EHHQ FRQVLVWHQWO\ DGKHUHQW WR WKHLU UHJLPHQV .DWNR -RKQVRQ )RZOHU DQG 7XUQHU f XVHG SKDUPDF\ UHILOO GDWD WR H[DPLQH DGKHUHQFH DPRQJ FKLOGUHQ RQ $57 7KLV VWXG\ FRQVLGHUHG FKLOGUHQ DGKHUHQW LI RYHU D RQH\HDU SHULRG WKH SURSRUWLRQ RI GD\V IRU ZKLFK PHGLFDWLRQ ZDV GLVSHQVHG WR WKH GD\V IRU ZKLFK WKH PHGLFDWLRQ ZDV SUHVFULEHG ZDV OHVV WKDQ RU HTXDO WR b 2QO\ b ZHUH

PAGE 37

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t $YRP 6LQJK HW DO f DQG GHVSLWH WKH XQLGLPHQVLRQDO DSSURDFK WKH VWXG\ LV XVHIXO LQ VXJJHVWLQJ WKDW SRRU DGKHUHQFH PD\ EH D SHUYDVLYH SUREOHP IRU IDPLOLHV RI FKLOGUHQ RQ $57 UHJLPHQV $ UHFHQW VWXG\ ZLWK FKLOGUHQ \HDUVROG ZDV WKH ILUVW NQRZQ VWXG\ WR DVVHVV $57 DGKHUHQFH PXOWLGLPHQVLRQDOO\ DPRQJ FKLOGUHQ )DUOH\ HW DO f 7KH VWXG\ H[DPLQHG DGKHUHQFH DPRQJ FKLOGUHQ XVLQJ 0(06 SKDUPDF\ UHILOOV FDUHJLYHU VHOI UHSRUWV DQG SK\VLFLDQ DQG QXUVH DVVHVVPHQWV 0HGLDQ DGKHUHQFH UDWLQJV ZHUH b IRU 0(06 b IRU SKDUPDF\ UHILOOV DQG b ZKHQ FDUHJLYHUV ZHUH DVNHG WR UHSRUW WKH PLVVHG GRVHV GXULQJ WKH GD\V SULRU WR LQWHUYLHZ WKRXJK 0(06 GDWD GHPRQVWUDWHG PLVVHG PHGLFDWLRQ HYHQWV RI WKH SUHVFULEHG HYHQWV IRU WKH UHVSRQGHQWV RYHU WKH GD\ SHULRG 3K\VLFLDQ DQG QXUVH UDWLQJV ZHUH DOVR DVVHVVHG DQG ZHUH VLJQLILFDQWO\ UHODWHG WR 0(06 GDWD NDSSD b S f 3KDUPDF\ UHILOO UDWLQJV DORQH GLG QRW VLJQLILFDQWO\ UHODWH WR YLUDO ORDG GDWD EXW ZKHQ 0(06 DQG SKDUPDF\ GDWD ZHUH FRPELQHG DQ DGKHUHQFH FXWRII RI b ZDV D JRRG SUHGLFWRU RI YLURORJLF VXFFHVV 7KLV VWXG\ LV FRQVLVWHQW ZLWK VWXGLHV RI +,9LQIHFWHG DGXOWV ZKLFK LOOXVWUDWH WKH XWLOLW\ RI 0(06 DQG SKDUPDF\ UHILOO KLVWRULHV IRU DVVHVVLQJ $57 DGKHUHQFH .DWNR HW DO

PAGE 38

0F1DEE HW DO 3DWHUVRQ HW DO 6YDUVWDG HW DO f 7KLV VWXG\ DOVR GHPRQVWUDWHV WKDW GHVSLWH SUHYLRXV UHVHDUFK VXJJHVWLQJ WKDW SURYLGHUV RYHUHVWLPDWH DGKHUHQFH 5DQG t :HHNV f VRPH SURYLGHUV PD\ EH SDUWLFXODUO\ DVWXWH DW DVVHVVLQJ SDWLHQW DGKHUHQFH 7DNHQ WRJHWKHU WKH UHYLHZHG VWXGLHV VXJJHVW WKDW QHDUSHUIHFW DGKHUHQFH PD\ EH GLIILFXOW WR DFKLHYH DPRQJ FKLOGUHQ DQG DGROHVFHQWV RQ $57 UHJLPHQV ,I b DGKHUHQFH RU JUHDWHU LV QHFHVVDU\ WR DFKLHYH RSWLPDO WKHUDSHXWLF EHQHILW )DUOH\ HW DO f WKH DERYH UHVXOWV VXJJHVW WKDW D ODUJH SHUFHQWDJH RI FKLOGUHQ PD\ QRW EH DFKLHYLQJ PD[LPXP EHQHILW IURP $57 $OWKRXJK WKHVHV VWXGLHV DUH XVHIXO LQ HVWLPDWLQJ WKH SUHYDOHQFH RI SRRU DGKHUHQFH DGGLWLRQDO VWXGLHV ZLWK ODUJHU VDPSOH VL]HV DQG PXOWLSOH DVVHVVPHQW PRGDOLWLHV 4XLWWQHU (VSHODJH ,HYHUV/DQGLV t 'URWDU Ef DUH QHFHVVDU\ LQ RUGHU WR XQGHUVWDQG WKH H[WHQW DQG QDWXUH RI DGKHUHQFH SUREOHPV DPRQJ IDPLOLHV RI FKLOGUHQ RQ $-57 2QO\ ZLWK D V\VWHPLF XQGHUVWDQGLQJ RI WKH SUREOHP FDQ UHVHDUFKHUV EHJLQ WR GHYHORS VSHFLILF WDUJHWHG LQWHUYHQWLRQV WKDW DUH HIIHFWLYH LQ SURPRWLQJ DGKHUHQFH LQ WKLV SRSXODWLRQ %DUULHUV WR $57 $GKHUHQFH 8QGHUVWDQGLQJ EDUULHUV WR DGKHUHQFH PD\ EH HVVHQWLDO LQ XQGHUVWDQGLQJ WKH SUREOHP RI SRRU DGKHUHQFH 3ULRU WR GHYHORSLQJ LQWHUYHQWLRQV WR LQFUHDVH DGKHUHQFH DPRQJ +,9LQIHFWHG FKLOGUHQ UHVHDUFKHUV PXVW LGHQWLI\ DQG XQGHUVWDQG WKH REVWDFOHV WR DGKHUHQFH 7R GDWH IHZ VWXGLHV KDYH DWWHPSWHG WR LGHQWLI\ WKHVH REVWDFOHV 7KRXJK EDUULHUV DUH DVVHVVHG DV SDUW RI WKH 3HGLDWULF $,'6 &OLQLFDO 7ULDOV *URXS 3$&7*f SURWRFROV IHZ NQRZQ VWXGLHV KDYH H[DPLQHG EDUULHUV DPRQJ \RXQJHU +,9LQIHFWHG FKLOGUHQ DQG WKHLU IDPLOLHV )XUWKHUPRUH RQO\ RQH NQRZQ VWXG\ KDV H[DPLQHG EDUULHUV WR

PAGE 39

DGKHUHQFH DPRQJ +,9LQIHFWHG DGROHVFHQWV DQG RQO\ VHYHUDO VWXGLHV KDYH H[DPLQHG EDUULHUV WR DGKHUHQFH DPRQJ +,9LQIHFWHG DGXOWV 7KH 3$&7* DGKHUHQFH PRGXOHV UHTXLUH WKH QXUVH WR FRQGXFW D VWUXFWXUHG LQWHUYLHZ WR DVVHVV NQRZOHGJH RI WKH PHGLFDWLRQ QDPHV GRVDJH IUHTXHQF\ RI GRVLQJ DQG QXPEHU RI PLVVHG GRVHV LQ SUHYLRXV GD\V 0RGXOH f DQG WKHQ DVVHVV EDUULHUV E\ UHDGLQJ D OLVW RI SUHYLRXVO\ LGHQWLILHG EDUULHUV DQG WKHQ DVNLQJ FDUHJLYHUV f2YHU WKH ODVW WZR ZHHNV KDYH DQ\ RI WKH IROORZLQJ EHHQ SUREOHPV IRU \RX ZLWK GUXJ QDPHf"f 0RGXOH f :KHQ )DUOH\ DQG FROOHDJXHV f DGPLQLVWHUHG WKH PRGXOHV WR D JURXS RI FDUHJLYHUV RI FKLOGUHQ XQGHU DJH FDUHJLYHUV LGHQWLILHG SUREOHPV DGKHULQJ WR PHGLFDWLRQV 7KH PRVW FRPPRQO\ HQGRUVHG SUREOHPV ZHUH fGLGQfW UHILOO UDQ RXWf fVFKHGXOLQJ LQWHUIHUHV ZLWK OLIHVW\OHf fFKLOG UHIXVHVf fPXOWLSOH FDUHWDNHUVf fIRUJRWf DQG fWDVWHf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f SURYLGHG WHHQDJHUV ZLWK D LWHP OLVW RI UHDVRQV WKH\ PD\ KDYH PLVVHG WKHLU PHGLFDWLRQ DQG DVNHG WKH WHHQV WR UDWH WKH UHDVRQV RQ D /LNHUW VFDOH WKDW UDQJHG IURP fVWURQJO\ GLVDJUHHf WR fVWURQJO\ DJUHHf 7KH LWHPV LQ RUGHU IURP PRVW KLJKO\ HQGRUVHG WR OHDVW KLJKO\ HQGRUVHG DUH fWRR PDQ\ SLOOVf fUHPLQGV PH ,fP +,9SRVLWLYHf fVLGH HIIHFWVf fLQWHUIHUHV ZLWK VFKHGXOHf fIRUJRWf fGLVFORVXUH

PAGE 40

FRQFHUQVf DQG fGHSUHVVHGf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f SDWLHQWV ZKR DGPLWWHG PLVVLQJ GRVHV ZHUH DVNHG WR OLVW DV PDQ\ UHDVRQV DV SRVVLEOH IRU ZK\ WKH\ PLVVHG WKHLU PHGLFDWLRQ &RPPRQ EDUULHUV OLVWHG ZHUH IRUJHWWLQJ bf VOHHSLQJ WKURXJK GRVHV bf EHLQJ DZD\ IURP KRPH bf FKDQJLQJ WKH WKHUDS\ URXWLQH bf EHLQJ EXV\ bf EHLQJ VLFN bf H[SHULHQFLQJ VLGH HIIHFWV bf DQG IHHOLQJ GHSUHVVHG bf ,Q D VWXG\ RI +,9LQIHFWHG ZRPHQ ZKR ZHUH DVNHG WR fZULWH DERXW D IXWXUH LQ ZKLFK \RX RQO\ KDYH WR WDNH RQH SLOO D GD\f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

PAGE 41

6HYHUDO VWXGLHV KDYH FRQGXFWHG TXDQWLWDWLYH DVVHVVPHQWV RI EDUULHUV EDVHG RQ SUHYLRXV TXDOLWDWLYH VWXGLHV 0DQQKHLPHU DQG FROOHDJXHV f DVNHG RYHU PRVWO\PDOH +,9LQIHFWHG DGXOWV WR FRPSOHWH D LWHP PHGLFDWLRQ EDUULHUV FKHVFNOLVW )RUJHWWLQJ WR WDNH WKH PHGLFDWLRQ ZDV WKH PRVW KLJKO\ HQGRUVHG EDUULHU IROORZHG E\ EHLQJ DZD\ IURP KRPH H[SHULHQFLQJ VLGH HIIHFWV DQG KDYLQJ GLIILFXOW\ WDNLQJ WKH SLOOV DW VSHFLILHG WLPHV 7KH IUHTXHQF\ RI UHDVRQV ZDV VWDEOH RYHU PRQWKV $PPDVVDUL DQG RWKHUV f DGPLQLVWHUHG D LWHP EDUULHUV FKHVFNOLVW WR PRVWO\PDOH +,9LQIHFWHG DGXOWV 3DUWLFLSDQWV ZHUH DVNHG WR UDWH WKH LPSRUWDQFH RI HDFK EDUULHU LQ WKH PLVVLQJ RU GLVFRQWLQXLQJ RI GUXJV RQ D IRXUSRLQW /LNHUW VFDOH 7KH PRVW KLJKO\ HQGRUVHG EDUULHUV UHODWHG WR WKH QXPEHU RI SLOOV FRQFHUQ DERXW IXWXUH VLGH HIIHFWV EHLQJ DZD\ IURP KRPH DQG FRQFHUQ DERXW RWKHUV JDLQLQJ DZDUHQHVV RI WKHLU PHGLFDWLRQn WDNLQJ DQG GLVHDVH VWDWXV :DOVK DQG FROOHDJXHV f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f DVNHG DGXOWV RQ $57 WR FRPSOHWH D LWHP EDUULHUV FKHVFNOLVW EDVHG RQ WKH $LGV &OLQLFDO 7ULDOV *URXS $&7*f %DVHOLQH $GKHUHQFH 4XHVWLRQQDLUH 3DUWLFLSDQWV ZHUH DVNHG WR HQGRUVH DQ LWHP LI LW SUHYHQWHG WKHP IURP WDNLQJ WKHLU PHGLFDWLRQV DV SUHVFULEHG 7KH PRVW KLJKO\ HQGRUVHG LWHPV ZHUH fVOHSW WKURXJK WKH GRVH WLPHf fKDG SUREOHPV WDNLQJ SLOOV ZLWK

PAGE 42

VSHFLDO LQVWUXFWLRQVf fKDG D FKDQJH LQ GDLO\ URXWLQHf fGLG QRW KDYH PHGLFDWLRQV ZLWK \RXf fEXV\ DQG GLG QRW ZDQW WR VWRS WR WDNH PHGLFDWLRQVf fVLPSO\ IRUJRWf fIHOW GHSUHVVHG RU RYHUZKHOPHGf fIHOW DQJU\ GHSUHVVHG RU KRSHOHVV WKDW \RX KDYH WR GHDO ZLWK WKLVf fZDQWHG WR IRUJHW WKH ZKROH WKLQJf DQG fVLGH HIIHFWVf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f H[DPLQHG UHJLPHQVSHFLILF EDUULHUV WR DGKHUHQFH DPRQJ PRVWO\ PDOH +,9LQIHFWHG DGXOWV 7KH\ GHYHORSHG WKH %DUULHUV WR $GKHUHQFH &KHFNOLVW %$&f D LWHP TXHVWLRQQDLUH WKDW PHDVXUHV JHQHUDO EDUULHUV WR +,9 WUHDWPHQW DQG EDUULHUV VSHFLILF WR WKH PHGLFDWLRQ UHJLPHQ 3DWLHQWV UDWH WKH GHJUHH WR ZKLFK HDFK LWHP UHIOHFWV WKHLU RZQ H[SHULHQFHV 7KH LWHPV ZHUH FRQVLVWHQW ZLWK EDUULHUV LGHQWLILHG GXULQJ TXDOLWDWLYH DQDO\VLV RI SDWLHQWV RQ +$$57 %RJDUW HW DO f 7KH PRVW KLJKO\ HQGRUVHG EDUULHUV ZHUH fWUHDWPHQW UHPLQGV PH WKDW DP +,9f f, GR QRW ZDQW RWKHUV WR NQRZ WKDW DP +,9f fZKHQ KDYH FOLQLF DSSRLQWPHQWV IRUJHW WR DVN VRPH RI P\ TXHVWLRQV DERXW WUHDWPHQWf f, KDYH WURXEOH UHPHPEHULQJ QDPHV RI PHGLFLQHV DQG ZKDW WKH\ DUH IRUf DQG f, GR QRW OLNH WKH ZD\ P\ PHGLFDWLRQ PDNHV P\ ERG\ IHHOf

PAGE 43

3DWLHQWV ZKR UHSRUWHG DW OHDVW ZHHNO\ PLVVHG GRVHV UHSRUWHG D VLJQLILFDQWO\ KLJKHU QXPEHU RI EDUULHUV WKDQ WKRVH ZKR ZHUH PRUH DGKHUHQW 7KLV VWXG\ LV GLIIHUHQW IURP WKH RWKHU VWXGLHV DV HQGRUVHPHQW RI EDUULHUV ZDV QRW QHFHVVDULO\ EDVHG RQ WKH UHODWLRQVKLS EHWZHHQ LWHPV DQG PLVVHG GRVHV RQO\ RQ WKH H[WHQW WR ZKLFK LWHPV UHIOHFWHG SDUWLFLSDQWVf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f LQSXW UHJDUGLQJ WKHLU H[SHULHQFH RI WKH +,9VSHFLILF GLVHDVH DQG UHJLPHQ EHKDYLRUV DQG LQ LGHQWLI\LQJ WKRVH IDFWRUV WKDW PD\ REVWUXFW GLVHDVH PDQDJHPHQW )XUWKHUPRUH TXDOLWDWLYH DVVHVVPHQW KDV EHHQ LPSOLFDWHG E\ WKH :RUOG +HDOWK 2UJDQL]DWLRQ 2n5HLOO\ f DV DQ LPSRUWDQW PHWKRGRORJ\ ZLWKLQ +,9$,'6 UHVHDUFK fWR XQGHUVWDQG EHKDYLRUV LQ WKHLU FRQWH[W DQG WKHUHE\ WR LGHQWLI\ EDUULHUV DQG SRWHQWLDO IDFLOLWDWRUV RI EHKDYLRU FKDQJHf S f 7KXV TXDOLWDWLYH PHWKRGV DUH QHFHVVDU\ LQ RUGHU WR XQGHUVWDQG WKH

PAGE 44

SUREOHP PRUH WKRURXJKO\ DQG SURYLGH WKH LQIRUPDWLRQ QHHGHG WR HIIHFWLYHO\ DSSO\ TXDQWLWDWLYH PHWKRGV DQG VSHFLILF IRFXVHG LQWHUYHQWLRQV WR WKH SUREOHP

PAGE 45

29(59,(: $1' +<327+(6(6 %DVHG RQ WKH UHVHDUFK OLWHUDWXUH D QHFHVVDU\ VWHS WRZDUGV WKH SURYLVLRQ RI LQWHUYHQWLRQV WR SURPRWH DGKHUHQFH LQ WKLV SRSXODWLRQ ZRXOG EH DQ H[DPLQDWLRQ RI WKH H[WHQW DQG QDWXUH RI DGKHUHQFH DPRQJ IDPLOLHV RI FKLOGUHQ RQ $57 UHJLPHQV 7KH UHVHDUFK LQGLFDWHV WKDW PXOWLSOH DVVHVVPHQW PHWKRGV DUH QHHGHG LQ RUGHU WR IXOO\ FDSWXUH WKH H[WHQW DQG QDWXUH RI DGKHUHQFH 4XLWWQHU HW DO Ef )RU WKLV UHDVRQ WKLV VWXG\ XVHG D PRUH VXEMHFWLYH EXW VSHFLILF PHDVXUH VWUXFWXUHG LQWHUYLHZVf DORQJ ZLWK D PRUH REMHFWLYH EXW JOREDO PHDVXUH SKDUPDF\ UHILOO KLVWRU\f RI DGKHUHQFH 6LQFH SUHYLRXV VWXGLHV KDG QRW DVNHG IDPLOLHV WR LGHQWLI\ WKH H[WHQW DQG QDWXUH RI WKH EDUULHUV WKDW KLQGHU DGKHUHQFH WR SHGLDWULF $57 UHJLPHQV WKH IXOO H[WHQW DQG QDWXUH RI WKH EDUULHUV UHOHYDQW WR WKLV SRSXODWLRQ ZHUH XQNQRZQ 7KHUHIRUH LQ WKLV VWXG\ D TXDOLWDWLYH DVVHVVPHQW ZDV FRQGXFWHG WR FRPSUHKHQVLYHO\ LGHQWLI\ WKH EDUULHUV WR DGKHUHQFH H[SHULHQFHG E\ IDPLOLHV RI FKLOGUHQ RQ $57 %DVHG RQ WKH UHYLHZ RI WKH OLWHUDWXUH WKH VSHFLILF DLPV RI WKH FXUUHQW VWXG\ ZHUH 3ULPDU\ $LPV f 7R GRFXPHQW FDUHJLYHUVf NQRZOHGJH RI PHGLFDWLRQ QDPHV GRVDJH GRVLQJ IUHTXHQFLHV GRVLQJ LQWHUYDOV DQG PHGLFDWLRQVSHFLILF IRRG DQG GULQN LQWDNH JXLGHOLQHV f 7R GRFXPHQW UDWHV RI DGKHUHQFH WR SUHVFULEHG PHGLFDWLRQWDNLQJ IUHTXHQFLHV LQWHUYDOV DQG PHGLFDWLRQVSHFLILF IRRG DQG GULQN LQWDNH JXLGHOLQHV

PAGE 46

f7R GRFXPHQW FDUHJLYHUVf SHUFHLYHG EDUULHUV WR DGKHUHQFH WR +,9 PHGLFDWLRQ UHJLPHQV 6HFRQGDU\ $LPV f 7R GRFXPHQW WKH XWLOLW\ RI WKH $GKHUHQFH ,QWHUYLHZ+,9 $,+,9f IRU REWDLQLQJ LQIRUPDWLRQ DERXW FDUHJLYHU NQRZOHGJH DQG DGKHUHQFH f 7R GRFXPHQW WKH XWLOLW\ RI D PHGLFDWLRQ GLVSOD\ FDUG LQ KHOSLQJ FDUHJLYHUV LGHQWLI\ PHGLFDWLRQ QDPHV f 7R GRFXPHQW WKH XWLOLW\ RI WKH KRXU UHFDOO LQWHUYLHZ 5,f IRU DVVHVVLQJ DGKHUHQFH WR SHGLDWULF +,9 UHJLPHQV f 7R GRFXPHQW WKH UHODWLRQVKLS EHWZHHQ UHJLPHQ NQRZOHGJH DQG DGKHUHQFH WR SHGLDWULF +,9 UHJLPHQV f 7R GRFXPHQW WKH UHODWLRQVKLS EHWZHHQ GLVHDVH VHYHULW\ DQG DGKHUHQFH WR SHGLDWULF +,9 UHJLPHQV +\SRWKHVHV &DUHJLYHU NQRZOHGJH RI WKH SUHVFULEHG UHJLPHQ ZLOO GLIIHU VLJQLILFDQWO\ IURP UHSRUWHG W\SLFDO DGKHUHQFH EHKDYLRU $,+,9f DV VRPH FDUHJLYHUV ZLOO DFWLYHO\ GHFLGH QRW WR JLYH WKHLU FKLOGUHQ VRPH PHGLFDWLRQV WKDW PD\ KDYH SRRU WDVWH RU VLJQLILFDQW VLGH HIIHFWV &DUHJLYHU NQRZOHGJH RI PHGLFDWLRQ QDPHV ZLOO QRW GLIIHU VLJQLILFDQWO\ IRU WKRVH ZKR XVHG WKH PHGLFDWLRQ GLVSOD\ FDUG WR LGHQWLI\ WKH QDPHV YHUVXV WKRVH ZKR GLG QRW

PAGE 47

$OO DGKHUHQFH PHDVXUHV 5, IUHTXHQF\ 5, LQWHUYDO 5, GLHWDU\ UHILOO UDWLQJV DQG QXUVH UDWLQJVf ZLOO FRUUHODWH VLJQLILFDQWO\ DQG SRVLWLYHO\ ZLWK RQH DQRWKHU $GKHUHQFH IUHTXHQF\ PHDVXUHV 5, IUHTXHQF\ UHILOO UDWLQJVf ZLOO VLJQLILFDQWO\ DQG QHJDWLYHO\ UHODWH WR GLVHDVH VHYHULW\ &'& VHYHULW\ UDWLQJV DQG YLUDO ORDGf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

PAGE 48

0(7+2' 3DUWLFLSDQWV 6WXG\ SDUWLFLSDQWV ZHUH SULPDU\ FDUHJLYHUV RI +,9VHURSRVLWLYH FKLOGUHQ DJHV FRQWDFWHG DW SHGLDWULF +,9 VSHFLDOW\ FOLQLFV LQ *DLQHVYLOOH DQG -DFNVRQYLOOH )ORULGD DQG %DOWLPRUH 0DU\ODQG 6HYHQW\WKUHH SULPDU\ FDUHJLYHUV PHW HOLJLELOLW\ FULWHULD DQG ZHUH DVNHG WR SDUWLFLSDWH LQ WKH VWXG\ 6L[W\WKUHH SULPDU\ FDUHJLYHUV bf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f ZLWK IHZHU $IULFDQ$PHULFDQ SDUWLFLSDQWV DW WKH *DLQHVYLOOH VLWH )XOO GHVFULSWLYH VWDWLVWLFV RQ SDUWLFLSDQW GHPRJUDSKLF LQIRUPDWLRQ FDQ EH IRXQG LQ 7DEOHV DQG &KLOG SDUWLFLSDQWV ZHUH SULPDULO\ $IULFDQ$PHULFDQ bf PDOH bf DQG DZDUH RI WKHLU +,9 GLDJQRVLV bf 7KH DYHUDJH FKLOG SDUWLFLSDQW ZDV \HDUV ROG 6' f DQG KDG EHHQ SUHVFULEHG $57 IRU DQ DYHUDJH RI \HDUV 7KH PRGDO YLUDO ORDG ZDV OHVV WKDQ XQGHWHFWDEOHf &DUHJLYHU SDUWLFLSDQWV ZHUH SULPDULO\

PAGE 49

$IULFDQ$PHULFDQ bf IHPDOH bf DQG UHSRUWHGO\ +,9 XQLQIHFWHG bf 0RVW FDUHJLYHUV ZHUH DGRSWLYH SDUHQWV bf ELRORJLFDO PRWKHUV bf RU JUDQGSDUHQWV bf 3URFHGXUH $W HDFK VWXG\ VLWH WKH UHVHDUFK QXUVH ZDV DVNHG WR KHOS WKH VXEMHFW UHFUXLWHU FRPSOHWH WKH HOLJLELOLW\ FKHFNOLVW $SSHQGL[ $f IRU HDFK SDWLHQW 1H[W WKH UHFUXLWHU DVNHG HOLJLEOH SDUHQWVJXDUGLDQV VHYHUDO SUHOLPLQDU\ TXHVWLRQV $SSHQGL[ $ DOVRf WR HQVXUH HOLJLELOLW\ 0XOWLSOH HOLJLELOLW\ FULWHULD ZHUH HVWDEOLVKHG IRU WKH VWXG\ 7DEOH f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f $OO LQWHUYLHZHUV ZHUH &DXFDVLDQ 1H[W WKH LQWHUYLHZHU FRPSOHWHG WKH GHPRJUDSKLF TXHVWLRQQDLUH $SSHQGL[ &f DQG WKH WHOHSKRQH FRQWDFW IRUP $SSHQGL[ 'f ZLWK WKH FDUHJLYHU )LQDOO\ WKH LQWHUYLHZHU FRPSOHWHG WKH 0HGLFDO 5HFRUG 5HSRUW )RUP $SSHQGL[ (f )ROORZLQJ WKH FOLQLF YLVLW SDUHQWVFDUHJLYHUV ZHUH FRQWDFWHG IRU WHOHSKRQH IROORZXS ZLWK WKH 5, $SSHQGL[ )f %HJLQQLQJ WZR ZHHNV IROORZLQJ WKH FOLQLF YLVLW SDUHQWV ZHUH DVNHG WR FRPSOHWH WKH 5, WKUHH WLPHV ZLWKLQ D WZRZHHN SHULRG LQFOXGLQJ

PAGE 50

7DEOH 3DUWLFLSDQW 'HPRJUDSKLF ,QIRUPDWLRQ 9DULDEOH &KLOGUHQ Q f &DUHHLYHUV UL f *HQGHU )HPDOH b b 0DOH b b (WKQLFLW\ $IULFDQ$PHULFDQ b b &DXFDVLDQ b b +LVSDQLF b b 2WKHU b b +,9 6WDWXV ,QIHFWHG b b .QRZ +,9 6WDWXV b b &DUHJLYHU 5HODWLRQVKLS %LRORJLFDO 3DUHQW b %LRORJLFDO *UDQGSDUHQW f§ b $GRSWLYH 3DUHQW f§ b )RVWHU 3DUHQW f§ b 5HODWLYH f§ b )DPLO\ )ULHQG f§ b &DUHJLYHU 0DULWDO 6WDWXV 6LQJOH RU /LYLQJ ZLWK 3DUWQHU b 0DUULHG f§ b 6HSDUDWHG RU 'LYRUFHG f§ b :LGRZHG b &DUHJLYHU (GXFDWLRQ *UDGH 6FKRRO b 6RPH +LJK 6FKRRO f§ b *UDGXDWHG +LJK 6FKRRO RU (DUQHG *(' f§ b 7HFKQLFDO9RFDWLRQDO 7UDLQLQJ f§ b 6RPH &ROOHJH f§ b %DFKHORU 'HJUHH RU +LJKHU f§ b &DUHJLYHU (PSOR\PHQW 6WDWXV (PSOR\HG )XOOWLPH b (PSOR\HG 3DUWWLPH f§ b (PSOR\HG 6WXGHQW f§ b 8QHPSOR\HG f§ b 'LVDEOHG f§ b 5HWLUHG f§ b +RPHPDNHU f§ b

PAGE 51

7DEOH $GGLWLRQDO 3DUWLFLQDQW 'HPRJUDSKLF ,QIRUPDWLRQ 9DULDEOH 0 6' &KLOG $JH
PAGE 52

RQH ZHHNHQG GD\ 7KUHH PRQWKV IROORZLQJ WKH FOLQLF YLVLW SKDUPDFLHV ZHUH FRQWDFWHG WR REWDLQ SKDUPDF\ UHFDOO KLVWRULHV IRU WKH SULRU WKUHH PRQWKV 0HDVXUHV $GKHUHQFH ,QWHUYLHZ+,9 $,+,9 $SSHQGL[ %f 7KLV VWUXFWXUHG LQWHUYLHZ ZDV GHVLJQHG E\ WKH DXWKRU WR DVVHVV W\SLFDO DGKHUHQFH WR SUHVFULEHG UHJLPHQV XQGHUVWDQGLQJ DQG NQRZOHGJH RI WKH SUHVFULEHG UHJLPHQ DQG EDUULHUV WR DGKHULQJ WR HDFK DVSHFW RI WKH SHGLDWULF +,9 UHJLPHQ 7KH LQWHUYLHZ LV FRQGXFWHG ZLWK FDUHJLYHUV ,W ZDV DGDSWHG IURP WKH 7UHDWPHQW $GKHUHQFH 4XHVWLRQQDLUH ZKLFK LV XVHG WR DVVHVV DGKHUHQFH WR SHGLDWULF &) UHJLPHQV 4XLWWQHU HW DO Df 'XH WR WKH QRYHOW\ RI WKLV PHDVXUH UHOLDELOLW\ DQG YDOLGLW\ KDYH QRW EHHQ HVWDEOLVKHG 7KH LQWHUYLHZHU EHJLQV E\ DVNLQJ SDUHQWVFDUHJLYHUV WR LQIRUP WKH LQWHUYLHZHU RI ZKDW WKH IDPLO\ KDV W\SLFDOO\ GRQH WR WUHDW WKH FKLOGfV LOOQHVV LQ WKH ODVW WZR ZHHNV 'LIILFXOW\ PDQDJLQJ WKH FKLOGfV UHJLPHQ LV QRUPDOL]HG LQ RUGHU WR SURYLGH D VDIH QRQMXGJPHQWDO FRQWH[W LQ ZKLFK KRQHVW UHVSRQGLQJ LV HQFRXUDJHG 0XUSK\ t 'LOORQ f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

PAGE 53

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fV UHJLPHQ WKH\ DUH DVNHG WR LGHQWLI\ RWKHU WKLQJV QRW \HW PHQWLRQHG WKDW PD\ PDNH PHGLFDWLRQWDNLQJ KDUG RU PD\ NHHS WKH FKLOG IURP WDNLQJ KLVKHU PHGLFDWLRQ DV UHFRPPHQGHG ,QWHUYLHZV ZHUH FRQGXFWHG E\ WUDLQHG XQGHUJUDGXDWH DVVLVWDQWV DQG JUDGXDWH VWXGHQWV ,QWHUYLHZHUV XQGHUZHQW WUDLQLQJ ZLWK WKH LQWHUYLHZ SURFHGXUHV DQG HQJDJHG LQ SUDFWLFH LQWHUYLHZV 7KH\ ZHUH UHTXLUHG WR DFKLHYH b DJUHHPHQW ZLWK WKH DXWKRUfV FRGLQJ RI D SUDFWLFH LQWHUYLHZ EHIRUH FRQGXFWLQJ LQWHUYLHZV IRU WKH VWXG\ 'HPRJUDSKLF 4XHVWLRQQDLUH $SSHQGL[ &O 7KLV TXHVWLRQQDLUH ZDV GHVLJQHG WR JDWKHU GHPRJUDSKLF LQIRUPDWLRQ RQ WKH SDUHQW RU FDUHJLYHU FKLOG DQG IDPLO\ 7KH TXHVWLRQQDLUH DVNV WKH FDUHJLYHU IRU WKH GDWH RI ELUWK JHQGHU DQG HWKQLFLW\ RI ERWK WKH FKLOG DQG FDUHJLYHU 4XHVWLRQV DVVHVV WKH FDUHJLYHUfV PDULWDO VWDWXV HGXFDWLRQ ZRUN VWDWXV DQG KRXVHKROG LQFRPH DV ZHOO DV +,9 VWDWXV DQG PRGH RI WUDQVPLVVLRQ IRU WKRVH ZKR DUH +,9 SRVLWLYH 4XHVWLRQV DERXW WKH FKLOG LQFOXGH WKH DJH RI GLDJQRVLV PRGH RI WUDQVPLVVLRQ GDWH RI $57 LQLWLDWLRQ DQG ZKHWKHU RU QRW WKH

PAGE 54

FKLOG NQRZV WKDW KH RU VKH LV +,9 SRVLWLYH )LQDOO\ TXHVWLRQV DERXW WKH IDPLO\ LQFOXGH ZKR OLYHV LQ WKH KRPH DQG ZKR KHOSV WKH FKLOG WDNH KLVKHU PHGLFDWLRQ 7HOHSKRQH &RQWDFW )RUP $SSHQGL[ 'f 7KLV IRUP ZDV GHVLJQHG WR FROOHFW WKH QDPHV RI WKH FDUHJLYHU DQG FKLOG SULPDU\ DQG VHFRQGDU\ WHOHSKRQH QXPEHUV DQG EHVW WLPHV WR FDOO 7KH LQIRUPDWLRQ ZDV XVHG LQ FRQGXFWLQJ WKH WHOHSKRQH DGKHUHQFH LQWHUYLHZV 0HGLFDO 5HFRUG 5HSRUW )RUP $SSHQGL[ (f 7KLV UHVHDUFKHUFRPSOHWHG IRUP ZDV GHVLJQHG E\ WKH DXWKRU WR DVVHVV WKH FOLQLFDO LQGLFDWLRQV RI WKH FKLOGfV GLVHDVH VHYHULW\ &'& VWDWXV DQG &' FRXQWf DQG WKH FKLOGfV FXUUHQWO\ SUHVFULEHG PHGLFDWLRQV 7KH IRUP DOVR LQFOXGHV D QXUVH UDWLQJ RI WKH IDPLO\fV DGKHUHQFH WR WKH FKLOGfV UHJLPHQ RQ D SRLQW /LNHUW VFDOH fQRW DW DOO DGKHUHQWf fDOZD\V DGKHUHQWff QXUVHV DUH DVNHG WR UDWH DGKHUHQFH EDVHG RQ WKHLU FOLQLFDO H[SHULHQFHV ZLWK HDFK SDWLHQW GXULQJ WKH SULRU VL[ PRQWKV ,Q SDUW WKH IRUP ZDV PRGHOHG DIWHU WKH 3UHVFULEHG 7UHDWPHQW )RUP 37)f ZKLFK KDV EHHQ XVHG WR LGHQWLI\ SUHVFULEHG WUHDWPHQW IRU FKLOGUHQ ZLWK &) 4XLWWQHU HW DO Ff 7ZR QXUVHV DQG RQH SK\VLFLDQ IURP WKH SHGLDWULF +,9 FOLQLF DW WKH 8QLYHUVLW\ RI )ORULGD UHYLHZHG WKH IRUP LQ RUGHU WR HQVXUH DSSURSULDWHQHVV RI WKH ZRUGLQJ DQG FRQWHQW KRXU 5HFDOO ,QWHUYLHZ 5, $SSHQGL[ ), 7KLV PHDVXUH ZDV DGDSWHG IURP WKH 5, XVHG WR DVVHVV DGKHUHQFH WR GLDEHWHV UHJLPHQV )UHXQG -RKQVRQ 6LOYHUVWHLQ t 7KRPDV -RKQVRQ 6LOYHUVWHLQ 5RVHQEORRP &DUWHU t &XQQLQJKDP -RKQVRQ HW DK f $OWKRXJK WKH 5, IRU GLDEHWHV LQFOXGHV ERWK SDUHQW DQG FKLOG UHSRUW WKLV VWXG\ LQFOXGHV RQO\ SDUHQW UHSRUW EDVHG RQ WKH GLIILFXOWLHV FKLOGUHQ PD\ KDYH UHFDOOLQJ GHWDLOV DERXW PHGLFDWLRQWDNLQJ

PAGE 55

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f H[FHSW UHSRUWV UHJDUGLQJ PHGLFDWLRQVSHFLILF IRRG DQG OLTXLG LQWDNH ZKLFK ZHUH FRPSDUHG ZLWK LQVWUXFWLRQV RQ SXEOLVKHG KDQGRXWV WKDW OLVW DOO FXUUHQW $57 PHGLFDWLRQV $EERWW /DERUDWRULHV f 7KH UHVXOWLQJ YDULDEOHV UHSUHVHQW SHUFHQWV RI DJUHHPHQW EHWZHHQ WKH UHSRUWHG PHGLFDWLRQUHODWHG EHKDYLRUV DQG WKH SUHVFULEHG UHJLPHQV 7KH ILQDO PHDVXUH PRGLILFDWLRQ LQFOXGHG DQ DVVHVVPHQW RI EDUULHUV WR DGKHUHQFH $W WKH HQG RI WKH UHFDOO LQWHUYLHZ LQIRUPDQWV ZHUH WROG WKDW PDQDJLQJ D PHGLFDWLRQ VFKHGXOH LV GLIILFXOW DQG PDQ\ IDPLOLHV KDYH GLIILFXOWLHV IROORZLQJ WKH UHJLPHQ 7KH\ ZHUH DVNHG WR OLVW WKH WKLQJV WKDW PDGH LW GLIILFXOW WR IROORZ WKH SK\VLFLDQf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

PAGE 56

b DJUHHPHQW ZLWK WKH DXWKRUfV FRGLQJ RI D SUDFWLFH LQWHUYLHZ EHIRUH FRQGXFWLQJ LQWHUYLHZV IRU WKH VWXG\ 3KDUPDF\ 5HILOO +LVWRU\ 7KH YDOLGLW\ RI SKDUPDF\ UHILOOV KLVWRU\ KDV EHHQ VXSSRUWHG E\ SUHYLRXV VWXGLHV )DUOH\ HW DK f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f E\ WKH DFWXDO QXPEHU RI GD\V WR RQH UHILOO 9DOXHV JUHDWHU WKDQ RQH ZHUH VFRUHG DV b DGKHUHQW DV WKH IDPLOLHV KDG UHILOOHG WKHLU SUHVFULSWLRQV SULRU WR WKH GDWH RQ ZKLFK WKH PHGLFDWLRQV ZRXOG EH QHHGHG LI WKH IDPLO\ ZHUH b DGKHUHQW ,I QR UHILOO ZDV GHWHFWHG ZLWKLQ WKH PRQWK SHULRG WKH IDPLO\ ZDV VFRUHG DV b DGKHUHQW

PAGE 57

'$7$ $1$/<6(6 'DWD ZHUH DQDO\]HG XVLQJ WKH 6WDWLVWLFDO 3DFNDJH IRU 6RFLDO 6FLHQFHV 9HUVLRQ 6366 f )LUVW LQ RUGHU WR WHVW IRU GHPRJUDSKLF GLIIHUHQFHV DFURVV VLWHV &KL VTXDUH DQG $QDO\VHV RI 9DULDQFH $129$Vf ZHUH SHUIRUPHG 1H[W GHVFULSWLYH GDWD ZHUH UHYLHZHG DQG WKH VNHZQHVV DQG NXUWRVLV RI WKH RXWFRPH YDULDEOHV ZHUH H[DPLQHG LQ RUGHU WR GHWHUPLQH ZKHWKHU RU QRW QRUPDOLW\ DVVXPSWLRQV ZHUH PHW $GKHUHQFH IUHTXHQF\ GDWD ZHUH YLHZHG FRQWLQXRXVO\ DQG WKHQ VSOLW DW FXWRIIV RI b DQG b ZKLFK ZHUH EDVHG RQ WKH UHVHDUFK OLWHUDWXUH )DUOH\ HW DK 3DWHUVRQ f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b DQG b ZHUH ERWK LQFOXGHG LQ DQDO\VHV LQ RUGHU WR GHWHUPLQH WKH EHVW FXWRII DQG YLUDO ORDG ZDV VSOLW WR FRSLHVPO WKH KLJKHVW QXPEHU DVVRFLDWHG ZLWK XQGHWHFWDEOH OHYHOV RI YLUXV LQ WKH EORRG

PAGE 58

$ VHULHV RI DQDO\VHV ZDV WKHQ FRQGXFWHG WR WHVW WKH ILIWK VHW RI K\SRWKHVHV 'XH WR WKH QRQQRUPDOLW\ RI WKH NQRZOHGJH YDULDEOHV 7DEOH f NQRZOHGJH RI PHGLFDWLRQ QDPHV DQG GLHWDU\ UHTXLUHPHQWV ZHUH VSOLW EHWZHHQ b DQG b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f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

PAGE 59

6LJQHG 5DQNV 7HVW ZDV WKHQ XVHG WR WHVW IRU GLIIHUHQFHV LQ WKH QXPEHU RI EDUULHUV UHSRUWHG GXULQJ WKH $,+,9 DQG WKH 5,

PAGE 60

5(68/76 &DUHJLYHU .QRZOHGJH RI WKH 3UHVFULEHG 5HJLPHQ 'HVFULSWLYH GDWD UHJDUGLQJ FDUHJLYHU NQRZOHGJH FDQ EH IRXQG LQ 7DEOH 5HVXOWV RI WKH $,+,9 VKRZHG WKDW FDUHJLYHUVf SHUFHQW NQRZOHGJH RI SUHVFULEHG PHGLFDWLRQ QDPHV UDQJHG IURP b WR b ZLWK D PHDQ RI b 6' f 7KLUW\HLJKW SHUFHQW FRQVXOWHG WKHLU RZQ OLVWV RI WKH FKLOGfV PHGLFDWLRQ LQIRUPDWLRQ DQG b FRQVXOWHG WKH PHGLFDWLRQ GLVSOD\ FDUG ZKLOH RQO\ b UHVSRQGHG ZLWKRXW VRPH DLG 6L[W\VHYHQ SHUFHQW RI FDUHJLYHUV ZHUH DEOH WR FRUUHFWO\ LGHQWLI\ DOO RI WKHLU FKLOGUHQfV PHGLFDWLRQ QDPHV 2QO\ RQH FDUHJLYHU ZDV XQDEOH WR FRUUHFWO\ LGHQWLI\ WKH SUHVFULEHG PHGLFDWLRQn WDNLQJ IUHTXHQF\ IRU KHU FKLOGUHQfV NQRZQ PHGLFDWLRQV :KHQ WKH PHGLFDWLRQ QDPHV ZHUH NQRZQ WKH SHUFHQW RI PHGLFDWLRQ GRVDJHV DFFXUDWHO\ UHSRUWHG E\ FDUHJLYHUV UDQJHG IURP b WR b ZLWK D PHDQ RI b 6' f 6L[W\QLQH SHUFHQW RI FDUHJLYHUV FRUUHFWO\ LGHQWLILHG WKH GRVDJHV IRU DOO RI WKHLU FKLOGUHQfV $57 PHGLFDWLRQV $GGLWLRQDOO\ ZKHQ WKH PHGLFDWLRQ QDPHV ZHUH NQRZQ DQG ZKHQ WKH FKLOGUHQ ZHUH SUHVFULEHG PHGLFDWLRQV ZLWK VSHFLILF GLHWDU\ UHTXLUHPHQWV b RI FDUHJLYHUV FRUUHFWO\ LGHQWLILHG WKH VSHFLILF GLHWDU\ UHTXLUHPHQWV IRU WKHLU FKLOGUHQfV PHGLFDWLRQV &DUHJLYHUV ZHUH QRW DVNHG GLUHFWO\ DERXW WKHLU NQRZOHGJH RI WKH SUHVFULEHG GRVLQJ LQWHUYDO DV WKDW LQIRUPDWLRQ ZDV DVVXPHG WR EH LQKHUHQW LQ WKH TXHVWLRQ DERXW IUHTXHQF\ 1HYHUWKHOHVV FDUHJLYHUV ZHUH DVNHG WR VWDWH WKH WLPHV DW ZKLFK WKH FKLOGUHQ W\SLFDOO\ UHFHLYH HDFK PHGLFDWLRQ 7KH UHSRUWHG LQWHUYDO PD\ EH XVHIXO LQ XQGHUVWDQGLQJ

PAGE 61

FDUHJLYHUVf SHUFHSWLRQV RI WKH SUHVFULEHG LQWHUYDOV DV FDUHJLYHU UHSRUWV PD\ KDYH EHHQ KLJKO\ LQIOXHQFHG E\ VRFLDO GHVLUDELOLW\ )RU WKH SXUSRVH RI DQDO\VLV WKH GDWD ZHUH VRUWHG LQWR fLQWHUYDO EXFNHWVf LQGLFDWLQJ WKH DPRXQW RI GHYLDWLRQ IURP WKH SUHVFULEHG LQWHUYDO )RU H[DPSOH LI PHGLFDWLRQ ZDV WDNHQ OHVV WKDQ PLQXWHV EHIRUH RU DIWHU WKH SUHVFULEHG LQWHUYDO WKH LQWHUYDO ZDV VFRUHG DV ff LI WKH PHGLFDWLRQ ZDV WDNHQ RQH KRXU EHIRUH RU DIWHU WKH SUHVFULEHG LQWHUYDO KDG SDVVHG LW ZDV VFRUHG ff ,QWHUYDOV ZHUH QRW FRPSXWHG IRU RQFH GDLO\ GRVLQJ DV WKH LQWHUYLHZ SURYLGHG RQO\ RQH GDWD SRLQW IRU GRVLQJ WLPHV RI RQFH GDLO\ PHGLFDWLRQV 'HVFULSWLYH VWDWLVWLFV UHJDUGLQJ LQWHUYDO GHYLDQFH DUH SURYLGHG LQ 7DEOH 5HVXOWV VKRZHG WKDW WKH PDMRULW\ RI GRVHV ZHUH UHSRUWHGO\ JLYHQ ZLWKLQ WKH SUHVFULEHG LQWHUYDOV b RI WKH UHSRUWHG W\SLFDO GRVLQJ LQWHUYDOV ZHUH OHVV WKDQ KRXUV GHYLDQW IURP WKH SUHVFULEHG LQWHUYDOV b ZHUH OHVV WKDQ KRXUV GHYLDQW IURP WKH SUHVFULEHG LQWHUYDOV DQG b ZHUH OHVV WKDQ KRXUV GHYLDQW IURP WKH SUHVFULEHG LQWHUYDOV :KHQ RQO\ WKH LQWHUYDOV IRU WKULFH GDLO\ PHGLFDWLRQV ZHUH H[DPLQHG RQO\ b RI WKH UHSRUWHG W\SLFDO LQWHUYDOV ZHUH OHVV WKDQ KRXUV GHYLDQW b ZHUH OHVV WKDQ KRXUV GHYLDQW DQG b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b EHWZHHQ WKH WZR UHSRUWV 7KHUHIRUH W\SLFDO DGKHUHQFH EHKDYLRU GDWD IURP WKH $,+,9 ZHUH QRW LQFOXGHG LQ IXUWKHU DQDO\VHV

PAGE 62

7DEOH 'HVFULSWLYH 6WDWLVWLFV IRU 1RQ'HPRJUDSKLF 6WXG\ 9DULDEOHV 9DULDEOH 0 6' 0HGLDQ 0RGH 5DQJH 6NHZQHVV .XUWRVLV b &DUHJLYHU .QRZOHGJH RI 0HGLFDWLRQ 1DPHV Q f b &DUHJLYHU .QRZOHGJH RI 0HGLFDWLRQ 'RVLQJ Q f .QRZOHGJH RI ,QWHUYDO 'HYLDQFH Q f b 5, )UHTXHQF\ $GKHUHQFH Q f 5, ,QWHUYDO 'HYLDQFH Q f b 5, 'LHWDU\ $GKHUHQFH Q f b 5HILOO $GKHUHQFH Q f b 1XUVHUHSRUWHG $GKHUHQFH Q f 1RWH 'DWD IRU SHUFHQW FDUHJLYHU NQRZOHGJH RI GRVLQJ IUHTXHQF\ ZDV QRW LQFOXGHG LQ WKLV WDEOH DV DOO EXW RQH FDUHJLYHU KDG FRUUHFW GRVLQJ IUHTXHQF\ NQRZOHGJH 'DWD IRU SHUFHQW FDUHJLYHU NQRZOHGJH RI GLHWDU\ UHTXLUHPHQWV ZDV QRW LQFOXGHG LQ WKLV WDEOH EHFDXVH WKH YDULDEOH LV GLFKRWRPRXV D6FDOH UDQJHG IURP fQRW DW DOO DGKHUHQWf WR fDOZD\V DGKHUHQWf

PAGE 63

7DEOH 7\SLFDO 3HUFHQW RI 'HYLDQFH IURP 3UHVFULEHG ,QWHUYDOV 'HYLDQFH $PRXQW b RI WZLFH GDLO\ GRVHV b RI WKULFH GDLO\ GRVHV b RI DOO GRVHV Q f Q f Q f /HVV WKDQ KRXU KRXUV KRXUV KRXUV KRXUV

PAGE 64

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b WR b 0 6' f DQG WKH PRGDO DGKHUHQFH WR IUHTXHQF\ ZDV b :KHQ WKH VDPSOH ZDV VSOLW EDVHG RQ D 5, IUHTXHQF\ DGKHUHQFH FXWRII RI b b RI SDUWLFLSDQWV ZHUH FRQVLGHUHG DGKHUHQW :KHQ D FXWRII RI b ZDV XVHG b ZHUH FRQVLGHUHG DGKHUHQW )RU WKRVH GRVHV UHFHLYHG DYHUDJH GHYLDWLRQ IURP WKH SUHVFULEHG GRVLQJ LQWHUYDOV UDQJHG IURP WR KRXUV ZLWK D PHDQ RI KRXU D PHGLDQ RI KRXUV 6' f DQG D PRGH RI ]HUR KRXUV 5HVXOWV VKRZHG WKDW b RI WKH GRVHV JLYHQ ZHUH GHYLDQW IURP SUHVFULEHG LQWHUYDO E\ DW OHDVW RQH KRXU DQG b ZHUH GHYLDQW E\ DW OHDVW KRXUV $OVR IRU WKRVH GRVHV UHFHLYHG DQG IRU WKRVH PHGLFDWLRQV ZLWK VSHFLILF GLHWDU\ UHTXLUHPHQWV WKH DYHUDJH SHUFHQW DGKHUHQFH WR GLHWDU\ UHTXLUHPHQWV UDQJHG IURP b WR b ZLWK D PHDQ RI b 6' f PHGLDQ RI b DQG PRGH RI b 7KHVH UHVXOWV YDULHG IURP WKH UHVXOWV RI WKH SKDUPDF\ UHILOO KLVWRULHV ZKLFK VKRZHG WKDW DGKHUHQFH WR SKDUPDF\ UHILOOV UDQJHG IURP b WR b ZLWK D PHDQ RI b 6' f PHGLDQ RI b DQG PRGH RI b :KHQ WKH VDPSOH ZDV VSOLW EDVHG RQ D SKDUPDF\

PAGE 65

UHILOO DGKHUHQFH FXWRII RI b b RI SDUWLFLSDQWV ZHUH FRQVLGHUHG DGKHUHQW :KHQ D FXWRII RI b ZDV XVHG b ZHUH FRQVLGHUHG DGKHUHQW 1XUVHV UDWHG HDFK FKLOGfV DGKHUHQFH GUDZLQJ IURP LQIRUPDWLRQ LQ WKH PHGLFDO UHFRUG DQG LQIRUPDWLRQ REWDLQHG GXULQJ LQWHUDFWLRQV ZLWK WKH FKLOGUHQ DQG WKHLU FDUHJLYHUV 7KH PHDQ QXUVH UHSRUW UDWLQJ ZDV 6' f RQ D SRLQW /LNHUW VFDOH 7KH PHGLDQ UDWLQJ ZDV DQG WKH PRGDO UDWLQJ ZDV 1R FKLOGUHQ UHFHLYHG D UDWLQJ RI f1RW DW DOO $GKHUHQWf 5HODWLRQVKLS EHWZHHQ $GKHUHQFH 0HDVXUHV &RUUHODWLRQV ZHUH SHUIRUPHG WR GHWHUPLQH UHODWLRQVKLSV EHWZHHQ DGKHUHQFH PHDVXUHV 7DEOH f 5, IUHTXHQF\ ZDV QRW LQFOXGHG LQ WKH FRUUHODWLRQ PDWUL[ GXH WR NXUWRVLV 5, LQWHUYDO DQG GLHWDU\ DGKHUHQFH UDWLQJV GLG QRW VLJQLILFDQWO\ FRUUHODWH ZLWK SKDUPDF\ UHILOO DGKHUHQFH UDWLQJV RU QXUVHUHSRUWHG DGKHUHQFH OHYHOV +RZHYHU SKDUPDF\ UHILOO DGKHUHQFH UDWLQJV VLJQLILFDQWO\ DQG SRVLWLYHO\ FRUUHODWHG ZLWK QXUVH UHSRUWHG DGKHUHQFH OHYHOV U S f VXFK WKDW KLJKHU SKDUPDF\ UHILOO UDWLQJV UHODWHG WR KLJKHU QXUVH UHSRUW UDWLQJV RI DGKHUHQFH 5HODWLRQVKLS EHWZHHQ $GKHUHQFH DQG 'LVHDVH 6HYHULW\ &KL 6TXDUH DQDO\VHV ZHUH SHUIRUPHG LQ RUGHU WR WHVW IRU UHODWLRQVKLSV EHWZHHQ DGKHUHQFH PHDVXUHV DQG GLVHDVH VHYHULW\ PHDVXUHV YLUDO ORDG &'& OHWWHU RU &'& QXPEHUf :KHQ SDUWLFLSDQWV ZHUH FDWHJRUL]HG DV fDGKHUHQW b DGKHUHQFH UDWLQJff DQG fQRQDGKHUHQW b DGKHUHQFH UDWLQJff QRQH RI WKH GLVHDVH VHYHULW\ PHDVXUHV ZDV VLJQLILFDQWO\ UHODWHG ZLWK 5, IUHTXHQF\ DGKHUHQFH RU UHILOO DGKHUHQFH UDWLQJV :KHQ D PRUH VWULQJHQW FULWHULRQ b DGKHUHQFHf ZDV XVHG 5, IUHTXHQF\ ZDV VLJQLILFDQWO\ DQG SRVLWLYHO\ UHODWHG WR &'& QXPEHU ; S f VXFK WKDW WKRVH ZLWK KLJKHU 5,

PAGE 66

IUHTXHQF\ DGKHUHQFH ZHUH PRUH OLNHO\ WR EH FODVVLILHG DV DW RQH WLPH KDYLQJ VHYHUH &' 7 /\PSKRF\WH VXSSUHVVLRQ ORZ ZKLWH EORRG FHOO FRXQW RU SRRU LPPXQH IXQFWLRQLQJf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b RU b EXW ZDV VLJQLILFDQWO\ DQG SRVLWLYHO\ UHODWHG WR UHILOO KLVWRU\ ZKHQ D b DGKHUHQFH FXWRII ZDV XVHG ; S f 5HVXOWV VKRZHG WKDW FDUHJLYHUV ZLWK b UHILOO DGKHUHQFH ZHUH PRUH OLNHO\ WR NQRZ WKH QDPHV RI WKHLU FKLOGfV PHGLFDWLRQ 1H[W LW ZDV K\SRWKHVL]HG WKDW NQRZOHGJH RI GRVLQJ LQWHUYDO ZRXOG VLJQLILFDQWO\ FRUUHODWH ZLWK 5, GRVLQJ LQWHUYDO 7KLV K\SRWKHVLV ZDV VXSSRUWHG U f VXFK WKDW DV NQRZOHGJH RI GRVLQJ LQWHUYDO GHYLDWHG IURP WKH SUHVFULEHG GRVLQJ LQWHUYDO VR GLG 5, GRVLQJ LQWHUYDO )LQDOO\ LW ZDV DOVR K\SRWKHVL]HG WKDW 5, GLHWDU\ DGKHUHQFH ZRXOG SRVLWLYHO\ UHODWH WR NQRZOHGJH RI GLHWDU\ UHTXLUHPHQWV KRZHYHU WKH UHVXOWV RI &KLVTXDUH DQDO\VHV ZHUH QRQVLJQLILFDQW 5HODWLRQVKLS EHWZHHQ $GKHUHQFH DQG 3UHVFULEHG 'RVLQJ )UHTXHQF\ 'XH WR WKH QRQSDUDPHWULF QDWXUH RI 5, IUHTXHQF\ .UXVNDO:DOOLV DQDO\VLV ZDV HPSOR\HG WR GHWHUPLQH WKH UHODWLRQVKLS EHWZHHQ 5, IUHTXHQF\ DGKHUHQFH DQG

PAGE 67

7DEOH &RUUHODWLRQV %HWZHHQ $GKHUHQFH 0HDVXUHV 5, LQWHUYDO 5, 'LHWDU\ 3KDUPDF\ 5HILOO 5, ,QWHUYDO f§ f§ 5, 'LHWDU\ Q f f§ f§ 3KDUPDF\ 5HILOO Q f Q f f§ 1XUVHUHSRUWHG Q f Q f r Q f 1RWH r J

PAGE 68

SUHVFULEHG GRVLQJ IUHTXHQF\ 7KH UHVXOW ZDV QRQVLJQLILFDQW VXJJHVWLQJ WKDW 5, IUHTXHQF\ DGKHUHQFH GLG QRW GLIIHU EDVHG RQ SUHVFULEHG GRVLQJ IUHTXHQF\ 1H[W XQLYDULDWH $QDO\VLV RI 9DULDQFH $129$f ZDV FRQGXFWHG LQ RUGHU WR GHWHUPLQH WKH UHODWLRQVKLS EHWZHHQ UHILOO DGKHUHQFH DQG SUHVFULEHG GRVLQJ IUHTXHQF\ 5HVXOWV VKRZHG WKDW SUHVFULEHG GRVLQJ IUHTXHQF\ ZDV VLJQLILFDQWO\ DQG SRVLWLYHO\ UHODWHG WR UHILOO DGKHUHQFH ) f S 3RVWKRF DQDO\VHV XVLQJ WKH 6FKHIIH WHVW VKRZHG WKDW SDUWLFLSDQWV WDNLQJ PHGLFDWLRQV ZLWK WZLFH GDLO\ GRVLQJ KDG KLJKHU UHILOO UDWHV WKDQ SDUWLFLSDQWV WDNLQJ PHGLFDWLRQV ZLWK RQFH GDLO\ GRVLQJ 5HODWLRQVKLS EHWZHHQ $GKHUHQFH DQG 0HGLFDWLRQ )RUPXODWLRQ 0DQQ:KLWQH\ 8 DQDO\VLV ZDV HPSOR\HG WR GHWHUPLQH WKH UHODWLRQVKLS EHWZHHQ 5, IUHTXHQF\ DGKHUHQFH DQG PHGLFDWLRQ IRUPXODWLRQ 5HVXOWV VKRZHG WKDW IRUPXODWLRQ ZDV VLJQLILFDQWO\ UHODWHG WR 5, IUHTXHQF\ DGKHUHQFH S f VXFK WKDW PHGLFDWLRQV LQ SLOO IRUP ZHUH VLJQLILFDQWO\ PRUH OLNHO\ WR KDYH KLJK 5, IUHTXHQF\ DGKHUHQFH 7KHVH UHVXOWV FRQWUDVW ZLWK UHVXOWV RI 8QLYDULDWH $QDO\VLV RI 9DULDQFH $129$f ZKLFK VKRZHG WKDW IRUPXODWLRQ ZDV VLJQLILFDQWO\ DQG SRVLWLYHO\ UHODWHG WR UHILOO DGKHUHQFH ) f S ([DPLQDWLRQ RI WKH PHDQV LQGLFDWHV WKDW PHGLFDWLRQV LQ OLTXLG RU SRZGHU IRUPXODWLRQ ZHUH VLJQLILFDQWO\ PRUH OLNHO\ WR KDYH KLJK UHILOO UDWHV WKDQ PHGLFDWLRQV LQ SLOO IRUP %DUULHUV WR $GKHUHQFH 'XULQJ WKH $,+,9 FDUHJLYHUV UHSRUWHG DQ DYHUDJH RI EDUULHUV WR WKHLU FKLOGUHQfV DGKHUHQFH 6' f ZLWK D PHGLDQ RI RQH EDUULHU DQG D PRGH RI ]HUR EDUULHUV %DUULHUV ZHUH UHSRUWHG YHUEDWLP DQG WKHQ FDWHJRUL]HG )RU D FRPSOHWH OLVW RI EDUULHUV UHSRUWHG UHIHU WR 7DEOH

PAGE 69

:KHQ FDUHJLYHUV UHSRUWHG EDUULHUV WKH\ ZHUH DVNHG DERXW WKH IUHTXHQF\ RI WKH EDUULHU RFFXUUHQFH DQG VWUHQJWK RI WKH EDUULHU LQ SUHYHQWLQJ PHGLFDWLRQWDNLQJ HJ f+RZ RIWHQ GRHV WKLV KDSSHQ"f DQG f+RZ RIWHQ GRHV WKLV SUREOHP NHHS WKH FKLOG IURP JHWWLQJ WKH PHGLFLQH"ff 7KH\ ZHUH DVNHG WR DQVZHU HDFK TXHVWLRQ XVLQJ D SRLQW /LNHUW VFDOH ZLWK LQGLFDWLQJ fQRW RIWHQ DW DOOf DQG LQGLFDWLQJ fDOZD\Vf ,Q UHVSRQVH WR WKH ILUVW TXHVWLRQ WKH PHDQ IUHTXHQF\ RI RFFXUUHQFH UDWLQJ ZDV 6' f ZLWK D PHGLDQ RI DQG D PRGH RI ,Q UHVSRQVH WR WKH VHFRQG TXHVWLRQ WKH PHDQ VWUHQJWK UDWLQJ ZDV ZLWK ERWK D PHGLDQ DQG PRGH RI RQH 6' f %DUULHUV 5HSRUWHG ZLWK WKH $,+,9 9HUVXV WKH 5, 7KH :LOFR[RQ 6LJQHG 5DQNV 7HVW ZDV FRQGXFWHG WR GHWHUPLQH ZKHWKHU DQ\ VLJQLILFDQW GLIIHUHQFHV H[LVWHG LQ WKH QXPEHU RI EDUULHUV UHSRUWHG GXULQJ WKH $,+,9 YHUVXV WKH 5, 5HVXOWV VKRZHG WKDW WKH QXPEHU RI EDUULHUV UHSRUWHG GXULQJ WKH $,+,9 ZDV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH QXPEHU UHSRUWHG GXULQJ WKH 5, &DUHJLYHUV FXPXODWLYHO\ PDGH EDUULHU UHSRUWV GXULQJ WKH $,+,9 ZKLOH WKH\ PDGH FXPXODWLYH EDUULHU UHSRUWV GXULQJ WKH WKUHH UHFDOO LQWHUYLHZV

PAGE 70

7DEOH 5HSRUWHG %DUULHUV IURP $,+,9 DQG 5, %DUULHU 7RWDO RI &DUHJLYHUV :KR 5HSRUWHG %DUULHU rf 0HDQ )UHTXHQF\ 0HDQ 6WUHQJWK 3LOOV DUH WRR ELJ WR VZDOORZ f 7DVWHV %DG f &KLOG DYRLGV UHVLVWV SURORQJV RU UHIXVHV WKH PHGLFDWLRQ f 6OHHSLQJ WKURXJK GRVLQJ WLPH PHGLFLQH LQWHUIHUHV ZLWK VOHHS VFKHGXOH f &KLOG RU IDPLO\ IRUJHWV f 7RR PDQ\ SLOOV WR WDNH f 1HHGV SDUHQWDO VXSHUYLVLRQUHPLQGLQJ f 6LGH HIIHFWV f 0HGLFDWLRQWDNLQJ LQWHUIHUHV ZLWK GDLO\ DFWLYLWLHV f f§ f§ 'LIILFXOW\ ZLWK WKH PHGLFLQH ZKHQ VLFN f &DQfW NHHS PHGLFLQH GRZQ f f§ f§ :RUU\ WKDW RWKHUV ZLOO ILQG RXW DERXW FKLOGfV +,9 VWDWXV f &KLOG GRHVQfW VZDOORZ SLOOV f &KLOG KLGHV WKH PHGLFLQH f KDYH GLIILFXOW\ WUDYHOLQJ WR JHW WKH UHILOOV f KDYH GLIILFXOW\ JHWWLQJ WKH UHILOO QR UHDVRQ VWDWHGf f ,W LV WRR KDUG WR NHHS XS ZLWK WKH PHGLFLQH f f§ f§

PAGE 71

5DQ RXW f f§ f§ 3KDUPDF\ GRHVQfW DOZD\V KDYH PHGLFLQH LQ VWRFN f f§ f§ 'LIILFXOW\ UHPHPEHULQJ WR WDNH SLOOV DORQJ ZKHQ JRLQJ RXW f f§ f§ 2XW RI URXWLQH f f§ f§ 'RHVQfW GULQN DOO RI WKH PHGLFLQH f f§ f§ 7RWDO QXPEHU RI EDUULHU UHSRUWV f f§ f§ 1RWH 1XPEHUV LQ SDUHQWKHVHV UHSUHVHQW WKH QXPEHU RI EDUULHUV UHSRUWHG GXULQJ 5, %DUULHUV SURYLGHG GXULQJ WKH 5, ZHUH QRW UDWHG DFFRUGLQJ WR IUHTXHQF\ DQG VWUHQJWK

PAGE 72

',6&866,21 &DUHJLYHU .QRZOHGJH RI WKH 3UHVFULEHG 5HJLPHQ 7KH ILUVW JRDO RI WKH FXUUHQW VWXG\ ZDV WR GRFXPHQW FDUHJLYHUVf NQRZOHGJH RI PHGLFDWLRQ QDPHV GRVDJH GRVLQJ IUHTXHQFLHV GRVLQJ LQWHUYDOV DQG PHGLFDWLRQVSHFLILF GLHWDU\ UHTXLUHPHQWV $OWKRXJK DOO EXW RQH FDUHJLYHU FRUUHFWO\ LGHQWLILHG WKH SUHVFULEHG PHGLFDWLRQWDNLQJ IUHTXHQF\ RI WKHLU FKLOGfV PHGLFDWLRQ b RI FDUHJLYHUV ZHUH XQDEOH WR FRUUHFWO\ LGHQWLI\ DW OHDVW RQH RI WKHLU FKLOGfV PHGLFDWLRQ QDPHV b RI FDUHJLYHUV IDLOHG WR FRUUHFWO\ LGHQWLI\ WKH GRVDJH IRU DW OHDVW RQH PHGLFDWLRQ DQG KDOI RI WKH FDUHJLYHUV IDLOHG WR FRUUHFWO\ LGHQWLI\ WKH VSHFLILF GLHWDU\ UHTXLUHPHQWV IRU DW OHDVW RQH PHGLFDWLRQ 7KHVH UHVXOWV VXJJHVW WKDW FDUHJLYHUV RI FKLOGUHQ RQ $57 PD\ KDYH PRUH GLIILFXOW\ UHPHPEHULQJ RU LGHQWLI\LQJ PHGLFDWLRQ QDPHV DQG GLHWDU\ UHTXLUHPHQWV DQG OHVV GLIILFXOW\ UHPHPEHULQJ RU LGHQWLI\LQJ PHGLFDWLRQWDNLQJ IUHTXHQFLHV WKDQ DGXOWV RQ $57 %DQJVEHUJ %URQVWRQH t +RIPDQ 6WRQH HW DO f &DUHJLYHUV PD\ KDYH PRUH GLIILFXOW\ UHPHPEHULQJ RU LGHQWLI\LQJ VSHFLILF LQIRUPDWLRQ DERXW PHGLFDWLRQV LQ WKHLU FKLOGfV UHJLPHQ EHFDXVH WKH\ DUH OHVV LQYROYHG LQ WKHLU FKLOGfV UHJLPHQ EHKDYLRUV WKDQ DGXOWV DUH LQ WKHLU RZQ UHJLPHQ EHKDYLRUV ,Q WKH IXWXUH LW PD\ EH ZRUWKZKLOH WR H[DPLQH WKH UHODWLRQVKLS EHWZHHQ FDUHJLYHU NQRZOHGJH DQG FDUHJLYHU LQYROYHPHQW :KLOH DOPRVW DOO FDUHJLYHUV H[KLELWHG DFFXUDWH NQRZOHGJH RI GRVLQJ IUHTXHQFLHV UHVXOWV LQGLFDWH WKDW VRPH FDUHJLYHUV PD\ EH XQDZDUH RI WKH H[DFW LQWHUYDOV DW ZKLFK WKH GRVHV DUH WR EH JLYHQ :KHQ FDUHJLYHUV ZHUH DVNHG WR VWDWH WKH WLPHV RI GD\ DW ZKLFK WKH FKLOGUHQ KDG W\SLFDOO\ WDNHQ WKHLU PHGLFLQH GXULQJ WKH SUHYLRXV WZR ZHHNV b RI WKH

PAGE 73

WRWDO UHSRUWHG W\SLFDO LQWHUYDOV ZHUH DW OHDVW KRXUV GHYLDQW IURP WKH SUHVFULEHG LQWHUYDOV 5HSRUWHG LQWHUYDOV ZHUH SDUWLFXODUO\ SUREOHPDWLF IRU WKULFH GDLO\ GRVLQJ LQ ZKLFK b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b DGKHUHQFH ZDV IRXQG GHVSLWH 0(06 GDWD WKDW VXJJHVWV SRRUHU DGKHUHQFH )DUOH\ HW DO f KRZHYHU WKHVH GDWD VXJJHVW WKDW FDUHJLYHU RYHUHVWLPDWLRQ RI WKHLU FKLOGfV $57 DGKHUHQFH PD\ EH PRUH H[WHQVLYH WKDQ FDUHJLYHU DQG FKLOG RYHUHVWLPDWLRQ RI DGKHUHQFH WR RWKHU FKURQLF LOOQHVV UHJLPHQV 'HODPDWHU )HVWD HW DK 3DVVHUR HW DK 4XLWWQHU HW DK F 5DSRII /LQGVOH\ t &KULVWRSKHUVRQ 6FKRQL HW DO f 3HUKDSV WKH SHGLDWULF +,9 FOLQLF HQYLURQPHQWV IRVWHU UHSRUWLQJ RI VRFLDOO\ GHVLUDEOH EHKDYLRU PRUH VR WKDQ WKH GLDEHWHV RU

PAGE 74

&) FOLQLF HYLURQPHQWV 3URYLGHUV LQ +,9 FOLQLFV W\SLFDOO\ H[SHFW KLJK OHYHOV RI DGKHUHQFH b )DUOH\ HW DO 3DWHUVRQ HW DK f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f LGHQWLILFDWLRQ RI PHGLFDWLRQV GXULQJ WKH $,+,9 &DUHJLYHUV ZHUH JLYHQ WKH RSWLRQ WR XVH WKH GLVSOD\ FDUG LI WKH\ KDG GLIILFXOW\ JHQHUDWLQJ PHGLFDWLRQ QDPHV DQG b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n WDNLQJ IUHTXHQFLHV LQWHUYDOV DQG PHGLFDWLRQVSHFLILF GLHWDU\ UHTXLUHPHQWV 5HJDUGLQJ PHGLFDWLRQWDNLQJ IUHTXHQFLHV UHVXOWV YDU\ JUHDWO\ EDVHG RQ WKH PRGH RI DVVHVVPHQW ZLWK KLJKHU DGKHUHQFH UDWLQJV UHVXOWLQJ IURP WKH 5, WKDQ SKDUPDF\ UHILOO DQG JHQHUDOO\

PAGE 75

KLJK DGKHUHQFH UDWLQJV UHVXOWLQJ IURP QXUVHUHSRUW 7KH UDWH IURP WKH 5, b DYHUDJH DGKHUHQFH WR IUHTXHQF\f VXJJHVWV ORZHU DGKHUHQFH WKDQ FDUHJLYHU VHOIUHSRUWHG DGKHUHQFH LQ )DUOH\ HW DO f ZKLFK IRXQG WKDW DOO FDUHJLYHUV GHQLHG WKDW WKHLU FKLOG KDG PLVVHG DQ\ GRVHV ZLWKLQ WKH SUHYLRXV GD\V KRZHYHU WKH SUHVHQW UHILOO UDWHV b DYHUDJH DGKHUHQFHf VXJJHVW KLJKHU DGKHUHQFH WKDQ WKH )DUOH\ HW DO f UHVXOWV RI SKDUPDF\ UHILOO KLVWRULHV DQG 0(06 PRQLWRULQJ ZKLFK IRXQG DYHUDJH DGKHUHQFH UDWLQJV RI b DQG b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f $V PHQWLRQHG SUHYLRXVO\ WKH H[SHFWDWLRQV RI KLJK DGKHUHQFH DQG H[WHQVLYH IRFXV RQ DGKHUHQFH LQ SHGLDWULF +,9 FOLQLFV PD\ LQKLELW DFFXUDWH VHOIUHSRUW 7KH 5, DOVR SURYLGHG GDWD DERXW GRVLQJ LQWHUYDOV DQG GLHWDU\ DGKHUHQFH 'DWD VKRZHG WKDW D VLJQLILFDQW SRUWLRQ RI WKH VDPSOH bf KDG DQ DYHUDJH GRVLQJ GHYLDWLRQ RI KRXUV IURP WKH SUHVFULEHG LQWHUYDO ZKLOH D PXFK ODUJHU SRUWLRQ bf KDG DQ DYHUDJH GHYLDWLRQ RI KRXU 7KHVH UHVXOWV DUH FRQVLVWHQW ZLWK WKH W\SLFDO LQWHUYDOV UHSRUWHG GXULQJ WKH $,+,9 LQ WKLV VWXG\ 7KH WKLUG VWXG\ K\SRWKHVLV ZDV WKDW DOO DGKHUHQFH PHDVXUHV ZRXOG FRUUHODWH VLJQLILFDQWO\ DQG SRVLWLYHO\ 7KH UHVXOWV RI WKH VWXG\ IDLO WR VXSSRUW WKLV K\SRWKHVLV DV RQO\ SKDUPDF\ UHILOO DQG QXUVHUHSRUW UDWLQJV ZHUH VLJQLILFDQWO\ DQG SRVLWLYHO\ FRUUHODWHG 'HVSLWH WKH K\SRWKHVLV LW LV VRPHZKDW LQWXLWLYH WKDW 5, LQWHUYDO DQG GLHWDU\ DGKHUHQFH

PAGE 76

ZRXOG QRW VWURQJO\ UHODWH WR PHDVXUHV RI DGKHUHQFH WR IUHTXHQF\ VLQFH LW LV RQH WKLQJ WR H[DPLQH ZKHWKHU RU QRW WKH GRVH ZDV HYHU UHFHLYHG DQG DQRWKHU WKLQJ WR H[DPLQH ZKHWKHU RU QRW WKH GRVH ZDV UHFHLYHG ZLWKLQ FHUWDLQ FRQVWUDLQWV )XUWKHUPRUH WKH ODFN RI VLJQLILFDQW UHODWLRQVKLSV EHWZHHQ DGKHUHQFH PHDVXUHV LV FRQVLVWHQW ZLWK WKH UHVXOWV RI DQRWKHU VWXG\ WKDW FRPSDUHG FDUHJLYHU VHOIUHSRUW RI +$$57 DGKHUHQFH WR SKDUPDF\ DQG 0(06 GDWD )DUOH\ HW DO f DQG D VWXG\ WKDW FRPSDUHG VHOIUHSRUW $57 DGKHUHQFH UDWHV WR SKDUPDF\ DQG 0(06 GDWD DPRQJ DGXOWV )ULFN *DO /DQH t 6HZHOO f 7KHVH UHVXOWV DUH DOVR VLPLODU WR WKH UHVXOWV RI VWXGLHV RI UHODWLRQVKLSV EHWZHHQ GLIIHUHQW DGKHUHQFH DVVHVVPHQW PHWKRGV DPRQJ FDUHJLYHUV RI FKLOGUHQ ZLWK RWKHU FKURQLF LOOQHVVHV *ODVFRZ HW DO -RKQVRQ HW DO f )LQDOO\ WKH VLJQLILFDQW DQG SRVLWLYH UHODWLRQVKLS EHWZHHQ SKDUPDF\ UHILOO GDWD DQG QXUVHUHSRUW UDWLQJV LV XQVXUSULVLQJ DV LW LV OLNHO\ WKDW QXUVH UDWLQJV ZHUH LQIOXHQFHG E\ QXUVHVf NQRZOHGJH RI SKDUPDF\ UHILOO GDWD DV QXUVHV KDYH UHSRUWHG SUHYLRXVO\ WKDW SKDUPDF\ UHILOO UHFRUGV REWDLQHG GXULQJ URXWLQH FOLQLFDO FDUH ZHUH LQIOXHQWLDO LQ QXUVHVf UDWLQJV RI SDWLHQWVf DGKHUHQFH )DUOH\ HW DO f 7KH DXWKRU DOVR K\SRWKHVL]HG WKDW PHGLFDWLRQ DGKHUHQFH ZRXOG VLJQLILFDQWO\ DQG QHJDWLYHO\ UHODWH WR GLVHDVH VHYHULW\ 1HYHUWKHOHVV ZKHQ DQ b DGKHUHQFH FXWRII ZDV XVHG YLUDO ORDG &'& OHWWHU DQG &'& QXPEHU FODVVLILFDWLRQ ZHUH QRW VLJQLILFDQWO\ UHODWHG ZLWK 5, DGKHUHQFH UDWLQJV RU UHILOO UDWLQJV 1LQHW\SHUFHQW UHILOO DGKHUHQFH FODVVLILFDWLRQ GLG QRW YDU\ EDVHG RQ GLVHDVH VHYHULW\ KRZHYHU b 5, IUHTXHQF\ DGKHUHQFH FODVVLILFDWLRQ GLG YDU\ EDVHG RQ &'& QXPEHU FODVVLILFDWLRQ VXFK WKDW WKRVH ZKR KDG HYHU KDG ORZ &' FRXQWV DQG WKXV KLJKHU &'& QXPEHU FODVVLILFDWLRQVf WHQGHG WR EH FODVVLILHG DV DGKHUHQW PRUH WKDQ WKRVH ZLWK KLVWRULFDOO\ KLJKHU &' FRXQWV DQG

PAGE 77

WKXV ORZHU &'& QXPEHU FODVVLILFDWLRQVf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t *D]]DUG f DQG DPRQJ FKLOGUHQ .DWNR HW DO f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b DGKHUHQFH PD\ QRW EH VXIILFLHQW IRU YLURORJLF VXFFHVV 7KH UHVHDUFKHU DOVR K\SRWKHVL]HG WKDW DGKHUHQFH ZRXOG VLJQLILFDQWO\ DQG SRVLWLYHO\ UHODWH WR NQRZOHGJH ,W ZDV K\SRWKHVL]HG WKDW 5, IUHTXHQF\ DQG UHILOO DGKHUHQFH ZRXOG UHODWH WR FDUHJLYHU NQRZOHGJH RI PHGLFDWLRQ QDPHV DQG GRVLQJ IUHTXHQFLHV $OWKRXJK QHLWKHU DGKHUHQFH PHDVXUH ZDV VLJQLILFDQWO\ DVVRFLDWHG ZLWK GRVLQJ IUHTXHQF\ NQRZOHGJH FDUHJLYHUV ZLWK b NQRZOHGJH RI PHGLFDWLRQ QDPHV ZHUH VLJQLILFDQWO\ PRUH OLNHO\ WR KDYH b UHILOO DGKHUHQFH WKDQ WKRVH ZLWK LPSHUIHFW NQRZOHGJH RI PHGLFDWLRQ QDPHV 7KLV LV DQ LPSRUWDQW ILQGLQJ DV LW VXJJHVWV WKDW

PAGE 78

FOLQLFLDQV PD\ EH DEOH WR JDLQ YDOXDEOH LQIRUPDWLRQ DERXW DGKHUHQFH VLPSO\ E\ DVNLQJ FDUHJLYHUV WR VWDWH WKH PHGLFDWLRQV WKDW WKHLU FKLOGUHQ DUH FXUUHQWO\ WDNLQJ 1HYHUWKHOHVV LW UHPDLQV XQFOHDU ZKHWKHU FDUHJLYHUV DUH PRUH DGKHUHQW EHFDXVH WKH\ NQRZ WKH PHGLFDWLRQ QDPHV RU WKH\ NQRZ WKH PHGLFDWLRQ QDPHV EHFDXVH WKH\ DUH DGKHUHQW )XUWKHU LQYHVWLJDWLRQ LV QHFHVVDU\ LQ RUGHU WR GHWHUPLQH WR ZKDW H[WHQW FDUHJLYHUVf ODFN RI NQRZOHGJH LV GXH WR ODFN RI HIIHFWLYH UHJLPHQVSHFLILF HGXFDWLRQ ODFN RI UHJLPHQ EHKDYLRUV DQG ODFN RI FDUHJLYHUVf LQYROYHPHQW LQ WKH UHJLPHQ EHKDYLRUV $,+,9 LQWHUYDO NQRZOHGJH ZDV K\SRWKHVL]HG WR FRUUHODWH VLJQLILFDQWO\ DQG SRVLWLYHO\ ZLWK 5, LQWHUYDO 7KH SUHVHQW VWXG\ VXSSRUWV WKLV UHODWLRQVKLS LQGLFDWLQJ WKDW FDUHJLYHU UHSRUWV RI WKH W\SLFDO LQWHUYDOV EHWZHHQ GRVLQJ RYHU D ZHHN SHULRG WHQG WR LQFUHDVH DV FDUHJLYHU UHSRUWV RI WKH W\SLFDO LQWHUYDOV EHWZHHQ GRVLQJ RYHU f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t 6WDUN f ZKLFK VKRZHG WKDW GLHWDU\ DGKHUHQFH DPRQJ FKURQLFDOO\ LOO FKLOGUHQ WHQGV WR KRYHU DW RU EHORZ b DQG NQRZOHGJH RI WKH GLHWDU\ UHJLPHQ LV LQVXIILFLHQW IRU GLHWDU\ DGKHUHQFH

PAGE 79

,W ZDV DOVR K\SRWKHVL]HG WKDW DGKHUHQFH IUHTXHQF\ PHDVXUHV ZRXOG VLJQLILFDQWO\ UHODWH WR SUHVFULEHG GRVLQJ IUHTXHQF\ DQG IRUPXODWLRQ 5HILOO DGKHUHQFH ZDV VLJQLILFDQWO\ DVVRFLDWHG ZLWK SUHVFULEHG GRVLQJ IUHTXHQF\ VXJJHVWLQJ WKDW PHGLFDWLRQV SUHVFULEHG IRU WZLFH GDLO\ GRVLQJ DUH VLJQLILFDQWO\ PRUH OLNHO\ WR EH UHILOOHG WKDQ PHGLFDWLRQV SUHVFULEHG IRU RQFH GDLO\ GRVLQJ 7KLV UHVXOW LV LQWHUHVWLQJ JLYHQ WKDW RQFH GDLO\ GRVLQJ LV RIWHQ SUHIHUUHG RYHU WZLFH DQG WKULFH GDLO\ GRVLQJ 5RVHQEDFK $OOLVRQ t 1DGOHU f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f ,W PD\ EH WKDW VRPHWLPHV LW LV WRR GLIILFXOW WR IROORZ WKH GLHWDU\ UHTXLUHPHQWV DQG WKHUHIRUH DQ HQWLUH GRVH LV PLVVHG 5REHUWV t 0DQQ f %RWK 5, IUHTXHQF\ DQG UHILOO DGKHUHQFH ZHUH VLJQLILFDQWO\ UHODWHG WR PHGLFDWLRQ IRUPXODWLRQ WKRXJK WKH UHVXOWV ZHUH LQ RSSRVLWH GLUHFWLRQV 5, IUHTXHQF\ DGKHUHQFH ZDV JHQHUDOO\ KLJKHU IRU SLOOV WKRXJK UHILOO DGKHUHQFH ZDV KLJKHU IRU OLTXLG DQG SRZGHU

PAGE 80

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fH[DFWO\ DV SUHVFULEHGf 2Q DYHUDJH FDUHJLYHUV UHSRUWHG DERXW RQH EDUULHU HDFK ZKLFK LV IDU ORZHU WKDQ PLJKW EH H[SHFWHG JLYHQ WKH ODUJH SHUFHQWDJH RI FDUHJLYHUV GHILQHG DV QRQDGKHUHQW EDVHG RQ SKDUPDF\ UHILOO UDWHV RI OHVV WKDQ b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

PAGE 81

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f &DUHJLYHUV LQ WKH SUHVHQW VWXG\ GLG IDLO WR PHQWLRQ VRPH EDUULHUV WKDW KDYH EHHQ KLJKO\ HQGRUVHG E\ +,9LQIHFWHG DGXOWV LQ SDUWLFXODU HPRWLRQDOSV\FKRORJLFDO EDUULHUV VXFK DV EHLQJ DQJU\ 0XUSK\ HW DK f RU GHSUHVVHG &KHVQH\ 0XUSK\ HW DK f RU QRW ZDQWLQJ WR EH UHPLQGHG RI +,9 &DW] HW DK f ZHUH QRW PHQWLRQHG $OVR GLIILFXOW\ ZLWK WKH GLHWDU\ UHVWULFWLRQV DQG WLPLQJ RI GRVHV &DW] HW DK 0DQQKHLPHU HW DK 0XUSK\ HW DK 5REHUWV t 0DQQ :DOVK HW DK f ZHUH QRW PHQWLRQHG DV EDUULHUV :KHQ FDUHJLYHUV UHSRUWHG EDUULHUV WKH\ ZHUH DVNHG WR UDWH WKH IUHTXHQF\ RI RFFXUUHQFH DQG WKH VWUHQJWK RI EDUULHUV LQ SUHYHQWLQJ PHGLFDWLRQWDNLQJ 0DQ\ FDUHJLYHUV UHSRUWHG KLJK EDUULHU IUHTXHQF\ UDWLQJV DQG ORZ EDUULHU VWUHQJWK UDWLQJV ZHUH DOPRVW DOZD\V JLYHQ 7KHVH UHVXOWV LPSO\ WKDW DOWKRXJK IDPLOLHV PD\ H[SHULHQFH PDQ\ WKLQJV

PAGE 82

WKDW PDNH FKLOGUHQf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

PAGE 83

UHJLPHQ LV XVHIXO LQ DVVHVVLQJ NQRZOHGJH WKH UHVXOWLQJ LQIRUPDWLRQ DERXW GDLO\ DGKHUHQFH LV OLPLWHG DQG GLG QRW FRUUHODWH ZLWK RWKHU PHDVXUHV RI DGKHUHQFH LQ WKLV VWXG\ 5, DQG SKDUPDF\ UHILOO KLVWRU\f 3RVVLEOH LPSURYHPHQWV WR WKH PHDVXUH PD\ LQFOXGH WKH DGGLWLRQ RI FKLOGUHQ DV LQIRUPDQWV D VWDWHPHQW HPSKDVL]LQJ WKH LPSRUWDQFH RI KRQHVW UHVSRQGLQJ DQG WKH DGDSWDWLRQ RI WKH PHDVXUH LQWR D FRPSXWHUL]HG VWUXFWXUHG LQWHUYLHZ D IRUPDW ZKLFK KDV EHHQ IRXQG WR SURYLGH PRUH DFFXUDWH UHVSRQGLQJ WKDQ SURYLGHU LQWHUYLHZ %DQJVEHUJ HW DO f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fV FOLQLFDO XWLOLW\ ,Q WKLV VWXG\ 5, LQWHUYLHZHUV DVNHG FDUHJLYHUV WR GHVFULEH WKHLU FKLOGfV SUHYLRXV GD\ DQG ZKHQ PHGLFDWLRQWDNLQJ ZDV PHQWLRQHG E\ WKH FDUHJLYHUV LQWHUYLHZHUV UHFRUGHG LQIRUPDWLRQ DERXW WKH PHGLFDWLRQ QDPHV GRVLQJ ZKHWKHU RU QRW

PAGE 84

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b DQG b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

PAGE 85

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fV FRRSHUDWLRQ ZLWK PHGLFDWLRQn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f &DUHJLYHUV RI +,9LQIHFWHG FKLOGUHQ PD\ EH PRUH LQFOLQHG WRZDUGV RYHUHVWLPDWLQJ DGKHUHQFH EHFDXVH RI WKH H[WHQVLYH IRFXV RQ DGKHUHQFH WKDW LV W\SLFDO LQ +,9 FOLQLF HQYLURQPHQWV DQG PD\ EH PRUH LQWHQVLYH WKDQ LQ RWKHU FOLQLF HQYLURQPHQWV 7KLV H[WHQVLYH DGKHUHQFH IRFXV LV OLNHO\ IRVWHUHG E\ FOLQLFLDQV EDVHG RQ NQRZOHGJH RI WKH GHWULPHQWDO HIIHFWV RI HYHQ RFFDVLRQDO ODSVHV LQ DGKHUHQFHf§FRQVHTXHQFHV WKDW DUH XQPDWFKHG LQ PRVW RWKHU FKLOGKRRG FKURQLF LOOQHVVHV &DUHJLYHUV RI +,9LQIHFWHG FKLOGUHQ DOVR PD\ EH XQFRPIRUWDEOH GLVFORVLQJ DGKHUHQFH SUREOHPV IRU IHDU WKDW WKH SURYLGHUV ZLOO UHSRUW WKHP WR VRFLDO VHUYLFHV IRU PHGLFDO QHJOHFW ZKLOH FDUHJLYHUV RI

PAGE 86

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f &KLOGUHQ ZLWK +,9 RIWHQ OLYH LQ HFRQRPLF DQG HQYLURQPHQW FRQGLWLRQV WKDW DUH OHVV VWDEOH VDIH DQG VHFXUH WKDQ HQYLURQPHQWV RI FKLOGUHQ ZLWK RWKHU FKURQLF LOOQHVVHV DQG VRPH +,9LQIHFWHG FKLOGUHQ OLYH ZLWK +,9LQIHFWHG FDUHJLYHUV ZKR KDYH WKHLU RZQ PHGLFDO FRQFHUQV RU ZLWK IRVWHU SDUHQWV RU GLVWDQW IDPLO\ PHPEHUV ZKR PD\ KDYH RWKHU VLJQLILFDQW GHPDQGV RQ WKHLU WLPH HJ WKHLU RZQ PHGLFDO UHJLPHQV DQG UHJLPHQV RI RWKHU FKLOGUHQ LQ WKH KRPHf \HW WKHVH UHVXOWV LPSO\ WKDW +,9 FKLOGUHQ DQG WKHLU IDPLOLHV DQG SHUKDSV WKHLU SURYLGHUV DUH DFWXDOO\ GRLQJ D EHWWHU MRE DGKHULQJ WKDQ WKRVH ZKR PD\ KDYH IHZHU SV\FKRVRFLDO VWUHVVRUV 7KH +,9LQIHFWHG JURXS PD\ GR EHWWHU EHFDXVH WKH UHJLPHQ LQFOXGHV IHZHU FRPSRQHQWV WKDQ UHJLPHQV IRU GLDEHWHV DQG &) EHFDXVH WKH SDWLHQWSURYLGHU FRPPXQLFDWLRQV DERXW WKH UHJLPHQ DUH PRUH FRQVWUXFWLYH RU GXH WR VRPH FRPELQDWLRQ RI WKHVH DQG RWKHU IDFWRUV 5HJDUGOHVV RI WKH UHDVRQV IRU WKH EHWWHU DGKHUHQFH LQ WKH +,9LQIHFWHG JURXS WKLV LV DQ LPSRUWDQW REVHUYDWLRQ DV LW GHPRQVWUDWHV WKH UHVLOLHQFH DQG DGDSWDELOLW\ RI +,9LQIHFWHG FKLOGUHQ DQG WKHLU IDPLOLHV )XUWKHUPRUH WDNHQ WRJHWKHU WKH UHVXOWV HPSKDVL]H WKH LPSRUWDQFH RI H[DPLQLQJ DGKHUHQFH WR

PAGE 87

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b DGKHUHQW WR UHILOOV 7KHUHIRUH WKLV VWXG\ DGGV WR WKH OLWHUDWXUH E\ SURYLGLQJ HPSLULFDO VXSSRUW IRU WKH QHHG IRU HIIHFWLYH LQWHUYHQWLRQV WR LPSURYH FKLOGUHQfV DGKHUHQFH WR $57 7KLUG WKLV VWXG\ UHSUHVHQWV WKH ILUVW NQRZQ TXDOLWDWLYH DVVHVVPHQW RI EDUULHUV WR FKLOGUHQfV $57 DGKHUHQFH 7KH UHVXOWLQJ OLVW RI FDUHJLYHUVf SHUFHLYHG EDUULHUV ZLOO EH KHOSIXO LQ GHYHORSLQJ D EDUULHUV FKHVFNOLVW WKDW FRXOG EH XVHG WR TXLFNO\ DVVHVV EDUULHUV LQ WKH FOLQLFDO VHWWLQJ 7KLV VWXG\ KDV DOVR LQWURGXFHG WZR QHZ WRROV IRU $57 DGKHUHQFH DVVHVVPHQW WKH $,+,9 DQG WKH 5, 7KH XWLOLW\ RI WKH $,+,9 LQ DVVHVVLQJ NQRZOHGJH DQG EDUULHUV DQG WKH XWLOLW\ RI WKH 5, LQ DVVHVVLQJ DGKHUHQFH WR SUHVFULEHG GRVLQJ LQWHUYDOV GLHWDU\ UHTXLUHPHQWV DQG EDUULHUV KDYH ERWK EHHQ VXSSRUWHG E\ WKLV VWXG\ ,Q WKH IXWXUH WKHVH

PAGE 88

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b RI HOLJLEOH SDUWLFLSDQWV DSSURDFKHG UHIXVHG WR SDUWLFLSDWH LQ WKH VWXG\ VHYHUDO VXEMHFWV IDLOHG WR FRPSOHWH WKH 5, DQG SKDUPDF\ UHILOO GDWD ZHUH QRW UHWULHYDEOH IRU RQH VXEMHFW 6HOIVHOHFWLRQ ELDV PD\ KDYH RFFXUUHG ,W PD\

PAGE 89

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t :HHNV f WKRXJK SKDUPDF\ UHILOO KLVWRU\ KDV EHHQ IRXQG WR VWURQJO\ UHODWH WR 0(06 ILQGLQJV )DUOH\ HW DO f 7KXV WKH SKDUPDF\ UHILOO KLVWRU\ DSSHDUV WR EH D JRRG DOWHUQDWLYH LQ WKH DEVHQFH RI 0(06 GDWD 7KLV VWXG\ DOVR QHJOHFWHG FKLOGUHQ DV LPSRUWDQW VRXUFHV RI DGKHUHQFH LQIRUPDWLRQ 5HVHDUFKHUV KDYH SUHYLRXVO\ GHPRQVWUDWHG WKH XWLOLW\ RI REWDLQLQJ DGKHUHQFH LQIRUPDWLRQ IURP ERWK FKLOGUHQ DQG FDUHJLYHUV -RKQVRQ f WKHUHIRUH WKH DEVHQFH RI FKLOGUHSRUW PHDVXUHV LV D ZHDNQHVV RI WKH SUHVHQW VWXG\ &KLOGUHSRUW LV SDUWLFXODUO\ LPSRUWDQW DV FKLOGUHQ EHFRPH ROGHU DQG PRUH UHVSRQVLEOH IRU PHGLFDWLRQWDNLQJ $GGLWLRQDOO\ FKLOG UHSRUWV RI EDUULHUV LV LPSRUWDQW IRU FRQVWUXFWLQJ DQ DFFXUDWH PRGHO RQ ZKLFK WR EDVH IXWXUH LQWHUYHQWLRQV 7KLV VWXG\ FRPSOHWHV DQ LPSRUWDQW VWHS LQ WKH GRFXPHQWDWLRQ RI EDUULHUV WR FKLOGUHQfV $57 DGKHUHQFH WKRXJK FKLOGUHSRUW RI DGKHUHQFH EHKDYLRUV DQG EDUULHUV ZLOO EH LPSRUWDQW LQ IXWXUH UHVHDUFK

PAGE 90

)LQDOO\ WKLV VWXG\ LV OLPLWHG E\ WKH VPDOO VDPSOH VL]H ,GHDOO\ DW OHDVW FDUHJLYHUV RI +,9LQIHFWHG FKLOGUHQ ZRXOG KDYH FRPSOHWHG WKLV VWXG\ WKHUHE\ LQFUHDVLQJ WKH VWDWLVWLFDO SRZHU WR VXSSRUW WKH K\SRWKHVHV 'XH WR WKH UHODWLYHO\ VPDOO QXPEHUV RI FKLOGUHQ IROORZHG DW HDFK FOLQLF f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f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

PAGE 91

7KH SUHVHQW VWXG\ KDV DOVR GRFXPHQWHG FDUHJLYHUVf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f DGKHUHQFH WR FKURQLF LOOQHVV UHJLPHQV VXJJHVWV WKDW ZH PXVW SUHSDUH IRU GHFUHDVHV LQ DGKHUHQFH GXULQJ WKLV GHYHORSPHQWDO VWDJH 7KH LPSDFW RI +,9 VWDWXV RQ DGROHVFHQWVf LGHQWLW\ DQG VH[XDO H[SHULHQFHV PD\ IXUWKHU LQKLELW PHGLFDWLRQ DGKHUHQFH EHKDYLRU WKHUIRUH LQFOXGLQJ ULVN EHKDYLRUV LQ WKH DVVHVVPHQW RI DGKHUHQFH DPRQJ +,9LQIHFWHG \RXWK PD\ DGG YDOXDEOH LQIRUPDWLRQ 0XFK UHVHDUFK LV QHFHVVDU\ LQ RUGHU WR LQIRUP FOLQLFLDQV DERXW WKH PRVW HIIHFWLYH GLVHDVH PDQDJHPHQW GXULQJ WKLV WXPXOWXRXV SHULRG SDUWLFXODUO\ JLYHQ WKH VHYHUH

PAGE 92

FRQVHTXHQFHV RI SRRU DGKHUHQFH DQG VXEVHTXHQW GUXJ UHVLVWDQFH &OLQLFLDQV DQG UHVHDUFKHUV PXVW DVN WKHPVHOYHV DQG DGROHVFHQWV f*LYHQ DQ DGROHVFHQWfV SHUVRQDO DWWULEXWHV VXSSRUW V\VWHP DQG KRPH HQYLURQPHQW ZKDW FDQ ZH H[SHFW LQ WHUPV RI $57 DGKHUHQFH"f

PAGE 93

5()(5(1&(6 $EERWW /DERUDWRULHV f 0\ 0HGLFDWLRQ 6FKHGXOH 1RUWK &KLFDJR ,/ $EERWW /DERUDWRULHV ,QF $OHVVDQGUR ) 9LQFHQ]R =* 0DUFR 6 0DUFHOOR t (QULFD 5 f &RPSOLDQFH ZLWK SKDUPDFRORJLF SURSK\OD[LV DQG WKHUDS\ LQ EURQFKLDO DVWKPD $QQDOV RI $OOHUJ\ f $PHULFDQ $FDGHP\ RI 3HGLDWULFV f 3V\FKRVRFLDO ULVNV RI FKLOG KHDOWK FRQGLWLRQV LQ FKLOGKRRG DQG DGROHVFHQFH 3HGLDWULFV f $PPDVVDUL $ 0XUUL 5 3H]]RWWL 3 7URWWD 03 5DYDVLR / 'H/RQJLV 3 /R&DSXWR 6 1DUFLVR 3 3DXOX]]L 6 &DURVL 1DSSD 6 3LDQR 3 ,]]R &0 /LFWQHU 0 5H]]D 0RQWIRUWH $$ ,SSROLWR 0RURQL 0 :X $: t $QWLQRUL $ f 6HOIUHSRUWHG V\PSWRPV DQG PHGLFDWLRQ VLGH HIIHFWV LQIOXHQFH DGKHUHQFH WR KLJKO\ DFWLYH DQWLUHWURYLUDO WKHUDS\ LQ SHUVRQV ZLWK +,9 LQIHFWLRQ -RXUQDO RI $FTXLUHG ,PPXQH 'HILFLHQF\ 6\QGURPHV $QWKRQ\ + 3D[WRQ 6 %LQHV t 3KHODQ 3 f 3V\FKRVRFLDO SUHGLFWRUV RI DGKHUHQFH WR QXWULWLRQDO UHFRPPHQGDWLRQV DQG JURZWK RXWFRPHV LQ FKLOGUHQ ZLWK F\VWLF ILEURVLV -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK f %DQJVEHUJ '5 %URQVWRQH $ t +RIPDQ 5 f $ FRPSXWHUEDVHG DVVHVVPHQW GHWHFWV UHJLPHQ PLVXQGHUVWDQGLQJ DQG QRQDGKHUHQFH IRU SDWLHQWV RQ +,9 DQWLUHWURYLUDO WKHUDS\ $,'6 &DUH f %DQJVEHUJ '5 +HFKW )0 &KDUOHERLV (' =RORSD $5 +RORGQL\ 0 6KHLQHU / %DPEXUJHU -' &KHVQH\ 0$ t 0RVV $ f $GKHUHQFH WR SURWHDVH LQKLELWRUV +,9 YLUDO ORDG DQG GHYHORSPHQW RI GUXJ UHVLVWDQFH LQ DQ LQGLJHQW SRSXODWLRQ $,'6 %DQJVEHUJ '5 +HFKW )0 &ODJXH + &KDUOHERLV (' &LFFDURQH &KHVQH\ 0 t 0RVV $ f 3URYLGHU DVVHVVPHQW RI DGKHUHQFH WR +,9 DQWLUHWURYLUDO WKHUDS\ -RXUQDO RI $FTXLUHG ,PPXQH 'HILFLHQF\ 6\QGURPHV %HFN ( )HQQHOO 5 6
PAGE 94

%HO]HU 0 ( )XFKV '1 /XIWPDQ *6 t 7XFNHU 'f $QWLUHWURYLUDO DGKHUHQFH LVVXHV DPRQJ +,9SRVLWLYH DGROHVFHQWV DQG \RXQJ DGXOWV -RXUQDO RI $GROHVFHQW +HDOWK %HUJ 6 'LVFKOHU :DJQHU 5DLD t 3DOPHU6KHYOLQ 1 f 0HGLFDWLRQ FRPSOLDQFH $ KHDOWK FDUH SUREOHP 7KH $QQDOV RI 3KDUPDFRWKHUDS\ VXSSOHPHQWf %LUNKHDG $WWDZD\ 16WUXQFN 5& 7RZQVHQG 0& t 7HXWVFK 6 f ,QYHVWLJDWLRQ RI D FOXVWHU RI GHDWKV RI DGROHVFHQWV IURP DVWKPD (YLGHQFH LPSOLFDWLQJ LQDGHTXDWH WUHDWPHQW DQG SRRU SDWLHQW DGKHUHQFH ZLWK PHGLFDWLRQV -RXUQDO RI $OOHUJ\ DQG &OLQLFDO ,PPXQRORJ\ %RFFXWL / &HODQR 0 *HOOHU 5t 3KLOOLSV .0 f 'HYHORSPHQW RI D VFDOH WR PHDVXUH FKLOGUHQnV PHWHUHG GRVH LQKDOHU DQG VSDFHU WHFKQLTXH $QQDOV RI $OOHUJ\ $VWKPD t ,PPXQRORJ\ %RJDUW / 0 &DW] 6/ .HOO\ -$ *UD\%HPKDUGW 0/ +DUWPDQQ %5 3WWR6DODM // +DFNO ./ t %ORRP )5 f 3V\FKRVRFLDO LVVXHV LQ WKH HUD RI QHZ $,'6 WUHDWPHQWV IURP WKH SHUVSHFWLYH SI SHUVRQV OLYLQJ ZLWK +,9 -RXUQDO RI +HDOWK 3V\FKRORJ\ f %UHWWOH 53 :LOVRQ $ 3RYH\ 6 0RUULV 6 0RUJDQ 5 /HHQ &/ +XWFKLQVRQ 6 /HZLV 6 t *RUH 6 f &RPELQDWLRQ WKHUDS\ IRU +,9 WKH HIIHFW RQ LQSDWLHQW DFWLYLW\ PRUELGLW\ DQG PRUWDOLW\ RI D FRKRUW RI SDWLHQWV ,QWHUQDWLRQDO -RXUQDO RI 67' t $,'6 f %URZQEULGJH t )LHOGLQJ 0 f 3V\FKRVRFLDO DGMXVWPHQW DQG DGKHUHQFH WR GDLO\ WUHDWPHQWV UHJLPHQV 3HGLDWULF 1HSKURORJ\ %URZQOHH'XIIHFN 0 3HWHUVRQ / 6LPRQGV -) *ROGVWHLQ .LOR & t ,LRHWWH 6 f 7KH UROH RI KHDOWK EHOLHIV LQ WKH UHJLPHQ DGKHUHQFH DQG PHWDEROLF FRQWURO RI DGROHVFHQWV DQG DGXOWV ZLWK GLDEHWHV PHOOLWXV -RXUQDO RI &RQVXOWLQJ DQG &OLQLFDO 3V\FKRORJ\ f %XW] $ 0 -R\QHU 0 *UHHQEHUJ)ULHGPDQ t +XWWRQ 1 f 3ULPDU\ FDUH IRU FKLOGUHQ ZLWK KXPDQ LPPXQRGHILFLHQF\ YLUXV LQIHFWLRQ -RXUQDO RI 3HGLDWULF +HDOWK &DUH &DUSHQWHU & &RRSHU '$ )LVFKO 0$ *DWHOO -0 *D]]DUG %* +DPPHU 60 +LUVFK 06 -DFREVHQ '0 .DW]HQVWHLQ '$ 0RQWDQHU 5LFKPDQ 6DDJ 06 6FKHFKWHU 0 6FKRROH\ 57 7KRPSVRQ 0$ 9HOOD 6
PAGE 95

&DVVHOO (f 7KH 1DWXUH RI 6XIIHULQJ DQG WKH *RDOV RI 0HGLFLQH 1HZ
PAGE 96

)DUOH\ -+LQHV 6 0XVN $ )HUUXV 6 t 7HSSHU 9 f $VVHVVPHQW RI DGKHUHQFH WR DQWLYLUDO WKHUDS\ LQ +,9LQIHFWHG FKLOGUHQ XVLQJ WKH 0HGLFDWLRQ (YHQWV 0RQLWRULQJ 6\VWHP 0(06f SKDUPDF\ UHILOO SURYLGHU DVVHVVPHQW FDUHJLYHU VHOIUHSRUW DQG DSSRLQWPHQW NHHSLQJ 0DQXVFULSW VXEPLWWHG IRU SXEOLFDWLRQ )DWNHQKDXHU 7KHLVHQ $ t 5RFNVWURK f 9LURORJLFDO WUHDWPHQW IDLOXUH RI SURWHDVH LQKLELWRU WKHUDS\ LQ DQ XQVHOHFWHG FRKRUW RI +,9LQIHFWHG SDWLHQWV $,'6 )) )HLQVWHLQ $5 f &OLQLFDO ELRVWDWLVWLFV ;;; %LRVWDWLVWLFDO SUREOHPV LQ nFRPSOLDQFH ELDVn &OLQLFDO 3KDUPDFRORJ\ DQG 7KHUDSHXWLFV f )HVWD 56 7DPDURII 0+ &KDVDORZ ) t /DQ]NRZVN\ 3 f 7KHUDSHXWLF DGKHUHQFH WR RUDO PHGLFDWLRQ UHJLPHQV E\ DGROHVFHQWV ZLWK FDQFHU /DERUDWRU\ DVVHVVPHQW -RXUQDO RI 3HGLDWULFV )LQQH\ -: +RRN 5)ULPDQ 3& 5DSRII 0$ t &KULVWRSKHUVHQ (5 f 7KH RYHUHVWLPDWLRQ RI DGKHUHQFH WR SHGLDWULF PHGLFDO UHJLPHQV &KLOGUHQnV +HDOWK &DUH f )UHXQG $ -RKQVRQ 6% 6LOYHUVWHLQ t 7KRPDV f $VVHVVLQJ GDLO\ PDQDJHPHQW RI FKLOGKRRG GLDEHWHV XVLQJ KRXU UHFDOO LQWHUYLHZV 5HOLDELOLW\ DQG VWDELOLW\ +HDOWK 3V\FKRORJ\ )ULFN 3$ *DO 3 /DQH 7: t 6HZHOO 3& f $QWLUHWURYLUDO PHGLFDWLRQ FRPSOLDQFH LQ SDWLHQWV ZLWK $,'6 $,'6 3DWLHQW &DUH DQG 67'nV )ULHGODQG *+ t :LOOLDPV $ f $WWDLQLQJ KLJKHU JRDOV LQ +,9 WUHDWPHQW 7KH FHQWUDO LPSRUWDQFH RI DGKHUHQFH $,'6 6XSSOHPHQW f 66 *DOODQW -( f 6WUDWHJLHV IRU ORQJWHUP VXFFHVV LQ WKH WUHDWPHQW RI +,9 LQIHFWLRQ -RXUQDO RI WKH $PHULFDQ 0HGLFDO $VVRFLDWLRQ 2/ *HUELQR 3 f )RUZDUG $QQDOV RI 3KDUPDFRWKHUDS\ 66 *ODVJRZ 5( 0F&DXO .' t 6FKDIHU /& f %DUULHUV WR UHJLPHQ DGKHUHQFH DPRQJ SHUVRQV ZLWK LQVXOLQGHSHQGHQW GLDEHWHV -RXUQDO RI %HKDYLRUDO 0HGLFLQH f *ROGIILHG 0 5 t 'n =XULOOD 7f $ EHKDYLRUDQDO\WLF PRGHO IRU DVVHVVLQJ FRPSHWHQFH &XUUHQW WRSLFV LQ &OLQLFDO &KLOG 3V\FKRORJ\ 9RO SS f 1HZ
PAGE 97

*XGDV / .RRFKHU *3 t :\SLM f 3HUFHSWLRQV RI PHGLFDO FRPSOLDQFH LQ FKLOGUHQ DQG DGROHVFHQWV ZLWK F\VWLF ILEURVLV -RXUQDO RI 'HYHORSPHQWDO DQG %HKDYLRUDO 3HGLDWULFV f *XSWD 3 0HOORUV .LQJVOH\ / 5LGGOHU 6 6LQJK 0. 6FKUHLEHU 6 &URQLQ 0 t 5LQDOGR &5 f +LJK YLUDO ORDG LQ VHPHQ RI KXPDQ LPPXQLGHILFLHQF\ YLUXV W\SH LQIHFWHG PHQ DW DOO VWDJHV RI GLVHDVH DQG LWV UHGXFWLRQ E\ WKHUDS\ ZLWK SURWHDVH DQG QRQQXFOHRVLGH UHYHUVH WUDQVFULSWDVH LQKLELWRUV 9LURORJ\ +DOO % 0RQWDQHU 5HLVV 3 &RRSHU 9HOOD 6 'RKQDQ\L & 0\HUV 0 /DQJH t &RQZD\ % f ,QGXFWLRQPDLQWHQDQFH DQWLUHWURYLUDO WKHUDS\ SURRI RI FRQFHSW $,'6 f )) +DQVRQ & / +HQJJHOHU 6: t %XUJKHQ *$ Df 0RGHO RI DVVRFLDWLRQV EHWZHHQ SV\FKRVRFLDO YDULDEOHV DQG KHDOWKRXWFRPH PHDVXUHV RI DGROHVFHQWV ZLWK ,''0 'LDEHWHV &DUH f +DQVRQ & / +HQJJHOHU 6: t %XUJKHQ *$ Ef 6RFLDO FRPSHWHQFH DQG SDUHQWDO VXSSRUW DV PHGLDWRUV RI WKH OLQN EHWZHHQ VWUHVV DQG PHWDEROLF FRQWURO LQ DGROHVFHQWV ZLWK LQVXOLQGHSHQGHQW GLDEHWHV PHOOLWXV -RXUQDO RI &RQVXOWLQJ DQG &OLQLFDO 3V\FKRORJ\ f +DQVRQ & / +HQJJHOHU 6: +DUULV 0$ %XUJKHQ *$ t 0RRUH 0 f )DPLO\ V\VWHP YDULDEOHV DQG WKH KHDOWK VWDWXV RI DGROHVFHQWV ZLWK LQVXOLQ GHSHQGHQW GLDEHWHV PHOOLWXV +HDOWK 3V\FKRORJ\ f +DQVRQ & / 'H *XLUH 06FKLQNHO $0 t +HQJJHOHU 6: f &RPSDULQJ VRFLDO OHDUQLQJ DQG IDPLO\ V\VWHPV FRUUHODWHV RI DGDSWDWLRQ LQ \RXWKV ZLWK ,''0 -RXUQDO RI 3HGLDWULF 3V\FKRORJ\ f +DQVRQ & / 'H*XLUH 06FKQLNHO $0 .ROWHUPDQ 2* *RRGPDQ -3 t %XFNLQJKDP %$ f 6HOIFDUH EHKDYLRUV LQ LQVXOLQGHSHQGHQW GLDEHWHV (YDOXDWLYH WRROV DQG WKHLU DVVRFLDWLRQV ZLWK JO\FHPLF FRQWURO -RXUQDO RI 3HGLDWULF 3V\FKRORJ\ f OHYHUV &( %URZQ 57 'URWDU &DSODQ 3LVKHYDU %6 t /DPEHUW 5* f .QRZOHGJH RI SK\VLFLDQ SUHVFULSWLRQV DQG DGKHUHQFH WR WUHDWPHQW DPRQJ FKLOGUHQ ZLWK F\VWLF ILEURVLV DQG WKHLU PRWKHUV -RXUQDO RI 'HYHORSPHQWDO t %HKDYLRUDO 3HGLDWULFV f n -DFREVRQ $ 0 +DXVHU 67 /DYRUL 3 :ROIVGRUI +HUVNRZLW] 5' 0LOOH\ -) *HOIDQG ( :HUWOLHE t 6WHLQ f $GKHUHQFH DPRQJ FKLOGUHQ DQG DGROHVFHQWV ZLWK LQVXOLQGHSHQGHQW GLDEHWHV PHOOLWXV RYHU D IRXU\HDU ORQJLWXGLQDO IROORZXS 7KH LQIOXHQFH RI SDWLHQW FRSLQJ DQG DGMXVWPHQW -RXUQDO RI 3HGLDWULF 3V\FKRORJ\

PAGE 98

-DQ] 1 % 0+ f 7KH KHDOWK EHOLHI PRGHO $ GHFDGH ODWHU +HDOWK (GXFDWLRQ 4XDUWHUO\ -RKQVRQ 6% f 0DQDJLQJ LQVXOLQGHSHQGHQW GLDEHWHV PHOOLWXV LQ DGROHVFHQFH $ GHYHORSPHQWDO SHUVSHFWLYH ,Q /:DOODQGHU t / 6LHJHO (GVf $GROHVFHQW +HDOWK 3UREOHPV %HKDYLRUDO SHUVSHFWLYHV SS f 1HZ
PAGE 99

/D *UHFD $0 t 6FKXPDQ :% f $GKHUHQFH WR SUHVFULEHG PHGLFDO UHJLPHQV +DQGERRN RI 3HGLDWULF 3V\FKRORJ\ QG HG SS f 1HZ
PAGE 100

0XUSK\ %& t 'LOORQ & f ,QWHUYLHZLQJ LQ $FWLRQ 3URFHVV DQG 3UDFWLFH 3DFLILF *URYH &$ %URRNV &ROH 0XUSK\ '$ 5REHUWV .0DUWLQ '0DUHOLFK : t +RIIPDQ f %DUULHUV WR DQWLUHWURYLUDO DGKHUHQFH DPRQJ +,9LQIHFWHG DGXOWV $,'6 3DWLHQW &DUH DQG 67'V f 1HZDFKHFN 3: 0F0DQXV 0 )R[ +% +XQJ << t +DOIRQ 1 f $FFHVV WR KHDOWK FDUH IRU FKLOGUHQ ZLWK VSHFLDO KHDOWK FDUH QHHGV 3HGLDWULFV f 2n5HLOO\ 2 f 7KH UROH RI TXDOLWDWLYH UHVHDUFK LQ WKH JOREDO SURJUDPPH RQ $,'6 DW WKH :RUOG +HDOWK 2UJDQL]DWLRQ ,Q ( < /DPEHUW 56 $VKHU\ t 5+ 1HHGOH (GVf 4XDOLWDWLYH 0HWKRGV RI 5HVHDUFK LQ 'UXJ $EXVH DQG +,9 5HVHDUFK 9RO SS f 5RFNYLOOH 0' 1DWLRQDO ,QVWLWXWH RI 'UXJ $EXVH 'LYLVLRQ RI (SLGHPLRORJ\ DQG 3UHYHQWLRQ 5HVHDUFK 2VWURS 1t *LOO 0f $QWLUHWURYLUDO PHGLFDWLRQ DGKHUHQFH DQG SHUVLVWHQFH ZLWK UHVSHFW WR DGKHUHQFH WRRO XVDJH $,'6 3DWLHQW &DUH DQG 67'V f 3DHV $ + %DNNHU $ t 6RH$JQLH & f 0HDVXUHPHQW RI SDWLHQW FRPSOLDQFH 3KDUPDFRORJ\ :RUOG 6FLHQFH 3DOHOOD )'HODQH\ .0 0RRUPDQ $& /RYHOHVV 02 )KUHU 6DWWHQ *$ $VFKPDQ '+ROPEHUJ 6' t 7KH +,9 2XWSDWLHQW 6WXG\ ,QYHVWLJDWRUV f 'HFOLQLQJ PRUELGLW\ DQG PRUWDOLW\ DPRQJ SDWLHQWV ZLWK DGYDQFHG KXPDQ LPPXQRGHILFLHQF\ YLUXV LQIHFWLRQ 1HZ (QJODQG -RXUQDO RI 0HGLFLQH 3DULHQWL -9HPGRQ 5 t %D]LQ & f 7KH 3LOOV ,GHQWLILFDWLRQ 7HVW $ WRRO WR DVVHVV DGKHUHQFH WR DQWLUHWURYLUDO WKHUDS\ -RXUQDO RI WKH $PHULFDQ 0HGLFDO $VVRFLDWLRQ f 3DVVHUR 0$ 5HPRU % t 6DORPRQ f 3DWLHQWUHSRUWHG FRPSOLDQFH ZLWK F\VWLF ILEURVLV WKHUDS\ &OLQLFDO 3HGLDWULFV 3DWHUVRQ '/ 6ZLQGHOOV 6 0RKU %UHVWHU 0 9HUJLV ( 6TXLHU & :DJHQHU 0 t 6LQJK 1 f $GKHUHQFH WR SURWHDVH LQKLELWRU WKHUDS\ DQG RXWFRPHV LQ SDWLHQWV ZLWK +,9 LQIHFWLRQ $QQDOV RI ,QWHUQDO 0HGLFLQH 4XLWWQHU $/ 'URWDU '$ ,HYHUV/DQGLV & $ t 6HLGQHU Df %HKDYLRUDO IDPLO\ V\VWHPV WKHUDS\ %)67f IRU WHHQDJHUV ZLWK F\VWLF ILEURVLV DQG WKHLU SDUHQWV 7UHDWPHQW DQG LPSOHPHQWDWLRQ PDQXDO 8QSXEOLVKHG 0DQXDO

PAGE 101

4XLWWQHU $/ 'URWDU ,HYHUV/DQGLV & 6ORFXP 1 6HLGQHU t -DFREVHQ Ff ,Q 'URWDU (Gf 3URPRWLQJ $GKHUHQFH WR 0HGLFDO 7UHDWPHQW LQ &KURQLF &KLOGKRRG ,OOQHVV SS f 0DKZDK 1/DZUHQFH (DUOEDXP $VVRFLDWHV 4XLWWQHU $/ t (VSHODJH '/ f 5HOLDELOLW\ DQG YDOLGLW\ RI D GDLO\ SKRQH GLDU\ PHDVXUH WR DVVHVV GDLO\ DFWLYLWLHV DQG IDPLO\ LQWHUDFWLRQV 8QSXEOLVKHG 0DQXVFULSW 4XLWWQHU $/ (VSHODJH '/ /HYHUV/DQGLV & t 'URWDU Ef 0HDVXULQJ DGKHUHQFH WR PHGLFDO WUHDWPHQWV LQ FKLOGKRRG FKURQLF LOOQHVV &RQVLGHULQJ PXOWLSOH PHWKRGV DQG VRXUFHV RI LQIRUPDWLRQ -RXUQDO RI &OLQLFDO 3V\FKRORJ\ LQ 0HGLFDO 6HWWLQJV f 5DQG &6 t :HHNV f 0HDVXULQJ DGKHUHQFH ZLWK PHGLFDWLRQ UHJLPHQV LQ FOLQLFDO FDUH DQG UHVHDUFK 7KH +DQGERRN RI +HDOWK %HKDYLRU &KDQJH QG (Gf 1HZ
PAGE 102

6FRWW % t 6OHDVPDQ -: f 3HGLDWULF +,9 ,QIHFWLRQ )ORULGD 'HSDUWPHQW RI +HDOWK &KLOGUHQnV 0HGLFDOV 6HUYLFHV 6LQJK 1 %HUPDQ 60 6ZLQGHOO 6 -XVWLV -& 0RKU -$ 6TXLHU & t :DJHQHU 00 f $GKHUHQFH RI KXPDQ LPPXQRGHILFLHQF\ YLUXVLQIHFWHG SDWLHQWV WR DQWLUHWURYLUDO WKHUDS\ &OLQLFDO ,QIHFWLRXV 'LVHDVHV f 6LQJK 1 6TXLHU & 6LYHN & :DJPHU 0 1JX\HQ 0+ t
PAGE 103

:DOVK -& +RPH 5 'DOWRQ 0 %XUJHVV $3 t *D]]DUG %* f 5HDVRQV IRU QRQDGKHUHQFH WR DQWLUHWURYLUDO WKHUDS\ 3DWLHQWVn SHUVSHFWLYHV SURYLGH HYLGHQFH RI PXOWLSOH FDXVHV $,'6 &DUH f :DOVK -& 0DQGDOLD 6 t *D]]DUG %* f 5HVSRQVHV WR D PRQWK VHOI UHSRUW RQ DGKHUHQFH WR DQWLUHWURYLUDO WKHUDS\ DUH FRQVLVWHQW ZLWK HOHFWURQLF GDWD DQG YLURORJLFDO WUHDWPHQW RXWFRPH $,'6 :DWVRQ & t )DUOH\ -f (IILFDF\ RI DQG DGKHUHQFH WR KLJKO\ DFWLYH DQWLUHWURYLUDO WKHUDS\ LQ FKLOGUHQ LQIHFWHG ZLWK KXPDQ LPPXQRGHILFLHQF\ YLPV W\SH 3HGLDWULF ,QIHFWLRXV 'LVHDVH -RXUQDO f :HLVEHUJ%HQFKHOO *ODVJRZ $0 7\QDQ :' :LUW] 3 7XUHN t :DUG f $GROHVFHQW GLDEHWHV PDQDJHPHQW DQG PLVPDQDJHPHQW 'LDEHWHV &DUH

PAGE 104

$33(1',; $ (/,*,%,/,7< &+(&./,67 5HVHDUFK 3HUVRQQHO $VN \RXUVHOI WKH IROORZLQJ TXHVWLRQV DERXW WKH SDWLHQW 3ODFH D FKHFEQDUN QH[W WR WKH EHVW DQVZHU 3OHDVH FRPSOHWH $// LWHPV ,V WKH FKLOG FXUUHQWO\ SUHVFULEHG DQWLUHWURYLUDO PHGLFDWLRQ" \HV QR +DV WKH FKLOG EHHQ SUHVFULEHG DQWLUHWURYLUDO PHGLFDWLRQ IRU DW OHDVW PRQWKV" \HV QR +DYH WKH FKLOGfV PHGLFDWLRQV EHHQ WKH VDPH LQ WKH ODVW PRQWKV" YHV QR ,V WKH FKLOG EHWZHHQ WKH DJHV RI DQG \HDUV" \HV QR ,V WKH FKLOG FXUUHQWO\ HQUROOHG LQ DQRWKHU DGKHUHQFH \HV QR VWXG\" $// LWHPV PXVW EH PDUNHG <(6 WR SURFHHG WR EHORZ 3UHOLPLQDU\ 4XHVWLRQV ,QFOXVLRQ &ULWHULDf 5HVHDUFK 3HUVRQQHO $VN WKH FDUHVLYHU ZKR KDV DFFRPSDQLHG WKH FKLOG WR FOLQLFf WKH IROORZLQJ TXHVWLRQV 3ODFH D FKHFNPDUN QH[W WR WKH EHVW DQVZHU 3OHDVH FRPSOHWH $// RI WKH LWHPV $UH \RX RQH RI WKH SHUVRQV UHVSRQVLEOH IRU \RXU FKLOG JHWWLQJ +,9 PHGLFDWLRQV" \HV QR ,Q WKH ODVW PRQWK KDV \RXU FKLOG OLYHG ZLWK \RX" \HV QR ,Q WKH QH[W PRQWKV GR \RX H[SHFW WKDW \RXU FKLOG ZLOO FRQWLQXH WR OLYH ZLWK \RX" \HV QR ,V (QJOLVK \RXU ILUVW ODQJXDJH" \HV QR

PAGE 105

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fV LOOQHVV LQ WKH SDVW WZR ZHHNV" +DV WDNH DQ\ PHGLFDWLRQV LQ WKH SDVW WZR ZHHNV" ,I WKH FDUHJLYHU DQVZHUV \HV :KDW PHGLFDWLRQV KDV \RXU FKLOG WDNHQ" ,I WKH FDUHJLYHU DQVZHUV QR SURFHHG WR 6HFWLRQ 0DUN WKH PHGLFDWLRQV RQ WKH UHFRUGLQJ VKHHW ,I WKH FDUHJLYHU GHVFULEHV WKH PHGLFDWLRQ Vf VD\ 'R \RX NQRZ WKH QDPH RI LW" ,I WKH FDUHJLYHU FDQQRW UHFDOO WKH QDPH ILUVW VD\ :HfOO FRPH EDFN WR WKDW RQH :KDW RWKHU PHGLFDWLRQV KDV \RXU FKLOG WDNHQ" :KHQ WKH FDUHJLYHU FDQ QR ORQJHU QDPH DQ\ PHGLFDWLRQV LI VRPH DUH OHIW XQQDPHG WKHQ 6+2: 7+( ',63/$< &$5' ,I GLVSOD\ FDUG LV XVHG PDUN '& RQ UHFRUGLQJ VKHHW ,I VXEMHFW XVHV QRWHV WR QDPH PHGLFDWLRQ DW DQ\ WLPH PDUN &6 FKHDW VKHHWf IRU WKDW PHGLFDWLRQ 2WKHUZLVH EHJLQ ZLWK WKH ILUVW PHGLFDWLRQ DQG VD\
PAGE 106

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f 6HFWLRQ 1H[W VD\ 1RZ WKDW KDYH OHDUQHG DOO DERXW WKH PHGLFDWLRQV WKDW W\SLFDOO\ WDNHV ,fG OLNH \RX WR WHOO PH DERXW WKH WKLQJV \RXU GRFWRU QXUVH RU SKDUPDFLVW WROG \RX WR GR 'XULQJ \RXU ODVW FOLQLF YLVLW EHIRUH WRGD\f ZKDW PHGLFDWLRQV GLG \RXU GRFWRU QXUVH RU SKDUPDFLVW UHFRPPHQG WKDW \RX JLYH WR \RXU FKLOG" 0DUN WKH PHGLFDWLRQV RQ WKH UHFRUGLQJ VKHHW ,I WKH FDUHJLYHU GHVFULEHV WKH PHGLFDWLRQ Vf VD\ 'R \RX NQRZ WKH QDPH RI LW" ,I WKH FDUHJLYHU FDQQRW UHFDOO WKH QDPH ILUVW VD\ :HfOO FRPH EDFN WR WKDW RQH :KDW RWKHU PHGLFDWLRQV KDV \RXU GRFWRU QXUVH RU SKDUPDFLVW UHFRPPHQGHG" :KHQ WKH FDUHJLYHU FDQ QR ORQJHU QDPH DQ\ PHGLFDWLRQV LI VRPH DUH OHIW XQQDPHG WKHQ 6+2: 7+( ',63/$< &$5' 2WKHUZLVH EHJLQ ZLWK WKH ILUVW PHGLFDWLRQ DQG VD\ /HWf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

PAGE 107

0DQ\ GLIIHUHQW WKLQJV FDQ PDNH LW KDUG IRU FKLOGUHQ WR JHW WKH PHGLFLQHV WKHLU GRFWRUV UHFRPPHQG :KDW NLQGV RI WKLQJV PDNH LW KDUG IRU \RXU FKLOG WR WDNH OLNH WKH GRFWRU UHFRPPHQGHG" 5HFRUG WKH UHVSRQVH Vf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fV LOOQHVV 7KDQN \RX VR PXFK IRU KHOSLQJ XV RXW KDYH MXVW D IHZ PRUH TXHVWLRQV IRU \RX DQG WKHQ ZHfOO EH WKURXJK 5HDG WKH FDUHJLYHU GHPRJUDSKLF TXHVWLRQQDLUH WR WKH SDWLHQW

PAGE 108

3ODFH D FKHFN LQ WKH ER[ QH[W WR DOO PHGLFDWLRQV SUHVFULEHG IRU WKLV SDWLHQW 7KHQ ZULWH LQ GRVLQJ LQIRUPDWLRQ DQG FKHFN RQH ER[ IRU IUHTXHQF\ DQG FKHFN DOO DSSOLFDEOH ER[HV XQGHU fVSHFLDO LQVWUXFWLRQVf 0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ LQ WDEV PJ RU FF LQGLFDWHf )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV $EDFDYLU VXFFLQDWH =LDJHQ 8 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 4T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU $PSUHQDYLU $JHQHUDVH : Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU &RPELYLU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 'HODYLUGLQH 5HVFULSWRU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’WLG Â’TLG Â’ T Â’ T Â’ T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 'LGDQRVLQH GGO 9LGH[ 'LGHR[\LQRVLQH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU (IDYLUHQ] 6XVWLYD '03 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 109

0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV ,QGLQDYLU &UL[LYDQ 0. / Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 4T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU /DPLYXGLQH 7& (SLYLU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 4T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 1HOILQDYLU 9LUDFHSW $* Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 4T 2T 4T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 1HYLUDSLQH 193 9LUDPXQH %,5* Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 4T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 5LWRQDYLU 1RUYLU $%7 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 4 T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 6DTXLQDYLU ,QYLUDVH )RUWRYDVH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 6WDYXGLQH G7 =HULW Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 4T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 110

0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV =DOFLWDELQH GG& +LYLG 'LGHR[\F\WLGLQH 5 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG 2ELG Â’ WLG Â’ TLG 2T 4T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU =LGRYXGLQH $=7 ='9 5HWURYLU $]LGRWK\PDGLQH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 111

$33(1',; & '(02*5$3+,& 48(67,211$,5( 'DWH FRPSOHWHG PRQWK \HDU 6D\ 7KH QH[W TXHVWLRQV DUH DERXW \RX DQG \RXU FKLOG
PAGE 112

'RHV \RXU FKLOG NQRZ WKDW KHVKH KDV +,9" 4
PAGE 113


PAGE 114

+RZ PDQ\ SHRSOH OLYH LQ WKH KRPH ZLWK \RX DQG \RXU FKLOG" 3OHDVH OLVW DOO ,QLWLDOV $JH 5HODWLRQVKLS WR &KLOG 'RHV VRPHRQH KHOS PDNH VXUH WKDW \RXU FKLOG WDNHV KLVKHU PHGLFLQH" Â’
PAGE 115

$33(1',; 7(/(3+21( &217$&7 )250 6XEMHFW 3ULPDU\ 3KRQH 1XPEHU B f 7KH EHVW QXPEHU IRU XV WR UHDFK \RXf :KHQ FDQ ZH H[SHFW WR UHDFK \RX DW WKLV QXPEHU" FLUFOH DOO WKDW DSSO\f :HHNGD\ PRUQLQJV :HHNGD\ DIWHUQRRQV :HHNGD\ HYHQLQJV :KDW DUH WKH %(67 WLPHV IRU XV :HHNHQG PRUQLQJV :HHNHQG DIWHUQRRQV :HHNHQG HYHQLQJV UHDFK \RX DW WKLV QXPEHU" 6HFRQGDU\ 3KRQH 1XPEHU B f 7KH VHFRQG EHVW QXPEHU IRU XV WR UHDFK \RX ,W FDQ EH D ZRUN QXPEHU FHOO SKRQH RU IULHQGnV KRXVHf :KHQ FDQ ZH H[SHFW WR UHDFK \RX DW WKLV QXPEHU" FLUFOH DOO WKDW DSSO\f :HHNGD\ PRUQLQJV :HHNGD\ DIWHUQRRQV :HHNGD\ HYHQLQJV :HHNHQG PRUQLQJV :HHNHQG DIWHUQRRQV :HHNHQG HYHQLQJV :KDW DUH WKH %(67 WLPHV IRU XV WR UHDFK \RX DW WKLV QXPEHU" :KHQ ZH FDOO ZH ZLOO VD\ +HOOR WKLV LV FDOOLQJ IURP 3URMHFW )5$0( ,I \RX DUH QRW DW WKDW QXPEHU ZH ZLOO WU\ EDFN ODWHU

PAGE 116

$33(1',; ( 0(',&$/ 5(&25' 5(3257 )250 :KDW LV WKH FKLOGfV FXUUHQW &'& VWDWXV FOLQLFDO VWDWXVf" &KHFN RQH &KHFN RQH Â’ $ Â’ Â’ % Â’ Â’ & Â’ &KLOGfV PRVW UHFHQW &' FRXQW FHOOV b 'DWH RI PRVW UHFHQW &' FRXQW PRQWK GD\ \HDU &KLOGfV PRVW UHFHQW YLUDO ORDG FRSLHV P/ 'DWH RI PRVW UHFHQW YLUDO ORDG PRQWK GD\ \HDU 7R WKH EHVW RI \RXU DELOLW\ SOHDVH UDWH WKH FKLOGfV DGKHUHQFH WR KLVKHU DQWLUHWURYLUDO WKHUDS\ UHJLPHQ &LUFOH WKH QXPEHU WKDW EHVW UHSUHVHQWV WKH H[WHQW WR ZKLFK WKH FKLOG KDV IROORZHG KLVKHU PHGLFDO UHJLPHQ ZLWKLQ WKH SDVW PRQWKV 1RW DW DOO 6RPHWLPHV $OZD\V $GKHUHQW $GKHUHQW $GKHUHQW 3UHVFULEHG 5HJLPHQ SOHDVH FRPSOHWH WKH IRUP RQ WKH QH[W SDJHf

PAGE 117

3ODFH D FKHFN LQ WKH ER[ QH[W WR DOO PHGLFDWLRQV SUHVFULEHG IRU WKLV SDWLHQW 7KHQ ZULWH LQ GRVLQJ LQIRUPDWLRQ DQG FKHFN RQH ER[ IRU IUHTXHQF\ DQG FKHFN DOO DSSOLFDEOH ER[HV XQGHU fVSHFLDO LQVWUXFWLRQVf 0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ LQ WDEV PJ RU FF LQGLFDWHf )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV $EDFDYLU VXFFLQDWH =LDJHQ 8 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 4T 4T 2T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU $PSUHQDYLU $JHQHUDVH : Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU &RPELYLU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 'HODYLUGLQH 5HVFULSWRU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 'LGDQRVLQH GGO 9LGH[ 'LGHR[\LQRVLQH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU (IDYLUHQ] 6XVWLYD '03 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 118

0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV ,QGLQDYLU &UL[LYDQ 0. / Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU /DPLYXGLQH 7& (SLYLU Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 1HOILQDYLU 9LUDFHSW $* Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 1HYLUDSLQH 193 9LUDPXQH %,5* Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 5LWRQDYLU 1RUYLU $%7 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 6DTXLQDYLU ,QYLUDVH )RUWRYDVH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 4T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 6WDYXGLQH G7 =HULW Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T Â’ T Â’ T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 119

0HGLFDWLRQ QDPH 2WKHU QDPHV )RUPXODWLRQ 'RVLQJ )UHTXHQF\ 6SHFLDO ,QVWUXFWLRQV =DOFLWDELQH GG& +LYLG 'LGHR[\F\WLGLQH 5 Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU =LGRYXGLQH $=7 ='9 5HWURYLU $]LGRWK\PDGLQH Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T Â’ T 4T 4TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T Â’ T Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG 2T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU 2WKHU $QWLUHWURYLUDO 0HGLFDWLRQ Â’ FDSVXOHV WDEOHWV Â’ SRZGHU Â’ OLTXLG Â’ TG Â’ ELG Â’ WLG Â’ TLG Â’ T 2T 2T 2TO Â’ WDNH ZLWK IRRG Â’ WDNH RQ HPSW\ VWRPDFK Â’ QR IDWW\ IRRGV Â’ RWKHU

PAGE 120

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f 'R \RX UHPHPEHU ZKDW WLPH \RXU FKLOG JRW XS \HVWHUGD\"f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fV GD\ \HVWHUGD\ 7KLV VKRXOG WDNH DERXW PLQXWHV /HWfV JHW VWDUWHG 'R \RX UHPHPEHU ZKDW WLPH \RXU FKLOG ZRNH XS \HVWHUGD\" 7KH UHPDLQGHU RI WKH TXHVWLRQV VKRXOG IORZ IURP KHUH :KHQ WKH LQIRUPDQW PHQWLRQV WKDW PHGLFDWLRQWDNLQJ RFFXUUHG UHFRUG LQIRUPDWLRQ DERXW WKH PHGLFDWLRQ QDPHV DQG GRVLQJ ,I WKH LQIRUPDQW HOXGHV WR PHGLFDWLRQWDNLQJ EXW GRHV QRW SURYLGH WKH VSHFLILF QDPHV DQG GRVLQJ DVN WKH LQIRUPDQW ZKLFK PHGLFDWLRQV WKH FKLOG WRRN DW D JLYHQ WLPH DQG KRZ PXFK WKH FKLOG UHFHLYHG RI HDFK PHGLFDWLRQ

PAGE 121

7R ILQG RXW LI WKH LQIRUPDQW REVHUYHG WKH FKLOG WDNLQJ WKH PHGLFDWLRQ RU HDWLQJ VD\ :HUH \RX ZLWK KLPKHU ZKHQ KHVKH WRRN KLV RU GLG \RX SUHWW\ PXFK OHW KLPKHU WDNH FDUH RI LW KLPKHUVHOI" ,I LQIRUPDQW VD\V WKDW KH RU VKH ZDV ZLWK WKH FKLOG VD\ 6R GLG \RX VHH KLPKHU WDNH KLVKHU 2QFH \RX KDYH JRQH WKURXJK WKH FKLOGfV HQWLUH SUHYLRXV GD\ DVN WKH SDUHQW ,V WKHUH LV DQ\WKLQJ HOVH WKDW \RX FDQ UHPHPEHU DERXW \RXU FKLOGnV GD\ \HVWHUGD\" ,V WKHUH DQ\WKLQJ HOVH WKDW \RX WKLQN ZRXOG EH JRRG IRU PH WR NQRZ" ,I WKH LQIRUPDQW KDV QRW PHQWLRQHG WKDW WKH FKLOG WRRN DQ\ PHGLFDWLRQ \HVWHUGD\ VD\ 6RPHWLPHV IDPLOLHV KDYH GD\V ZKHQ WKH FKLOGUHQ GRQfW JHW DQ\ RI WKHLU SUHVFULEHG PHGLFDWLRQ HLWKHU EHFDXVH WKH\ IRUJRW RU WKH\ UDQ RXW RI PHGLFDWLRQ RU IRU PDQ\ RWKHU UHDVRQV :DV \HVWHUGD\ RQH RI WKRVH GD\V IRU \RXU IDPLO\" ,I \RX DVN WKLV TXHVWLRQ GHVLJQDWH WKDW RQ WKH IRUP E\ ZULWLQJ f$4f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f DQG DVN WKH UHVSRQGHQW :KDW RWKHU WKLQJV \HVWHUGD\ NHSW \RXU FKLOG IURP JHWWLQJ KLVKHU PHGLFDWLRQ H[DFWO\ OLNH WKH GRFWRU VXJJHVWHG" 5HFRUG WKH UHVSRQVH DQG VD\ 2I DOO WKH WKLQJV \RX MXVW OLVWHG ZRXOG OLNH \RX WR WHOO PH ZKLFK RQH PDGH LW WKH KDUGHVW IRU \RXU FKLOG WR JHW KLVKHU PHGLFDWLRQ \HVWHUGD\ 5HPLQG WKHP RI WKH LWHPV RQ WKH OLVW LI QHFHVVDU\f 5HFRUG WKH UHVSRQVH DQG VD\ *UHDW 1RZ IURP WKH RWKHU RQHV \RX VDLG WHOO PH ZKLFK PDGH LW WKH KDUGHVW IRU \RXU FKLOG WR JHW KLVKHU PHGLFDWLRQ \HVWHUGD\ 5HPLQG WKHP RI WKH LWHPV RQ WKH OLVW LI QHFHVVDU\f 5HFRUG WKH UHVSRQVH DQG SURFHHG ZLWK WKH UDQNLQJ XQWLO DOO WKH EDUULHUV DUH OLVWHG LQ RUGHU :KHQ \RX DUH GRQH ZLWK WKH LQWHUYLHZ VD\ 7KDQN \RX IRU VKDULQJ WKLV LQIRUPDWLRQ ZLWK XV 7KLV LQIRUPDWLRQ LV YHU\ LPSRUWDQW DQG KHOSV XV WR XQGHUVWDQG ZKDW LW LV OLNH IRU \RXU IDPLO\ WR PDQDJH \RXU FKLOGfV LOOQHVV ,I DSSURSULDWHf VD\ :H ZLOO EH FDOOLQJ \RX DJDLQ VRRQ WR OHDUQ PRUH DERXW ZKDW WKLQJV DUH OLNH IRU \RXU IDPLO\ ,fOO VSHDN ZLWK \RX WKHQ 7KDQN \RX *RRGE\H

PAGE 122

%,2*5$3+,&$/ 6.(7&+ 6WHSKDQLH / 0DUKHIND ZDV ERP LQ -RKQVWRZQ 3HQQV\OYDQLD RQ $XJXVW 6KH ZDV WKH ILIWK FKLOG RI %DUEDUD DQG -RKQ 0DUKHIND 6WHSKDQLH VSHQW KHU ILUVW VL[ \HDUV RI OLIH LQ :LQGEHU 3HQQV\OYDQLD EHIRUH PRYLQJ WR 6W &ODLUVYLOOH 2KLR 6WHSKDQLH JUDGXDWHG IURP 6W &ODLUVYLOOH +LJK 6FKRRO LQ 6KH DWWHQGHG 0LDPL 8QLYHUVLW\f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

PAGE 123

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

PAGE 124

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ )RQGD (\OHU 9 3URIHVVRU RI 3V\FKRORJ\ 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI +HDOWK 3URIHVVLRQV DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 'HFHPEHU 'HDQ &ROOHJH RI +HDOWK 3URIHVVLRQV


BIOGRAPHICAL SKETCH
Stephanie L. Marhefka was bom in Johnstown, Pennsylvania, on August 3, 1975.
She was the fifth child of Barbara and John Marhefka. Stephanie spent her first six years
of life in Windber, Pennsylvania, before moving to St. Clairsville, Ohio. Stephanie
graduated from St. Clairsville High School in 1993. She attended Miami Universitys
Western College Program in the School of Interdisciplinary Studies, and received her
Bachelor of Philosophy, Cum Laude, in May of 1997. In August of 1997 she began her
graduate study at the University of Florida. She received her Master of Science degree in
May of 1999. Stephanie completed the University of Maryland Pediatric Psychology
Clinical Internship in July of 2002. After completing her Doctor of Philosophy degree,
Stephanie will complete a one-year postdoctoral fellowship in the Division of
Immunology in the University of Maryland Department of Pediatrics. She intends to
pursue an academic career and continue working with HIV-infected children and
adolescents.
114


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EI91NE5BT_YNZ6ZK INGEST_TIME 2014-11-07T19:16:05Z PACKAGE AA00024439_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


RESULTS
52
Caregiver Knowledge of the Prescribed Regimen 52
Relationship between Knowledge and Typical Adherence Behavior 53
Utility of Medication Display Card 56
Adherence Rates 56
Relationship between Adherence Measures 57
Relationship between Adherence and Disease Severity 57
Relationship between Adherence and Knowledge 58
Relationship between Adherence and Prescribed Dosing Frequency 58
Relationship between Adherence and Medication Formulation 60
Barriers to Adherence 60
Barriers Reported with the AI-HIV versus the 24RI 61
DISCUSSION 64
Caregiver Knowledge of the Prescribed Regimen 64
Adherence Rates 66
Barriers to Adherence 72
Implications 74
Study Strengths 79
Study Limitations 80
Future Directions 82
REFERENCES 85
APPENDIX
A ELIGIBILITY CHECKLIST 96
B ADHERENCE INTERVIEW-HIV 97
C DEMOGRAPHIC QUESTIONNAIRE 103
D TELEPHONE CONTACT FORM 107
E MEDICAL RECORD REPORT FORM 108
F 24-HOUR RECALL INTERVIEW 112
BIOGRAPHICAL SKETCH 114
vi


29
considered adherent by this more stringent criterion. Like the previous study, this study
is limited by the method of adherence assessment, as refill histories are subject to error
when patients have extra medication at home or when patients refill prescriptions at
pharmacies that are not listed in the research databases. The approach in this study was
also unidimensional, failing to account for differences in adherence based on individual
medications, and adherence to food and drink intake guidelines. However, the utility of
refill histories in estimating adherence has been supported by the literature (Laine et al.,
2000; Monane, Gurwitz, Monane, & Avom, 1993; Singh et al., 1999) and despite the
unidimensional approach, the study is useful in suggesting that poor adherence may be a
pervasive problem for families of children on ART regimens.
A recent study with children <13 years-old was the first known study to assess
ART adherence multidimensionally among children (Farley et al., 2002). The study
examined adherence among 26 children using MEMS, pharmacy refills, caregiver self-
reports, and physician and nurse assessments. Median adherence ratings were 81% for
MEMS, 79% for pharmacy refills, and 100% when caregivers were asked to report the
missed doses during the 3 days prior to interview, though MEMS data demonstrated 31
missed medication events of the 126 prescribed events for the 20 respondents over the 3-
day period. Physician and nurse ratings were also assessed, and were significantly
related to MEMS data (kappa= 74%, p< .05). Pharmacy refill ratings alone did not
significantly relate to viral load data, but when MEMS and pharmacy data were
combined, an adherence cut-off of 80% was a good predictor of virologic success. This
study is consistent with studies of HIV-infected adults, which illustrate the utility of
MEMS and pharmacy refill histories for assessing ART adherence (Katko et al., 2001;


77
intervention targets. First, the barrier assessments suggest that drug manufacturers
should work to modify medication attributes in order to decrease the size of pills,
improve the taste of both pills and liquids, reduce the number of pills required, and
generally simplify the regimen. Second, the barrier assessments suggest that healthcare
providers, including physicians, nurses, pharmacists, social workers, and psychologists,
should work with families to help eliminate medication scheduling problems, and educate
families on the use of appropriate strategies to observe medication-taking, help remember
the medication schedule, and help improve childrens cooperation with medication
taking.
Finally, results of the study imply that pediatric HIV differs significantly from
other childhood chronic illnesses. The differences between pediatric HIV and other
childhood chronic illnesses may be extensive, though the literature has not systematically
examined them. Two such differences emerged in this study. First, results of this study
imply that caregivers of HIV-infected children may be more susceptible to socially
desirable adherence self-reporting than caregivers of children with CF (Quittner et al.,
2000b). Caregivers of HIV-infected children may be more inclined towards
overestimating adherence because of the extensive focus on adherence that is typical in
HIV clinic environments, and may be more intensive than in other clinic environments.
This extensive adherence focus is likely fostered by clinicians based on knowledge of the
detrimental effects of even occasional lapses in adherenceconsequences that are
unmatched in most other childhood chronic illnesses. Caregivers of HIV-infected
children also may be uncomfortable disclosing adherence problems for fear that the
providers will report them to social services for medical neglect, while caregivers of


7
are needed to have a beneficial effect (Urquhart, 1989). Due to perceived ineffectiveness
the drug may never become available, when it actually may be beneficial for those
patients who take the medication as prescribed.
Finally, perhaps the most obvious implication of poor adherence is poor health
outcomes. In many chronic illnesses, and with many disease management behaviors, the
ultimate consequence of long-term poor adherence is death. Before reaching that level of
severity, poor adherence may lead to illness progression and complications, increased
hospitalizations, and an increase in the number and frequency of disease management
behaviors necessary in order to maintain or improve health, particularly for illnesses such
as HIV, cystic fibrosis (CF), hypertension, and renal disease. In other illnesses, such as
diabetes and asthma, poor adherence over even a short period of time may result in death
(Birkhead, Attaway, Strunk, Townsend, & Teutsch, 1989; Delamater, 2000; Gerbino,
1993). With the exception of palliative treatments, poor adherence with most disease
management behaviors results in increased morbidity and mortality.
Assessment of Medication Adherence
The assessment of adherence is a complex problem for which there is no easy
solution. Although several assessment methods have been used to measure adherence, no
perfect method has been identified (Rapoff, 1999). In the absence of a gold standard
for adherence assessment, researchers have suggested that multiple assessment modalities
should be used in order to obtain the most comprehensive information. Adherence
measurements have been characterized as direct and indirect assessments (Rapoff,
1999). Direct assessments such as biological assays, direct observations, and electronic
monitoring are often difficult to obtain and can be costly (Rand & Weeks, 1998).


82
Finally, this study is limited by the small sample size. Ideally, at least 100
caregivers of HIV-infected children would have completed this study, thereby increasing
the statistical power to support the hypotheses. Due to the relatively small numbers of
children followed at each clinic (60-140) and the exclusion criteria set for the study, only
63 families were recruited. As a result of the small sample size, it was impossible
compare adherence across all the different medications. Although the small sample size
did limit the power of the study, this study adds important information to the literature
regarding caregiver knowledge of the prescribed regimen, adherence rates, perceived
barriers to adherence, and assessment of knowledge, adherence to interval and dietary
requirements, and barriers to adherence among HIV-infected children and their
caregivers and other children with chronic illness and their caregivers.
Future Directions
The current study has begun to close a significant gap in the literature pertaining
to childrens adherence to ART. In particular, this study has provided information
suggesting that the AI-HIV and 24RI are useful measures for assessing caregiver
knowledge, adherence behavior, and caregiver-perceived barriers to adherence among
this population. Future studies should examine ways to maximize the utility and
determine the reliability and validity of these instruments, and specifically, convergent
validity analyses should consider both patient-specific goals of ART and patient levels of
drug resistance as they relate to virologic success. Studies should also assess the utility
of the AI-HIV and 24RI in assessing ART adherence based on child self-report, and
caregiver and child-report combined.


ACKNOWLEDGEMENTS
I am grateful for so many people who contributed to the successful completion of
my doctoral dissertation. I would like to thank my doctoral committee members for their
guidance throughout this very difficult and overwhelming process. I am particularly
grateful for the help of my chair, Dr. Jim Rodrigue, who agreed to chair my dissertation
even when it did not fit neatly into his silo. I also owe special thanks to Dr. Sam Sears,
who chaired my masters thesis and continued to support and encourage me after I made
the very difficult decision to change my area of concentration to pediatrics and
subsequently change my primary mentor.
This dissertation would not have been possible without the support and
collaboration of directors and staff of the pediatric immunology clinics at the University
of Florida, Gainesville, the University of Florida, Jacksonville, and the University of
Maryland at Baltimore. I owe special thanks to Dr. John Sleasman, Dr. Judy Lew, Carla
Duff, Dr. Mobeen Rathmore, Dr. Lauriann Sanders, Dr. Vicki Tepper, Dr. John Farley,
Dr. Douglas Watson, Dr. Peter Vink, Marie Parks, and Angelo Seda, for allowing me to
invade their clinics, helping me to make sense of hectic clinic environments, and teaching
me about pediatric HIV clinical practice. Doctors Vicki Tepper and John Farley were
especially monumental in mentoring me through the University of Maryland Institutional
Review Board processes and providing me time and resources for completing my
dissertation during internship and the first months of my fellowship.
11


31
adherence among HIV-infected adolescents and only several studies have examined
barriers to adherence among HIV-infected adults.
The PACTG adherence modules require the nurse to conduct a structured
interview to assess knowledge of the medication names, dosage, frequency of dosing, and
number of missed doses in previous 3 days (Module 1), and then assess barriers by
reading a list of previously identified barriers and then asking caregivers: Over the last
two weeks, have any of the following been problems for you with (drug name)?
(Module 2). When Farley and colleagues (2002) administered the modules to a group of
20 caregivers of children under age 13, 6 caregivers identified problems adhering to 17
medications. The most commonly endorsed problems were didnt refill, ran out,
scheduling interferes with lifestyle, child refuses, multiple caretakers, forgot,
and taste. All of the caregivers who identified problems had missed medication during
the prior two weeks, according to MEMS data. Conversely, not all of those who had
missed doses identified problems. Thus, this study identified some barriers that may be
common among caregivers of HIV-infected children and suggests that problem or barrier
identification may be one indicator of poor adherence. However, this study was limited
by the small sample size and the close-ended format of the barrier questions.
Another study assessed barriers to adherence among adolescents with HIV.
Belzer and colleagues (1999) provided 31 teenagers with a 7-item list of reasons they
may have missed their medication, and asked the teens to rate the reasons on a Likert
scale that ranged from 1= strongly disagree to 2= strongly agree. The items, in order
from most highly endorsed to least highly endorsed, are: too many pills, reminds me
Im HIV-positive, side effects, interferes with schedule, forgot, disclosure


2
When HIV adherence is poor, the potential consequences are severe; poor
adherence may be detrimental to both the individual and society. When children and
families are nonadherent to their anti-HIV medication regimens, HIV-infected children
may experience rapid physical decline with few therapeutic options (Butz, Joyner,
Greenberg-Friedman, & Hutton, 1998; Fatkenhauer, Theisen, & Rockstroch, 1997). In
terms of the impact on society, poor adherence may mean increased rates of transmission
of HIV, and increased difficulty treating the virus (Wainberg & Friedland, 1998). This is
particularly alarming, since adolescents are among the largest newly-infected HIV
population (UNAIDS, 2000). Clearly, poor adherence among HIV-infected children and
adults is a major public health concern.
Unfortunately, few studies have examined adherence to anti-HIV medication
regimens among children. As a result, little is known about the extent and nature of
adherence problems in this population. Thus, an understanding of adherence is necessary
in order to develop specific, focused, and effective adherence interventions. The major
purpose of this study was to assess both the extent of medication adherence problems in
this population and the specific barriers that prevent consistent adherence.
This paper begins with a review of adherence to chronic illness regimens,
including the prevalence of poor adherence, the definition of adherence, implications of
poor adherence, and the assessment of medication adherence and barriers to adherence.
The paper then focuses on HIV as a new chronic illness affecting both children and
adults. Information about the prevalence of HIV among children is presented, as well as
general information about HIV and the corresponding antiretroviral therapy (ART)
regimen. Next, implications of poor adherence to ART and the assessment of medication


I also wish to extend my gratitude towards the many individuals who helped me
with my literature review and data collection. Numerous research assistants retrieved
articles, collected clinical data, and spent many hours conducting telephone interviews
for this study. Joe Palmer was particularly gracious about working on weekends, putting
in numerous hours the week before my proposal, and driving to Jacksonville weekly for
data collection. Hanna Frost was my most enduring research assistant, and I am quite
grateful for her continued efforts, especially during my internship year. Bill Weisner was
not a research assistant per se, but he deserves my appreciation, nevertheless, as he
graciously acted as a substitute research assistant when deadlines drew near.
Several individuals deserve special thanks for reviewing my manuscript at various
stages of the process. Despite numerous other commitments, Dr. Vicki Tepper was kind
enough to read my draft before I submitted it to my chair. Dr. Sylvie Lombardo was also
extremely helpful with my result sectionMerci! Jack Rusher graciously read through
my discussion and helped to improve my grammar and tense consistency, and Dr.
Jennifer Brown remained patient as I directed many dissertation questions her way.
Finally, I wish to thank my friends and family for their continued support and
encouragement throughout my tenure as a graduate student. I will always be grateful for
the ways in which my sweetheart, Jack Rusher, encouraged me to persist, despite
seemingly insurmountable hardships, and the way that he believed in me even when I
struggled to believe in myself. I will also never forget that Brandy Werba, Jane Querido,
and Karen Bearss stood in the bathroom with me as I cried on the day I reached my first
substantial dissertation roadblock. Similarly, I will remember the way that my sister
Suzanne encouraged me over the telephone to keep working. Moreover, I will not forget
iii


11
knowledge among this group, though they did report that knowledge was widely variable
and related to self-reports of adherence behavior.
DiGirolamo, Quittner, Ackerman, & Stevens (1997) were among the first to
report a comprehensive assessment of knowledge of the self-care regimen for children
with CF. They developed the Treatment Adherence Questionnaire-Cystic Fibrosis
(TAQ-CF), with a child and parent-version 10-item self-report measure that assesses
adherence and knowledge of physicians treatment recommendations for airway
clearance, aerosol treatments, and pancreatic enzyme use. The measure has been found
to have good test-retest reliability (a= .62-.88).
levers and others (1999) used the TAQ-CF and found that 19.5% of mothers
incorrectly identified the prescribed frequency of airway clearance treatments, and 32%
incorrectly identified the prescribed frequency of aerosol medications. Eleven and a half
percent of caregivers incorrectly identified the greatest quantity of enzymes their children
should take with a meal, and 29.8% incorrectly identified the greatest quantity of
enzymes to be taken with a snack, though few of the enzyme identifications were grossly
incorrect. Children demonstrated even poorer regimen knowledge. Both child and
caregiver knowledge were predictive of self-reports of adherence.
Ricker, Delamater, and Hsu (1998) also reported the comprehensive assessment
of knowledge of the self-care regimen for children with CF. They asked each child to
indicate the prescribed amount of chest physiotherapy, the type of antibiotic,
multivitamin, and pancreatic supplement, and corresponding frequencies, and dosing
times. Regrettably, although the researchers collected these important data, they did not
report any information about the knowledge-related findings.


INTRODUCTION
Estimates suggest that 18% of children in the United States have a chronic illness
(Newacheck, McManus, Fox, Hung, & Halfon, 2000). Among children with chronic
illness, the prevalence of poor adherence to medical regimens varies according to the
study sample, the specifics of regimen requirements, the assessment of adherence, and
the criteria used to classify children as adherent or poorly adherent (Rapoff, 1999).
Nevertheless, estimates suggest that 21-52% of children with chronic illness may not
fully adhere to their regimens (Alessandro, Vincenzo, Marco, Marcello, & Enrica, 1994;
Conley & Salvatierra, 1996; Ettenger et al., 1991; Festa, Tamaroff, Chasalow, &
Lanzowsky, 1992; Meyers, Thompson, & Weiland, 1996; Schoni, Horak, & Nikolaizik,
1995; Weisberg-Benchell et al., 1995). Adverse effects of poor adherence include health
care costs, clinical decision-making, conclusions drawn from clinical trials, morbidity
and mortality (Rapoff, 1999).
The Human Immunodeficiency Virus (HIV) is a chronic illness that affects
children as well as adults. Medication regimens for HIV-infected children require
multiple medications to be taken at specific times throughout the day (Scott & Sleasman,
1999). The pills are often large and difficult to swallow, and many of the liquid
medications have a bad taste. For these and other reasons, families may have difficulty
adhering to pediatric HIV medication regimens. This is particularly problematic, as poor
adherence to anti-HIV medications is a serious public health concern.
1


81
be that people with lower adherence were less likely to participate or less likely to
complete the 24RI. If that is true, then the reported rates of adherence may overestimate
adherence among the pediatric HIV population.
The next group of considerations relates to adherence measurement. Although
problems exist with all adherence measurement techniques, electronic monitoring devices
have become a preferred method for measuring adherence to HAART. This study was
unable to use electronic monitoring because of cost, and instead assessed adherence
behaviors with clinical interview, a caregiver self-report recall interview, nurse report,
and pharmacy refill histories. Historically, self-report and provider assessment of
adherence tend to grossly overestimate adherence (Rand & Weeks, 1998), though
pharmacy refill history has been found to strongly relate to MEMS findings (Farley et al.,
2002). Thus, the pharmacy refill history appears to be a good alternative in the absence
of MEMS data.
This study also neglected children as important sources of adherence information.
Researchers have previously demonstrated the utility of obtaining adherence information
from both children and caregivers (Johnson, 1995); therefore, the absence of child-report
measures is a weakness of the present study. Child-report is particularly important as
children become older and more responsible for medication-taking. Additionally, child-
reports of barriers is important for constructing an accurate model on which to base future
interventions. This study completes an important step in the documentation of barriers to
childrens ART adherence, though child-report of adherence behaviors and barriers will
be important in future research.


61
When caregivers reported barriers, they were asked about the frequency of the
barrier occurrence and strength of the barrier in preventing medication-taking (e.g., How
often does this happen? and How often does this problem keep the child from getting
the medicine?). They were asked to answer each question using a 5-point Likert scale,
with 1 indicating not often at all and 5 indicating always. In response to the first
question, the mean frequency of occurrence rating was 3.6 (SD= 1.6), with a median of 4
and a mode of 5. In response to the second question, the mean strength rating was 1.6,
with both a median and mode of one (SD= 1.2).
Barriers Reported with the AI-HIV Versus the 24RI
The Wilcoxon Signed Ranks Test was conducted to determine whether any
significant differences existed in the number of barriers reported during the AI-HIV
versus the 24RI. Results showed that the number of barriers reported during the AI-HIV
was not significantly different from the number reported during the 24RI. Caregivers
cumulatively made 37 barrier reports during the AI-HIV while they made 32 cumulative
barrier reports during the three recall interviews.


18
Long duration of the regimen, and specific regimen attributes have also been
identified as important barriers to adherence. First, long duration of the regimen may
hinder adherence. It has been suggested that adherence is a greater problem with chronic
versus acute regimens (Rapoff, 1999). Relatedly, with childhood chronic illnesses
studies have found that adherence tends to decrease over time (Brownbridge & Fielding,
1994; Hanson, DeGuire, Schinkel, Henggeler, & Burghen, 1992; Hanson, Henggeler,
Harris, Burghen, & Moore, 1989; Jacobson et al., 1990). It may be that children begin to
resent medication taking and may decide to stop taking it regularly. Also, as families
adjust to the medication regimen, parents may begin to rely on their children to manage
their regimens independently, providing opportunities for children to skip, hide, or dump
doses.
Specific attributes of the regimen may also make adherence difficult.
Behaviorists have well established the organisms tendency to avoid aversive stimuli, and
this may be why complex regimen components, like chest physiotherapy, have been
associated with poor adherence (Passero, Remor, & Salomon, 1981). Behaviorists have
also established the organisms tendency to perform behaviors that are reinforcing, and
this may explain the difficulty children with diabetes experience with abstaining from
sweet and fatty foods (Glasgow, McCaul, & Schafer, 1986).
The Human Immunodeficiency Virus: A New Chronic Illness
Worldwide, more than 34 million people are infected with HIV; of those, 1.3
million are children (UNAIDS, 2000). Each day, almost 2,000 children are newly
infected with the virus. Specifically, in the United States more than 12,000 children are
living with HIV (UNAIDS, 2000) and more than 7,000 children are currently living with


41
African-American (66.7%), female (90.5%), and reportedly HIV uninfected (69.8%).
Most caregivers were adoptive parents (28.6%), biological mothers (27%), or
grandparents (23.8%).
Procedure
At each study site, the research nurse was asked to help the subject recruiter
complete the eligibility checklist (Appendix A) for each patient. Next, the recruiter asked
eligible parents/guardians several preliminary questions (Appendix A also) to ensure
eligibility. Multiple eligibility criteria were established for the study (Table 3).
Next, eligible participants underwent informed consent procedures in accordance
with ERB approval from all three study sites. If children were accompanied to the clinic
by foster parents or other caregivers, written informed consent was first obtained from the
parent or legal guardian. After written informed consent was obtained from the parent or
legal guardian, written informed consent was obtained from the non-parent caregiver
informant.
After completing informed consent procedures, caregivers were interviewed by
trained undergraduate psychology students or graduate students in clinical psychology,
using the AI-HIV (Appendix B). All interviewers were Caucasian. Next, the interviewer
completed the demographic questionnaire (Appendix C) and the telephone contact form
(Appendix D) with the caregiver. Finally, the interviewer completed the Medical Record
Report Form (Appendix E).
Following the clinic visit, parents/caregivers were contacted for telephone
follow-up with the 24RI (Appendix F). Beginning two weeks following the clinic visit,
parents were asked to complete the 24RI three times within a two-week period, including


50
A series of analyses was then conducted to test the fifth set of hypotheses. Due to
the non-normality of the knowledge variables (Table 4), knowledge of medication names
and dietary requirements were split between <100% and 100% knowledge, as it was
expected than any knowledge failure would relate to nonadherence categorization. Chi-
square analyses were conducted to determine the relationships of 24RI frequency
adherence and refill adherence to caregiver knowledge of medication names. Caregiver
knowledge of frequency was eliminated from analyses due to lack of variance. Chi-
square analysis was also employed to test the relationship between 24RI dietary
adherence and knowledge of dietary requirements. Correlations were then preformed to
test the relationship between 24RI interval deviance and interval knowledge.
For the sixth set of hypotheses, both parametric and nonparametric statistics were
used. To determine the relationship between 24RI frequency adherence and dosing
frequency, Kruskal-Wallis analysis was conducted due to the non-normality of the
dependent variable. Next, to determine the relationship between refill adherence and
dosing frequency, analysis of variance (ANOVA) was used, as the data were normally
distributed. Post-hoc analysis was conducted with the Scheffe test, as it is a conservative
post-hoc test. The Mann-Whitney U Test was used to determine the relationship between
24RI frequency adherence and medication formulation, as it was determined to be the
best test of differences between two groups when the dependent variable is
nonparametric. ANOVA was once again used when the dependent variable was changed
to refill adherence.
Barriers to adherence were recorded verbatim and categorized by the author.
Descriptive statistics regarding the barrier data were then examined. The Wilcoxon
50


19
AIDS, the advanced manifestation of the virus (CDC, 1996). Nationally, HIV has been
ranked as the seventh leading cause of death among children ages 1 to 4 (Ventura, Peters,
Martin, & Maurer, 1997). Improving mortality and morbidity among these children and
adolescents is contingent upon HIV-infected children receiving appropriate treatment
(Palella et al., 1998).
HIV is a retrovirus that is transmitted through exposure to infected blood and
sexual fluids, typically through needle sharing, unprotected sex, or mother-to-child
transmission (CDC, 1998). HIV disrupts the immune functioning of infected persons.
As a result, people with HIV have difficulty fighting off certain bacteria, viruses and
other microbes. Therefore, they are prone to develop opportunistic infections, which take
advantage of the compromised immune system. Over 100 microorganisms have been
identified as opportunistic, including: Candida allbicans, Varicella-Zoster Virus, herpes
simplex virus, measles, cytomegalovirus, congenital syphyilis, and pneumonocytic carini
pneumonia (PCP). If not treated successfully, such opporuntistic infections can lead to
AIDS-related fatalities (Butz et al., 1998, Palella et al., 1998). Also, HIV can lead to
neurological disease, neuropsychological impairment, cardiac abnormalities and
hematologic problems, among others (Scott & Sleasman, 1999). With no known cure, the
end result of HIV is death.
In the past, HIV was considered a debilitating illness characterized by rapid
deterioration and demise. However, the development of new drug therapies for HIV-
infected children and adults has led to increased life-spans for infected individuals
(Martino et al., 2001, Palella et al., 1998). Martino and colleagues (2001) found that
when children were prescribed at least 3 anti-retroviral medications, the risk of death


RESULTS
Caregiver Knowledge of the Prescribed Regimen
Descriptive data regarding caregiver knowledge can be found in Table 4. Results
of the AI-HIV showed that caregivers percent knowledge of prescribed medication
names ranged from 0% to 100%, with a mean of 86% (SD= .26). Thirty-eight percent
consulted their own lists of the childs medication information and 27% consulted the
medication display card, while only 39% responded without some aid. Sixty-seven
percent of caregivers were able to correctly identify all of their childrens medication
names. Only one caregiver was unable to correctly identify the prescribed medication
taking frequency for her childrens known medications. When the medication names
were known, the percent of medication dosages accurately reported by caregivers ranged
from 0% to 100%, with a mean of 83% (SD= .29). Sixty-nine percent of caregivers
correctly identified the dosages for all of their childrens ART medications. Additionally,
when the medication names were known and when the children were prescribed
medications with specific dietary requirements, 50% of caregivers correctly identified the
specific dietary requirements for their childrens medications.
Caregivers were not asked directly about their knowledge of the prescribed dosing
interval, as that information was assumed to be inherent in the question about frequency.
Nevertheless, caregivers were asked to state the times at which the children typically
receive each medication. The reported interval may be useful in understanding
52


43
Table 2. Additional Particinant Demographic Information
Variable
M
SD
Child Age
8.76
2.97
Years Prescribed ART
6.75
2.98
Caregiver Age
46.02
13.68
Household Yearly Income
23020.07
16751.57
Table 3. Inclusion Criteria
Caregiver Must Have:
1. Considered herself/himself one of the primary people responsible for the target childs
medication-taking
2. Reported that English was her/his primary language
Child Must Have Been:
1. Perinatally HIV-infected
2. Age 2-12 years
3. Prescribed the same anti-retroviral medication for the past 3 months
4. Living with the parent/caregiver during the last month
5. Planning to live with the parent/caregiver for the next three months
6. Not enrolled in an adherence intervention study
7. Not a sibling of an enrolled child


15
dietary behaviors (e.g., grams of carbohydrate, grams of fat, and total calories). Over a
three-month time period, moderate support was found for the stability of the measure,
with greater stability for blood-glucose testing and dietary behaviors compared to
exercise and injection behaviors (Freund et ah, 1991). Altogether, the findings suggest
that the 24-hour recall interview is a moderately reliable adherence measure with good
construct validity.
Recall interviews have received support for their utility in the research domain;
however, they lack clinical practicality. First, they require a trained interviewer who is
available in the evenings and on weekends. Although it may be relatively easy to find
such a person in an academic research center, it may be more difficult to find such a
person in the clinical environment. Second, they require that the patients and parents are
accessible by telephone in the evenings and on weekends. This may be practical for
middle class families with daily routines, but may be more difficult for disadvantaged
families who may be without phone service or for families in which parents work
evenings and weekends. These difficulties may explain why recall interviews have not
generally been integrated into most medical settings. Finally, while recall interviews can
detect children who are nonadherent, as they are conducted currently, recall interviews
provide little information about the barriers to adherence and potential targets for
intervention. While it is helpful to assess the extent of adherence, information regarding
appropriate targets for intervention would benefit both patients and providers.
While recall interviews may be impractical for clinical settings and do not identify
specific barriers to adherence, the structured or semi-structured interview is an
appropriate method to assess adherence and barriers to adherence in the clinical setting.


the sanctuary that was the Gainesville swing dancing community, and how my fellow
dancers kept my feet moving. I know that most of my friends and family members
cannot begin to understand the processes of graduate school and dissertation completion;
1 am grateful to them for trying to understand, and supporting me even when they did not
understand. With some confidence I can finally say that I am done with school!
IV


54
Table 4. Descriptive Statistics for Non-Demographic Study Variables
Variable
M
SD
Median
Mode
Range
Skewness
Kurtosis
% Caregiver
Knowledge of
Medication Names
(n= 62)
86.46
25.18
100.00
100.00
0-100
-2.24
4.96
% Caregiver
Knowledge of
Medication Dosing
(n= 59)
84.28
28.90
100.00
100.00
0-100
-1.90
2.8
Knowledge of
Interval Deviance
(n=51)
.73
.99
0
0
0-4
1.42
1.62
% 24RI Frequency
Adherence
(n= 55)
91.55
14.28
100.00
100.00
43-100
-2.01
3.5
24RI Interval
Deviance
(n=51)
1.00
1.03
.67
0
0- 4.67
1.41
2.47
% 24RI Dietary
Adherence
(n= 42)
75.00
30.05
84.00
100.00
0-100
1.02
.01
% Refill
Adherence
(n= 59)
62.05
35.74
70.85
100.00
0-100
-.63
-.99
% Nurse-reported
Adherence3
(n= 63)
5.92
1.22
6.00
7.00
2-7
-1.38
-1.58
Note. Data for percent caregiver knowledge of dosing frequency was not included in this
table, as all but one caregiver had correct dosing frequency knowledge. Data for percent
caregiver knowledge of dietary requirements was not included in this table because the
variable is dichotomous. aScale ranged from 1 not at all adherent to 7 always
adherent.


86
Belzer, M. E., Fuchs, D.N., Luftman, G.S., & Tucker, D.J. (1999). Antiretroviral
adherence issues among HIV-positive adolescents and young adults. Journal of
Adolescent Health, 25, 316-319.
Berg, J. S., Dischler, J., Wagner, D. J., Raia, J., & Palmer-Shevlin, N. (1993).
Medication compliance: A health care problem. The Annals of Pharmacotherapy,
27(supplement), 2-21.
Birkhead, G., Attaway, N.J., Strunck, R.C., Townsend, M.C., & Teutsch, S.
(1989). Investigation of a cluster of deaths of adolescents from asthma: Evidence
implicating inadequate treatment and poor patient adherence with medications. Journal
of Allergy and Clinical Immunology, 84, 484-491.
Boccuti, L., Celano, M., Geller, R.J., & Phillips, K.M. (1996). Development of a
scale to measure children's metered dose inhaler and spacer technique. Annals of
Allergy, Asthma, & Immunology, 77, 217-221.
Bogart, L. M., Catz, S.L., Kelly, J.A., Gray-Bemhardt, M.L., Hartmann, B.R.,
Ptto-Salaj, L.L., Hackl, K.L., & Bloom, F.R. (2000). Psychosocial issues in the era of
new AIDS treatments from the perspective pf persons living with HIV. Journal of Health
Psychology, 5(4), 500-516.
Brettle, R.P., Wilson, A., Povey, S., Morris, S., Morgan, R., Leen, C.L.,
Hutchinson, S., Lewis, S., & Gore, S. (1998). Combination therapy for HIV: the effect
on inpatient activity, morbidity and mortality of a cohort of patients. International
Journal of STD & AIDS. 9(2), 80-87.
Brownbridge, G., & Fielding, D. M. (1994). Psychosocial adjustment and
adherence to daily treatments regimens. Pediatric Nephrology, 8. 744-749.
Brownlee-Duffeck, M., Peterson, L., Simonds, J.F., Goldstein, D., Kilo, C., &
Iioette, S. (1987). The role of health beliefs in the regimen adherence and metabolic
control of adolescents and adults with diabetes mellitus. Journal of Consulting and
Clinical Psychology, 55(2), 139-144.
Butz, A. M., Joyner, M., Greenberg-Friedman, D., & Hutton, N. (1998). Primary
care for children with human immunodeficiency virus infection. Journal of Pediatric
Health Care, 12. 10-19.
Carpenter, C., Cooper, D.A., Fischl, M.A., Gatell, J.M., Gazzard, B.G., Hammer,
S.M., Hirsch, M.S., Jacobsen, D.M., Katzenstein, D.A., Montaner, J., Richman, D., Saag,
M.S., Schechter, M., Schooley, R.T., Thompson, M.A., Vella, S., Yeni, P.G., &
Volberding, P.A. (2000). Antiretroviral therapy in adults. Journal of the American
Medical Association, 283(31. 381-391.


46
child knows that he or she is HIV positive. Finally, questions about the family include
who lives in the home and who helps the child take his/her medication.
Telephone Contact Form (Appendix D)
This form was designed to collect the names of the caregiver and child, primary
and secondary telephone numbers, and best times to call. The information was used in
conducting the telephone adherence interviews.
Medical Record Report Form (Appendix E)
This researcher-completed form was designed by the author to assess the clinical
indications of the childs disease severity (CDC status and CD4 count), and the childs
currently prescribed medications. The form also includes a nurse rating of the familys
adherence to the childs regimen on a 7-point Likert scale (1 = not at all adherent, 7 =
always adherent); nurses are asked to rate adherence based on their clinical experiences
with each patient during the prior six months. In part, the form was modeled after the
Prescribed Treatment Form (PTF), which has been used to identify prescribed treatment
for children with CF (Quittner et al., 2000c). Two nurses and one physician from the
pediatric HIV clinic at the University of Florida reviewed the form in order to ensure
appropriateness of the wording and content.
24-hour Recall Interview (24RI Appendix FI
This measure was adapted from the 24RI used to assess adherence to diabetes
regimens (Freund, Johnson, Silverstein, & Thomas, 1991; Johnson, Silverstein,
Rosenbloom, Carter, & Cunningham, 1986; Johnson et ah, 1992). Although the 24RI for
diabetes includes both parent and child report, this study includes only parent report,
based on the difficulties children may have recalling details about medication-taking.


79
individual childhood chronic illnesses, and striving to understand the specific
environmental contexts that may promote or inhibit adherence.
Study Strengths
This study has many strengths. First, this was the first known study to assess
knowledge of the prescribed regimen among caregivers of children on ART. Thus, this
study fills a significant gap in the literature, and points clinicians and researchers to
valuable information about an important first step in adherence intervention: educating
children and caregivers about prescribed regimens.
Second, this study is one of few known studies to assess adherence to pediatric
ART regimens, and one of only two known studies to assess adherence to pediatric ART
regimens multidimensionally. Moreover, this study has successfully replicated the
findings of another recent study that measured adherence to pediatric ART regimens and
found over half to be less than 80% adherent to refills. Therefore, this study adds to the
literature by providing empirical support for the need for effective interventions to
improve childrens adherence to ART.
Third, this study represents the first known qualitative assessment of barriers to
childrens ART adherence. The resulting list of caregivers perceived barriers will be
helpful in developing a barriers chescklist that could be used to quickly assess barriers in
the clinical setting.
This study has also introduced two new tools for ART adherence assessment: the
AI-HIV and the 24RI. The utility of the AI-HIV in assessing knowledge and barriers and
the utility of the 24RI in assessing adherence to prescribed dosing intervals, dietary
requirements, and barriers have both been supported by this study. In the future, these


65
total reported typical intervals were at least 2 hours deviant from the prescribed intervals.
Reported intervals were particularly problematic for thrice daily dosing, in which 59% of
intervals for medications were at least 2 hours deviant from the prescribed interval.
These results suggest that caregivers may lack knowledge of appropriate dosing intervals
or they may have difficulty adhering to the specified intervals when those desired
intervals are known, particularly when medications are to be taken thrice daily.
It was also hypothesized that caregiver knowledge of the regimen would differ
significantly from reported typical regimen behavior during the previous 2 weeks, as
some caregivers would actively decide not to give their children some medications.
Given the nature of the AI-HIV, which first asked about typical behavior, and then asked
about knowledge of the prescribed regimen, it was thought that caregivers would feel free
to disclose any such differences. Nevertheless, caregivers failed to report any differences
at all between typical regimen behaviors and their knowledge of the prescribed regimen.
It seems likely that even with interviewer efforts to normalize adherence problems,
caregivers responded to the social desirability of the pediatric HIV clinic context. These
data are consistent with other self-report clinical data from caregivers of HIV-infected
children, in which a tendency to report 100% adherence was found, despite MEMS data
that suggests poorer adherence (Farley et al., 2002); however, these data suggest that
caregiver overestimation of their childs ART adherence may be more extensive than
caregiver and child overestimation of adherence to other chronic illness regimens
(Delamater, 2000; Festa et ah, 1992; Passero et ah, 1981; Quittner et ah, 2000c; Rapoff,
Lindsley, & Christopherson, 1985; Schoni et al, 1995). Perhaps the pediatric HIV clinic
environments foster reporting of socially desirable behavior more so than the diabetes or


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Clinical and
Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Stephen R. Boggs
Associate Professor of Clinical and
Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Alexandra L. Quittnr
Professor of Clinical and Health
Psychology


DATA ANALYSES
Data were analyzed using the Statistical Package for Social Sciences, Version
10.0 (SPSS 10.0). First, in order to test for demographic differences across sites, Chi-
square and Analyses of Variance (ANOVAs) were performed. Next, descriptive data
were reviewed and the skewness and kurtosis of the outcome variables were examined in
order to determine whether or not normality assumptions were met. Adherence
frequency data were viewed continuously and then split at cut-offs of 80% and 90%,
which were based on the research literature (Farley et ah, 2002; Paterson, 2000). The
first hypothesis was not tested statistically, as descriptive data were sufficient to support
the null hypothesis. For the second hypothesis, a Mann-Whitney U Test was conducted
in order to determine whether or not use of the display card was associated with
knowledge of medication names. This was determined to be the best test for determining
differences between the two groups when the test variable was nonparametric. The third
hypothesis was that all adherence measures would correlate. Pearson correlations were
performed, though due to the kurtosis of 24RI frequency adherence and inability to
transform this variable, it was left out of the correlation matrix. Chi-square analyses were
then conducted to test the fourth hypothesis. Adherence cut-offs of 80% and 90% were
both included in analyses, in order to determine the best cut-off, and viral load was split
to + 400 copies/ml, the highest number associated with undetectable levels of virus in the
blood.
49


34
special instructions, had a change in daily routine, did not have medications with
you, busy and did not want to stop to take medications, simply forgot, felt
depressed or overwhelmed, felt angry, depressed, or hopeless that you have to deal
with this, wanted to forget the whole thing, and side effects. After completing the
barriers chescklist open-ended discussions were held to discuss barriers, as well as
successful strategies. During the focus groups, several themes emerged. One theme was
lack of knowledge or false beliefs about the importance of taking the medicine regularly,
at certain intervals, and with specific food guidelines. Participants also pointed to the
complexity of the instructions, the poor taste, pill size, and side effects, as problematic.
Another theme was difficulty with the scheduling of medications, and particular concerns
were raised about taking medication in social situations. Finally, participants stated that
animosity toward the healthcare provider and confusion about the medication-taking
instructions the provider gave both negatively impact adherence.
Catz and colleagues (2000) examined regimen-specific barriers to adherence
among 72 mostly male HIV-infected adults. They developed the Barriers to Adherence
Checklist (BAC), a 56-item questionnaire that measures general barriers to HIV
treatment, and barriers specific to the medication regimen. Patients rate the degree to
which each item reflects their own experiences. The items were consistent with barriers
identified during qualitative analysis of patients on HAART (Bogart et al., 2000). The
most highly endorsed barriers were: treatment reminds me that I am HIV+, I do not
want others to know that I am HIV+, when I have clinic appointments, I forget to ask
some of my questions about treatment, I have trouble remembering names of medicines
and what they are for, and I do not like the way my medication makes my body feel.


10
understanding of the regimen, though the authors did not provide more specific
information regarding the knowledge assessment. Adherence behavior was measured via
self-report. Most who reported frequent nonadherence demonstrated poor regimen
knowledge, supporting the relationship between regimen knowledge and adherence.
Anthony, Paxton, Bines, and Phelan (1999) assessed maternal nutritional knowledge
specific to CF. No information was provided about how nutritional knowledge was
assessed, though disease-specific nutritional knowledge did positively relate to height,
weight, and self-reports of dietary adherence behaviors. While these findings do support a
relationship between knowledge and adherence behaviors, these studies fail to provide
important methodological information about their assessments.
Beck and colleagues (1980) were among the first to assess knowledge of
prescribed medication regimens. They used a questionnaire to assess child and caregiver
knowledge of the use, dose, side effects, and importance of each medication prescribed
after renal transplantation. A total knowledge score was given, based on the percentage
of questions answered correctly. At baseline, the mean knowledge score ranged from 7%
to 89%, with a mean of 60%, suggesting that a large portion of the sample lacked
significant knowledge about the medication regimen. Knowledge did not significantly
correlate with adherence behavior.
Gudas, Koocher, and Wypij (1991) assessed knowledge of childrens CF
regimens. They asked children ages 5-20 to state the medications that they take for CF,
and the impact that the medications, chest physiotherapy, and diet have on the disease.
Unfortunately, the researchers did not provide descriptive data about the degree of


47
This measure is also different in this study based on the resulting variables. When used
to assess adherence to diabetes regimens, the recall interview yields 13 variables related
to insulin injection, blood-glucose checking, diet and exercise. Since the HIV regimen
only requires medication-taking and related food and liquid intake, the recall interviews
conducted in this study yielded three variables: medication-taking frequency, interval
deviance, and dietary adherence. Caregiver reports were compared to the prescribed
regimen (as defined in the medical record), except reports regarding medication-specific
food and liquid intake, which were compared with instructions on published handouts
that list all current ART medications (Abbott Laboratories, 2001). The resulting
variables represent percents of agreement between the reported medication-related
behaviors and the prescribed regimens.
The final measure modification included an assessment of barriers to adherence.
At the end of the recall interview, informants were told that managing a medication
schedule is difficult, and many families have difficulties following the regimen. They
were asked to list the things that made it difficult to follow the physicians
recommendations regarding medication taking the previous day, and the things that kept
them from following the recommendations exactly the previous day. The reported
barriers were coded for purposes of analyses. This method was designed to capture
additional barriers that were not mentioned in the adherence interviews, as it allows
families to report actual barriers they faced during a given day.
As with the Adherence Interview-HIV, the interviews were conducted by trained
undergraduate assistants and graduate students. Interviewers underwent training of the
interview procedures and engaged in practice interviews. They were required to achieve


9
skill in urine testing and self-injection. The Diabetes Knowledge and Management Skills
Assessment Questionnaire (Brownlee-Duffeck et al., 1987) consists of 36 multiple-choice
questions that inquire about diet, insulin injections, insulin reactions, urine testing, foot
care, and general information about ketoacidosis and hyperglycemia. The medical
regimen for diabetes lends itself to this type of knowledge assessment, as the techniques
for self-care are generally consistent across individuals, and appropriate problem-solving
is consistent across individuals and is crucial to successful self-management. This type
of knowledge assessment is very specific to the regimen; however, it is difficult to
imagine how this type of assessment might be useful in assessing knowledge of
prescribed medication regimens for conditions such as HIV or cancer.
Similarly, the Metered Dose Inhaler Checklist (MDI) has been developed for
measuring inhaler/spacer technique among children with pulmonary disorders (Boccuti,
Celano, Geller, & Phillips, 1996; Celano, Geller, Phillips, & Ziman, 1998). The measure
was designed for providers to use as they observe patient use of inhalers/spacers. Six or
seven of the skills assessed are considered crucial for effective administration, while four
of the skills assessed are considered helpful to drug delivery; scores reflect these
differences. Similar to the diabetes tools, the MDI is a very promising technique for the
assessment of knowledge pertinent to pulmonary disease management, but does not apply
to assessment of prescribed medication regimen knowledge.
Several other studies have reportedly assessed child and caregiver regimen
knowledge, although they have failed to operationalize their knowledge assessment.
Tebbi and colleagues (1986) assessed regimen knowledge among 46 children with cancer
and their parents. Interview questions concerned knowledge of the medications and


66
CF clinic evironments. Providers in HIV clinics typically expect high levels of adherence
(80-95%; Farley et al., 2002; Paterson et ah, 2000), and often focus on adherence
extensively, because the consequences of even occasional deviance from the prescribed
regimen can be severe; unfortunately, the extensive focus on adherence may reinforce the
reporting of adherent behaviors, even when the reports are inaccurate. In the future, it
may be worthwhile to examine patient-provider interactions to assess the extent to which
providers may inadvertently reinforce overreporting of adherence behaviors.
Additionally, it may be worthwhile to modify the self-report adherence assessment of the
AI-HIV in order to promote more honest responding.
Another study aim was to document the utility of the medication display card for
caregivers identification of medications during the AI-HIV. Caregivers were given the
option to use the display card if they had difficulty generating medication names, and
27% of caregivers exercised that option. It was hypothesized that knowledge of
medication names would not differ significantly based on use of the display card, because
the use of the display card would eliminate most problems in medication identification.
This hypothesis was supported by the present study, and suggests that the display card is
a useful component of the AI-HIV, as it helps to trigger recall, yet fails to lead the
informant to a particular response.
Adherence Rates
The study also aimed to document rates of adherence to prescribed medication
taking frequencies, intervals, and medication-specific dietary requirements. Regarding
medication-taking frequencies, results vary greatly based on the mode of assessment,
with higher adherence ratings resulting from the 24RI than pharmacy refill, and generally


106
20. How many people live in the home with you and your child? Please list all:
Initials
Age
Relationship to Child
22. Does someone help make sure that your child takes his/her
medicine?
Yes
1. List all people who help. (Examples: you, your child, your spouse, relative,
neighbor, child's sister, school nurse, etc.)
2. Circle the word that best describes how often that person helps.
(For example: Rarely= once or twice per month
Often- at least once per week
Always= at least once per day)
No-Leave below blank
Initials
Age
Relationship to Child
How often does
he or she help?
Rarely
Often
Always
Rarely
Often
Always
Rarely
Often
Always
Rarely
Often
Always


specific dietary requirements for at least one medication. Adherence to medication
taking frequency varied by assessment modality. Results of the 24RI suggest that 87% of
children were at least 80% adherent; however, results of the pharmacy refill history
suggest that only 46% of children were at least 80% adherent. When adherence to the
prescribed interval was examined with the 24RI, 47.3% of doses given were deviant from
the prescribed interval by at least one hour, while 17.3% of doses given were deviant by
at least 2 hours. Results of the 24RI also suggest that the average child was adherent to
medication-specific dietary requirements 75% of the time. Twenty-two different barriers
were reported and fall into three general categories: a) medication-specific attributes,
such as the size or taste of pills; b) problems with scheduling or routine; or c) problems
with the child resisting, refusing, or hiding the medication.
This study supports the assertion that a significant proportion of HIV-infected
children are not receiving their medication exactly as it is prescribed. The study also
suggests that a large portion of caregivers lack knowledge about their childs HIV
medication regimen. Implications for adherence-related interventions are discussed.
viii


APPENDIX C
DEMOGRAPHIC QUESTIONNAIRE
Date completed: /
month / year
Say: The next questions are about you and your child. Your answers will help us learn
more about your family.
Questions about Your Child
Please answer the following questions about your child with the best answer:
1. Childs Date of Birth: _____
2. Childs Gender: MALE
FEMALE
3. Childs Ethnic background: African American
White/ Caucasian
Asian
Hispanic
Native American
OTHER:
4.How old was your child when the doctors first said he/she had HIV?
years old
5.How did your child get HIV? was bom with HIV
blood transfusion
sexual contact
I dont know
other (explain):
103


Place a check in the box next to all medications prescribed for this patient. Then, write in dosing information and check one box for
frequency, and check all applicable boxes under special instructions.
Medication
name
Other names
Formulation
Dosing
(in tabs, mg
or cc-
indicate)
Frequency
Special Instructions
Abacavir
succinate
Ziagen, 1592U89
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Qq4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Amprenavir
Agenerase, 141W94
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Combivir
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Delavirdine
Rescriptor
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 q 6 q 8 q 12
take with food
take on empty stomach
no fatty foods
other:
Didanosine
ddl, Videx,
Dideoxyinosine
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Efavirenz
Sustiva, DMP-266
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:


Medication name
Other names
Formulation
Dosing
Frequency
Special Instructions
Zalcitabine
ddC, Hivid,
Dideoxycytidine,
R024-2027
capsules/
tablets
powder
liquid
q.d. Ob.i.d. t.i.d. q.i.d.
Oq4 Qq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Zidovudine
AZT, ZDV, Retrovir,
Azidothymadine
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Other Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Other Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Other Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Other Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 q 12
take with food
take on empty stomach
no fatty foods
other:


92
Murphy, B.C. & Dillon, C. (1998). Interviewing in Action: Process and Practice.
Pacific Grove, CA: Brooks/ Cole.
Murphy, D.A., Roberts, K.J., Martin, D.J., Marelich, W., & Hoffman, D. (2000).
Barriers to antiretroviral adherence among HIV-infected adults. AIDS Patient Care and
STDs, 14(1)47-58.
Newacheck, P.W., McManus, M., Fox, H.B. Hung, Y.Y., & Halfon, N. (2000).
Access to health care for children with special health care needs. Pediatrics, 105(4), 760-
766.
O'Reilly, K. O. (1995). The role of qualitative research in the global programme
on AIDS at the World Health Organization. In E. Y. Lambert, R.S. Ashery, & R.H.
Needle (Eds.), Qualitative Methods of Research in Drug Abuse and HIV Research (Vol.
157, pp. 27-38). Rockville, MD: National Institute of Drug Abuse, Division of
Epidemiology and Prevention Research.
Ostrop, N.J., & Gill, M.J. (2000). Antiretroviral medication adherence and
persistence with respect to adherence tool usage. AIDS Patient Care and STDs, 14(7),
351-358.
Paes, A. H., Bakker, A., & Soe-Agnie, C. J. (1998). Measurement of patient
compliance. Pharmacology World Science, 20, 7-77.
Palella, F.J., Delaney, K.M., Moorman, A.C., Loveless, M.O., Fhrer, J., Satten,
G.A., Aschman, D.J., Holmberg, S.D., & The HIV Outpatient Study Investigators
(1998). Declining morbidity and mortality among patients with advanced human
immunodeficiency virus infection. New England Journal of Medicine, 338, 853-860.
Parienti, J.J., Vemdon, R., & Bazin, C. (2001). The Pills Identification Test: A
tool to assess adherence to antiretroviral therapy. Journal of the American Medical
Association, 285(4), 412.
Passero, M.A., Remor, B., & Salomon, J. (1981). Patient-reported compliance
with cystic fibrosis therapy. Clinical Pediatrics. 20. 264-268.
Paterson, D.L., Swindells, S., Mohr, J., Brester, M., Vergis, E., Squier, C.,
Wagener, M., & Singh, N. (2000). Adherence to protease inhibitor therapy and
outcomes in patients with HIV infection. Annals of Internal Medicine. 133.21-30.
Quittner, A.L., Drotar, D.A., Ievers-Landis, C. A., & Seidner, D. (2000a).
Behavioral family systems therapy (BFST) for teenagers with cystic fibrosis and their
parents: Treatment and implementation manual. Unpublished Manual.


Medication name
Other names
Formulation
Dosing
Frequency
Special Instructions
Indinavir
Crixivan, MK-639, L-
735,524
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Lamivudine
3TC, Epivir
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Nelfinavir
Viracept, AG-1343
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Nevirapine
NVP, Viramune,
BIRG-587
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Ritonavir
Norvir, ABT-538
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Saquinavir
Invirase, Fortovase
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Qq6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Stavudine
d4T, Zerit
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 q 6 q 8 q 12
take with food
take on empty stomach
no fatty foods
other:


71
It was also hypothesized that adherence frequency measures would significantly
relate to prescribed dosing frequency and formulation. Refill adherence was significantly
associated with prescribed dosing frequency, suggesting that medications prescribed for
twice daily dosing are significantly more likely to be refilled than medications prescribed
for once daily dosing. This result is interesting, given that once daily dosing is often
preferred over twice and thrice daily dosing (Rosenbach, Allison, & Nadler, 2002). The
small number of children in the study who were prescribed thrice daily dosing likely
explains why thrice daily dosing was not negatively associated with adherence, though
the higher refill rates among twice daily dosing versus once daily dosing are particularly
surprising, as all children with once daily medications also have twice daily medications;
therefore, it would seem easy for children or caregivers just to add the extra pill or pills
once a day, or at least refill the once daily medications while refilling twice daily
medications. One plausible explanation for this finding is that once-daily medication, at
least when taken along with twice-daily medication is, in fact, easy to forget, as it is not
taken as frequently as the other medication. Another explanation is related to the dietary
requirements. Common once-daily medications are didanosine and efavirenz.
Didanosine is to be taken on an empty stomach, at least 30 minutes before or 2 hours
after eating, and efavirenz is not to be taken with a high-fat meal (Abbott Laboratories,
2001). It may be that sometimes it is too difficult to follow the dietary requirements, and
therefore, an entire dose is missed (Roberts & Mann, 2000).
Both 24RI frequency and refill adherence were significantly related to medication
formulation, though the results were in opposite directions. 24RI frequency adherence
was generally higher for pills, though refill adherence was higher for liquid and powder


30
McNabb et al., 2001; Paterson et al., 2000; Svarstad et al., 1999). This study also
demonstrates that despite previous research suggesting that providers overestimate
adherence (Rand & Weeks, 1998), some providers may be particularly astute at assessing
patient adherence.
Taken together, the reviewed studies suggest that near-perfect adherence may be
difficult to achieve among children and adolescents on ART regimens. If 80% adherence
or greater is necessary to achieve optimal therapeutic benefit (Farley et al., 2002) the
above results suggest that a large percentage of children may not be achieving maximum
benefit from ART. Although theses studies are useful in estimating the prevalence of
poor adherence, additional studies with larger sample sizes and multiple assessment
modalities (Quittner, Espelage, Ievers-Landis, & Drotar, 2000b) are necessary in order to
understand the extent and nature of adherence problems among families of children on
AJR.T. Only with a systemic understanding of the problem can researchers begin to
develop specific, targeted interventions that are effective in promoting adherence in this
population.
Barriers to ART Adherence
Understanding barriers to adherence may be essential in understanding the
problem of poor adherence. Prior to developing interventions to increase adherence
among HIV-infected children, researchers must identify and understand the obstacles to
adherence. To date, few studies have attempted to identify these obstacles. Though
barriers are assessed as part of the Pediatric AIDS Clinical Trials Group (PACTG)
protocols, few known studies have examined barriers among younger HIV-infected
children and their families. Furthermore, only one known study has examined barriers to


69
thus, lower CDC number classifications). These results suggest that caregivers of
children with greater CDC number classification may feel greater stimulus demand to
report more adherent behaviors than caregivers of children with lower CDC number
classification, even when such reporting is inconsistent with refill behavior.
Viral load was not significantly different based on adherence categorization, a
finding that is inconsistent with some findings which have shown that self-reported
adherence was significantly and negatively associated with viral load among adults
(Duong et al., 2001; Knobel et al., 2002; Mannheimer et al., 2002; Wagner et al., 2001;
Walsh, Mandalia, & Gazzard, 2002) and among children (Katko et al., 2001). Several
factors may help to explain why viral load was not significantly related to adherence in
this study. First, viral load is impacted by adherence over time, takes time to adjust to
changes in adherence, and is also highly impacted by drug resistance, which can occur
over time even when adherence is generally consistent. Also, viral load was only
assessed at baseline; therefore, no follow-up data were available in order to compare
adherence behavior to viral load as an outcome variable. Another possibility may be that
the cutoff of 90% adherence may not be sufficient for virologic success.
The researcher also hypothesized that adherence would significantly and
positively relate to knowledge. It was hypothesized that 24RI frequency and refill
adherence would relate to caregiver knowledge of medication names and dosing
frequencies. Although neither adherence measure was significantly associated with
dosing frequency knowledge, caregivers with 100% knowledge of medication names
were significantly more likely to have > 90% refill adherence than those with imperfect
knowledge of medication names. This is an important finding, as it suggests that


55
Table 5. Typical Percent of Deviance from Prescribed Intervals
Deviance Amount
% of twice daily doses
% of thrice daily doses
% of all doses
(n= 166)
(n= 27)
(n= 193)
Less than 1 hour
57.2
25.9
53.7
1-2 hours
25.9
14.8
25.5
2-3 hours
9.6
22.2
9.6
3-4 hours
4.8
25.9
7.4
> 4 hours
2.4
11.1
3.7


60
prescribed dosing frequency. The result was non-significant, suggesting that 24RI
frequency adherence did not differ based on prescribed dosing frequency. Next,
univariate Analysis of Variance (ANOVA) was conducted in order to determine the
relationship between refill adherence and prescribed dosing frequency. Results showed
that prescribed dosing frequency was significantly and positively related to refill
adherence, F(184, 2)= 3.89, p < .05. Post-hoc analyses using the Scheffe test showed that
participants taking medications with twice daily dosing had higher refill rates than
participants taking medications with once daily dosing.
Relationship between Adherence and Medication Formulation
Mann-Whitney U analysis was employed to determine the relationship between
24RI frequency adherence and medication formulation. Results showed that formulation
was significantly related to 24RI frequency adherence (p< .01), such that medications in
pill form were significantly more likely to have high 24RI frequency adherence. These
results contrast with results of Univariate Analysis of Variance (ANOVA), which showed
that formulation was significantly and positively related to refill adherence, F (185, 2)=
4.9, p< .05. Examination of the means indicates that medications in liquid or powder
formulation were significantly more likely to have high refill rates than medications in
pill form.
Barriers to Adherence
During the AI-HIV, caregivers reported an average of .89 barriers to their
childrens adherence (SD = 1.11), with a median of one barrier, and a mode of zero
barriers. Barriers were reported verbatim and then categorized. For a complete list of
barriers reported, refer to Table 7.


33
Several studies have conducted quantitative assessments of barriers, based on
previous qualitative studies. Mannheimer and colleagues (2002) asked over 1,000
mostly-male HIV-infected adults to complete a 10-item medication barriers chescklist.
Forgetting to take the medication was the most highly endorsed barrier, followed by
being away from home, experiencing side effects, and having difficulty taking the pills at
specified times. The frequency of reasons was stable over 12 months.
Ammassari and others (2001) administered a 12-item barriers chescklist to 358
mostly-male HIV-infected adults. Participants were asked to rate the importance of each
barrier in the missing or discontinuing of drugs on a four-point Likert scale. The most
highly endorsed barriers related to the number of pills, concern about future side effects,
being away from home, and concern about others gaining awareness of their medication
taking and disease status.
Walsh and colleagues (2001) assessed barriers to adherence among 157 HIV-
infected individuals ages 16 and older. Participants completed a 20-item questionnaire
that listed barriers to adherence and requested them to rate on a five-point Likert scale
how often each item led to missed doses. Items with the highest frequency ratings related
to forgetting, oversleeping, eating at the wrong time, being busy, being in social
situations, side effects, feeling ill, and not wanting to be reminded of HIV status.
Murphy, Roberts, Martin, Marelich, and Hoffman (2000) asked 39 adults on ART
to complete a 23-item barriers chescklist based on the Aids Clinical Trials Group
(ACTG) Baseline Adherence Questionnaire. Participants were asked to endorse an item
if it prevented them from taking their medications as prescribed. The most highly
endorsed items were: slept through the dose time, had problems taking pills with


45
threatening manner, in order to promote honest responding. Then, the informant is asked
to recall the recommended regimen.
For each medication in the recommended regimen, informants are asked how
often, when, and within what special food and drink intake guidelines it was
recommended to be taken. Also for each medication, informants are asked to identify
difficulties with taking each medication, and those things that sometimes keep them from
taking the medication as recommended. This allows for identification of which
medications may have the most negative medication attributes, and may serve as a cue to
remind informants of specific barriers. After the informants discuss all of the
medications they believe to be in the childs regimen, they are asked to identify other
things, not yet mentioned, that may make medication-taking hard or may keep the child
from taking his/her medication as recommended.
Interviews were conducted by trained undergraduate assistants and graduate
students. Interviewers underwent training with the interview procedures and engaged in
practice interviews. They were required to achieve 80% agreement with the authors
coding of a practice interview before conducting interviews for the study.
Demographic Questionnaire (Appendix Cl
This questionnaire was designed to gather demographic information on the parent
or caregiver, child, and family. The questionnaire asks the caregiver for the date of birth,
gender, and ethnicity of both the child and caregiver. Questions assess the caregivers
marital status, education, work status, and household income, as well as HIV status and
mode of transmission for those who are HIV positive. Questions about the child include:
the age of diagnosis, mode of transmission, date of ART initiation, and whether or not the


27
As part of another large multi-center study, Stone and others (2001) used a
structured interview to assess regimen knowledge among 289 HIV-infected women.
Participants were presented with a card bearing photographs of all current FDA-approved
ART medications and corresponding medication names. They were then asked whether
or not they were currently taking each of the medications. Participants were then asked
to report the dosing frequency and dietary requirements related to each endorsed
medication. Sixty-three percent of the patients reported correct information about dosing
frequency and dietary requirements for each endorsed medication. Of those with
incorrect information, 8% reported incorrect information about both frequency and
dietary requirements, 17% reported incorrect information about frequency only, and 12%
reported incorrect information solely about the dietary requirements. The relationship
between regimen knowledge and adherence behavior was not reported.
Another multi-site study used a computer-assisted structured interview (CASI) to
assess regimen knowledge among HIV-infected adults (Bangsberg, Bronstone, &
Hoffman, 2002). One hundred and forty-one patients were presented with a computer
screen bearing names and images of ART medications. Patients were instructed to click
on the images of medications currently included in their regimen. The program then
prompted patients to provide information about the number of pills per dose, number of
doses per day, and associated dietary requirements for each medication. Adherence was
assessed in a 3-day recall format as part of the CASI. More than half of all patients made
at least one error regarding their medication regimens. Fourteen percent failed to identify
at least one prescribed medication, 18% incorrectly reported the number of daily doses
for at least one prescribed medication, and 19% incorrectly reported the dietary


23
pill count adherence significantly predicted viral load, such that a 10% difference in
adherence was associated with a doubly increased viral load, suggesting that 90%
adherence or greater is necessary for viral suppression. McNabb and colleagues (2001)
also used the MEMS to assess adherence among 40 mostly-male HIV-infected adults.
Results of this study are similar to the results of Bansberg et al (2000), in suggesting that
> 90% adherence is necessary for full viral suppression. While these studies suggest that
at least 90% adherence is necessary for virologic success, one study among a pediatric
population suggests that lower levels may be sufficient (Farley et al., 2002). Farley and
colleagues used MEMS and pharmacy refill data to assess adherence to one ART drug
over a 6-month period. MEMS data alone suggested that an adherence rate of > 80% was
robustly associated with virologic success, and when pharmacy refill data were combined
with MEMS data, the rate of > 80% adherence was even more strongly associated with
virologic success. Thus, among children, at least 80% adherence may be necessary for
optimal health.
Poor adherence to the HIV regimen may also have severe health consequences
such as the development of fatal opportunistic infections and drug resistance. When the
virus is not completely suppressed, potentially fatal opportunistic infections may develop.
If not treated successfully, opportunistic infections such as PCP lead to AIDS-related
fatalities. Moreover, when medications do not fully suppress the virus, the virus rapidly
mutates into strains that are resistant to drug therapies (Fatkenheuer et al., 1997).
Mutations occur naturally with HIV, but are prevented when ART stops viral replication
completely. When the virus is partially suppressed, as occurs with intermittent or
incomplete adherence, drugs may render common strains inactive, but may not prevent


35
Patients who reported at least weekly missed doses reported a significantly higher
number of barriers than those who were more adherent. This study is different from the
other studies, as endorsement of barriers was not necessarily based on the relationship
between items and missed doses, only on the extent to which items reflected participants
experiences.
Altogether, several studies have examined barriers to adherence among HIV-
infected adults. Several appropriate tools have been developed for identifying barriers to
ART adherence among adults, and one tool has been developed for pediatrics. Many of
the barriers assessed by these measures may be relevant to pediatric HIV regimens;
however, in order to determine the nature and extent of the adherence barriers for
families of children on ART, an extensive examination of the barriers facing these
families is necessary. Given the lack of research examining the adherence barriers to
pediatric ART regimens, an in-depth assessment of medication-taking behaviors and
barriers is warranted. Quantitative assessment methods are insufficient for studying this
problem, as they may fail to include important barriers. Alternatively, qualitative
assessment is indicated. Such a thorough approach is essential in eliciting the families
input regarding their experience of the HIV-specific disease and regimen behaviors, and
in identifying those factors that may obstruct disease management. Furthermore,
qualitative assessment has been implicated by the World Health Organization (O'Reilly,
1995) as an important methodology within HIV/AIDS research to understand behaviors
in their context and thereby to identify barriers and potential facilitators of behavior
change (p. 34). Thus, qualitative methods are necessary in order to understand the


105
14. Your Education (check highest grade or year completed)
gradeschool
some high school
graduated high school
GED
technical/ vocational training
some college
Associate degree
Bachelors degree
some graduate training
Masters degree
Doctoral Degree (J.D., M.D., Ph.D., or equivalent)
15. Your Work Status: Employed Full Time
Employed Part-Time
Unemployed
Disabled
Retired
Homemaker
Student, unemployed
Student, employed part-time
Student, employed full time
16. Your Job title (if unemployed or retired, list most recent job title):
17.Please specify the monthly income (in dollars) in your household.
$
18. Are you HIV positive? Yes
No
19. If you are HIV positive, how did you get HIV (or how do you think you got HIV)?
I am not HIV positive
I was bom with HIV
blood transfusion
sexual contact
needle sharing/ drug use
I dont know
other (explain):


74
that make childrens medication-taking difficult, ultimately, these barriers rarely prevent
children from receiving doses of medication.
The present study fails to support the hypothesis that caregivers would report
more barriers during the 24RI procedure than during the AI-HIV. It was thought that the
24RI would be more effective than the AI-HIV in eliciting barrier reports because the
diary format of the 24RI would help conjure memories of recent barriers. Although the
24RI did not result in a greater number of barriers reported, qualitative analysis showed
that individual caregivers tended to report different barriers during the 24RI than during
the AI-HIV. Therefore, it can be concluded that both the daily diary format and the
clinical interview format are effective methods for obtaining information about barriers to
adherence.
Implications
Overall, this study suggests that the AI-HIV, 24RI, and pharmacy refill histories
are valuable means of assessing adherence to ART regimens among HIV-infected
children. This study demonstrates that the AI-HIV is a useful tool for measuring
knowledge of the prescribed regimen, including knowledge about medication names,
dosages, dosing frequencies, dosing intervals, and dietary requirements, and for
collecting information about barriers to adherence. In particular, using the AI-HIV to
assess knowledge of prescribed medication names may help clinicians to detect overall
adherence problems, and using the AI-HIV to assess knowledge of prescribed dosing
intervals may help clinicians detect problems with dose timing. Nevertheless, the AI-
HIV may require some modifications in order to improve its utility. Although the format
of inquiring first about the typical adherence behaviors and then about the prescribed


78
children with other chronic illnesses may not share that fear. Moreover, caregivers who
are HIV-infected themselves may fail to report lower levels of adherence due to
additional concerns that reports of poor adherence may reflect their poor adherence to
their own regimens. Another difference may be that caregivers of HIV-infected children,
a group that includes more African-Americans than some other chronic illesses, may be
uncomfortable reporting adherence problems to Caucasian interviewers. Secondly, even
the most conservative results of the present study suggest that children on ART achieve
higher adherence levels than children with other chronic illness regimens (Rapoff, 1999).
Children with HIV often live in economic and environment conditions that are less stable,
safe, and secure than environments of children with other chronic illnesses, and some
HIV-infected children live with HIV-infected caregivers who have their own medical
concerns, or with foster parents or distant family members who may have other
significant demands on their time (e.g., their own medical regimens and regimens of
other children in the home); yet these results imply that HIV children and their families,
and perhaps their providers, are actually doing a better job adhering than those who may
have fewer psychosocial stressors. The HIV-infected group may do better because the
regimen includes fewer components than regimens for diabetes and CF, because the
patient-provider communications about the regimen are more constructive, or due to
some combination of these and other factors. Regardless of the reasons for the better
adherence in the HIV-infected group, this is an important observation as it demonstrates
the resilience and adaptability of HIV-infected children and their families. Furthermore,
taken together, the results emphasize the importance of examining adherence to


99
Many different things can make it hard for children to get the medicines
their doctors recommend. What kinds of things make it hard for your child to
take like the doctor recommended? Record the
response (s), and show the respondent the rating card.
Ask the respondent: How often does this happen?
And then: How often does this problem keep your child from getting the
medicine like the doctor recommended?
What other things keep your child from getting
when and how the doctor suggested? Record the response, and say:
Now, go to the next medication on the list and repeat this set of questions for each
medication on the list.
When you reach the end of the recommended medication list, say:
The information you have given me will be very helpful. I am beginning to understand
some of the things that make it hard for families with children who have HIV. Are
there other things that keep your child from getting the medications the doctor
recommended? Record the response, and say:
Wow! The information you have given me today has really helped me to learn more
about what it is like for your family to manage s illness. Thank you
so much for helping us out! I have just a few more questions for you, and then well
be through. Read the caregiver demographic questionnaire to the patient.


APPENDIX D
TELEPHONE CONTACT FORM
Subject #:
Primary Phone Number: _( )
(The best number for us to reach you.)
When can we expect to reach you at this number? (circle all that
apply)
Weekday mornings
Weekday afternoons
Weekday evenings
What are the BEST times for us I
Weekend mornings
Weekend afternoons
Weekend evenings
reach you at this number?
Secondary Phone Number: _( )
(The second best number for us to reach you. It can be a work number, cell phone, or
friend's house.)
When can we expect to reach you at this number? (circle all that
apply)
Weekday mornings
Weekday afternoons
Weekday evenings
Weekend mornings
Weekend afternoons
Weekend evenings
What are the BEST times for us to reach you at this number?
When we call, we will say: Hello, this is calling from Project FRAME. If
you are not at that number, we will try back later.
107


Medication name
Other names
Formulation
Dosing
Frequency
Special Instructions
Zalcitabine
ddC, Hivid,
Dideoxycytidine,
R024-2027
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Zidovudine
AZT, ZDV, Retrovir,
Azidothymadine
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q 4 q 6 Qq8 Qql2
take with food
take on empty stomach
no fatty foods
other:
Other
Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q 4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Other
Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 q 12
take with food
take on empty stomach
no fatty foods
other:
Other
Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
Oq4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:
Other
Antiretroviral
Medication
capsules/
tablets
powder
liquid
q.d. b.i.d. t.i.d. q.i.d.
q4 Oq6 Oq8 Oql2
take with food
take on empty stomach
no fatty foods
other:


67
high adherence ratings resulting from nurse-report. The rate from the 24RI (92% average
adherence to frequency) suggests lower adherence than caregiver self-reported adherence
in Farley et al. (2002), which found that all caregivers denied that their child had missed
any doses within the previous 3 days; however, the present refill rates (75% average
adherence) suggest higher adherence than the Farley et al. (2002) results of pharmacy
refill histories and MEMS monitoring, which found average adherence ratings of 46%
and 54%, respectively. Thus, the results of the present study suggest that when
compared to pharmacy refill histories, the 24RI tends to overestimate adherence to ART
frequency, but it may overestimate ART adherence less than other methods of self-report.
When these results are compared to child adherence to diabetes and CF regimens, these
results suggest that caregivers of HIV-infected children may overestimate adherence
significantly more than caregivers of children with other chronic illnesses (Quittner et al.,
2000b). As mentioned previously, the expectations of high adherence and extensive
focus on adherence in pediatric HIV clinics may inhibit accurate self-report.
The 24RI also provided data about dosing intervals and dietary adherence. Data
showed that a significant portion of the sample (17.3%) had an average dosing deviation
of > 2 hours from the prescribed interval, while a much larger portion (47.3%) had an
average deviation of > 1 hour. These results are consistent with the typical intervals
reported during the AI-HIV in this study.
The third study hypothesis was that all adherence measures would correlate
significantly and positively. The results of the study fail to support this hypothesis, as
only pharmacy refill and nurse-report ratings were significantly and positively correlated.
Despite the hypothesis, it is somewhat intuitive that 24RI interval and dietary adherence


6
regards to other medications during the same interval (Hall et al., 1998). Unfortunately,
much of the literature has examined adherence unidimensionally, failing to assess each
aspect of the regimen and each metric individually.
The Implications of Poor Adherence
The impact of poor adherence in children with chronic illnesses is potentially
severe and long-standing. Poor adherence adversely affects health care costs, clinical
decision-making, and conclusions drawn from clinical trials, in addition to health
outcomes (Rapoff, 1999). When patients are nonadherent to medical regimens, increased
health care costs are incurred and health care resources are wasted. In the United States,
the health care costs of poor adherence in both children and adults is estimated at $100
billion per year (Berg, Dischler, Wagner, Raja, & Palmer-Shelvin, 1993). Costs include
wasted medication and unused therapeutic equipment, and increased morbidity leading to
more frequent clinic appointments, emergency care, and hospitalization.
Poor adherence also adversely affects clinical decision-making. Providers may
make clinical decisions based on the assumption that a patient is adherent, even when the
patient is not (Rapoff, 1999). This may lead to the provision of increased time and
medical resources, in the form of additional medication, therapies, or medical procedures.
In clinical trials, poor adherence may lead to an underestimation of the
effectiveness of medications, known as the compliance bias (Feinstein,1974). If a person
involved in a clinical trial fails to take medications appropriately but does not inform the
investigator of her poor adherence, the investigator may assume that the medications
were taken as prescribed. If several participants in a clinical trial repeat this behavior, the
investigator may erroneously conclude that the drug is ineffective or that higher dosages


104
6. Does your child know that he/she has HIV? QYes
No
7. When did your child first begin taking antiretroviral medication for HIV?
/
month / year
8. How many times was your child hospitalized in the last 6 months?
Questions about You
Please answer the following questions about you with the best answer:
9. Your Date of Birth:
/
month/year
10. Your Gender:
MALE
FEMALE
11. Your Ethnic background: African American
White/ Caucasian
Asian
Hispanic
Native American
OTHER:
12. Your relationship to the child: Mother (biological)
13. Your Marital Status:
Father (biological)
Grandparent
Adoptive parent
Foster parent
Relative
Family friend
Single
Living with a partner
Married
Separated
Divorced
Widowed


56
Utility of Medication Display Card
Twenty-seven percent of caregivers used the medication display card to identify
the names of prescribed medication. A Mann-Whitney U Test was conducted in order to
determine whether or not the use of the display card was associated with knowledge of
the medication names. Display card use was not significantly related to knowledge of
medication names.
Adherence Rates
Descriptive statistics based on adherence rates are presented in Table 4. As
measured with the 24RI, adherence to the prescribed medication-taking frequency ranged
from 43% to 100% (M= .92, SD= .14), and the modal adherence to frequency was 100%.
When the sample was split based on a 24RI frequency adherence cut-off of 80%, 87% of
participants were considered adherent. When a cut-off of 90% was used, 75% were
considered adherent. For those doses received, average deviation from the prescribed
dosing intervals ranged from 0 to 4.67 hours, with a mean of 1 hour, a median of .67
hours (SD= 1.03), and a mode of zero hours. Results showed that 47.3% of the doses
given were deviant from prescribed interval by at least one hour, and 17.3% were deviant
by at least 2 hours. Also for those doses received and for those medications with specific
dietary requirements, the average percent adherence to dietary requirements ranged from
0% to 100%, with a mean of 75% (SD= .30), median of 84%, and mode of 100%. These
results varied from the results of the pharmacy refill histories, which showed that
adherence to pharmacy refills ranged from 0% to 100%, with a mean of 62% (SD= .36),
median of 71%, and mode of 100%. When the sample was split based on a pharmacy


20
decreased by 70%. Now that HIV-infected children are living into adolescence and
young-adulthood, the issue of adherence to their medication regimens is becoming
increasingly more important.
Typically, the preferred pediatric antiretroviral medication regimen consists of
combination ART, specifically, highly active antiretroviral therapy (HAART; Scott &
Sleasman, 1999). With standard HAART regimens, children are prescribed at least three
medications to be taken two or three times per day. Children may take their medication
in pill form or in liquid doses either because they are unable to swallow pills or because
some pills are not available in small doses appropriate to body weight. Many of the
antiretroviral medications have unpalatable taste and gastrointestinal side effects. Side
effects for antiretroviral therapy (ART) include gastrointestinal discomfort, nausea,
vomiting, diarrhea, headache, peripheral neuropathy, parasthesis, rash, mouth ulcers, dry
skin, and others, including organ damage (Carpenter et al., 2000). Side effects range in
frequency, duration, and severity. To combat side effects, as well as opportunistic
infections and other complications, healthcare providers may prescribe additional
medications (Kelly, Otto-Slaj, Sikkema, Pinkerton, & Bloom, 1998), such as Zantac,
Septra, and a variety of others.
With ART and other drugs to combat side effects, infections, and complications,
medication regimens for HIV-infected children can be complex. Many of the
medications have special instructions that are difficult to follow. For example, to ensure
proper absorption, some medications should be taken with food, while others should not
(Kelly et ah, 1998). Others require a large amount of fluid take. Additionally, to prevent
drug interactions and to maintain the appropriate level of medications in the bloodstream,


91
La Greca, A.M., & Schuman, W.B. (1995). Adherence to prescribed medical
regimens. Handbook of Pediatric Psychology (2nd ed., pp. 55-140). New York, NY:
Guilford Press.
Laine, C., Newschaffer, C.J., Zhang, D., Cosier, L., Hauck, W.W., & Turner, B.J.
(2000). Adherence to antiretroviral therapy by pregnant women infected with human
immunodeficiency virus: A pharmacy claims-based analysis. Obstetrics-Gynecology,
95(2), 167-173.
Leventhal, H., Safer, M.A., & Panagis, D.M. (1983). The impact of
communications on the self-regulation of health beliefs, decisions, and behavior. Health
Education Quarterly, 10(1), 3-29.
Mackner, L.M., McGrath, A.M., & Stark, L.J. (2001). Dietary recommendations
to prevent and manage chronic pediatric health conditions: Adherence, intervention, and
future directions. Journal of Developmental & Behavioral Pediatrics, 22(2), 130-143.
Mannheimer, S., Friedland, G., Matts, J., Child, C., & Chesney, M. (2002). The
consistency of adherence to antiretroviral therapy predicts biologic outcomes for human
immunodeficiency virus-infected persons in clinical trials. Clinical Infectious Diseases,
34, 1115-1121.
Martino, M. de, Tovo, P.A., Balducci, M., Galli, L., Gabiano, C., Rezza, G., &
Pezzotti P. (2001). Reduction in mortality with availability of antiretroviral therapy for
children with perinatal HIV-1 infection. Italian Register for HIV Infection in Children
and the Italian National AIDS Registry. Journal of the American Medical Association,
284(2), 190-197.
McNabb, J., Ross, J.W., Abrila, K., Turley, C, Nightingale, C.H., Nicolau, D.P.
(2001). Adherence to highly active antiretroviral therapy predicts virologic outcome at
an inner-city human immunodeficiency virus clinic. Clinical Infectious Diseases. 33.
700-705.
Melvin, A. J. (1999). Anti-retroviral therapy for HIV-infected children-toward
maximal effectiveness. Pediatric Infectious Disease Journal, 18(8). 723-724.
Meyers, K.E., Thomson, P.D., & Weiland, H. (1996). Noncompliance in children
and adolescents after renal transplantation. Transplantation. 62. 186-189.
Monane, M., Gurwitz, J. H., Monane, S., & Avom, J. (1993). Compliance issues
in medical practice. Hospital Physician. 35-39.
Monane, M., Bohn, R.L., Gurwitz, J.H., Glynn, R.J., & Avron, J. (1994).
Noncompliance with congestive heart failure therapy in the elderly. Archives of Interval
Medicine, 154, 433-437.


51
Signed Ranks Test was then used to test for differences in the number of barriers reported
during the AI-HIV and the 24RI.
51


68
would not strongly relate to measures of adherence to frequency, since it is one thing to
examine whether or not the dose was ever received, and another thing to examine
whether or not the dose was received within certain constraints. Furthermore, the lack of
significant relationships between adherence measures is consistent with the results of
another study that compared caregiver self-report of HAART adherence to pharmacy and
MEMS data (Farley et al., 2002) and a study that compared self-report ART adherence
rates to pharmacy and MEMS data among adults (Frick, Gal, Lane, & Sewell, 1998).
These results are also similar to the results of studies of relationships between different
adherence assessment methods among caregivers of children with other chronic illnesses
(Glascow et al., 1987; Johnson et al., 1986). Finally, the significant and positive
relationship between pharmacy refill data and nurse-report ratings is unsurprising, as it is
likely that nurse ratings were influenced by nurses knowledge of pharmacy refill data, as
nurses have reported previously that pharmacy refill records obtained during routine
clinical care were influential in nurses ratings of patients adherence (Farley et al.,
2002).
The author also hypothesized that medication adherence would significantly and
negatively relate to disease severity. Nevertheless, when an 80% adherence cut-off was
used, viral load, CDC letter, and CDC number classification were not significantly
related with 24RI adherence ratings or refill ratings. Ninety-percent refill adherence
classification did not vary based on disease severity; however, 90% 24RI frequency
adherence classification did vary based on CDC number classification, such that those
who had ever had low CD4 counts (and thus, higher CDC number classifications) tended
to be classified as adherent more than those with historically higher CD4 counts (and


83
The present study has also documented caregivers perceived barriers to ART
adherence. The next step will be to conduct a similar assessment based on child-report.
The resulting lists should be combined with barrier lists used in the adult literature to
devise a barriers checklist that can be used to quickly assess adherence in the clinical
setting.
This study has also provided important information about significant deficits in
caregiver regimen knowledge and pharmacy refill adherence, and daily barriers to
regimen adherence. Efforts should now be undertaken to develop effective adherence
interventions that focus on increasing regimen knowledge, increasing adherence
behaviors, and decreasing barriers to adherence among HIV-infected children and their
families. Researchers should also continue to work to identify other important targets for
intervention. This study has begun to clarify some of the issues important to ART
adherence among children, but it is only the beginning. Future studies should attempt to
clarify these findings, and should conduct research into the role of child responsibility for
medication-taking, and the relationship between child responsibility and adherence.
Finally, perinatally HIV-infected children on ART are quickly growing into
adolescents and young adults. The literature regarding adolescents adherence to chronic
illness regimens suggests that we must prepare for decreases in adherence during this
developmental stage. The impact of HIV status on adolescents identity and sexual
experiences may further inhibit medication adherence behavior; therfore, including risk
behaviors in the assessment of adherence among HIV-infected youth may add valuable
information. Much research is necessary in order to inform clinicians about the most
effective disease management during this tumultuous period, particularly given the severe


3
adherence and barriers to adherence among children on ART are reviewed. The literature
review concludes with a summary, specific aims, and hypothCeses for the following
study, which examines the extent and nature of adherence and barriers to adherence
among HIV-infected children and their families. The method, results, and discussion of
the findings follow.


87
Cassell, E.J. (1991). The Nature of Suffering and the Goals of Medicine. New
York, NY: Oxford University Press.
Catz, S. L., Kelly, J.W., Bogart, L.M., Benotsch, E.G., & McAuliffe. (2000).
Patterns, correlates, and barriers to medication adherence among persons prescribed new
treatments for HIV disease. Health Psychology, 19(2), 124-133.
CDC- Centers for Disease Control (1996). AIDS among children United States,
1996. Morbidity and Mortality Weekly Report, 1005-1010.
CDC- Centers for Disease Control (1998, November). What is HIV? Retrieved
August 16, 2002, from http://www.cdc.gov/hiv/pubs/faq/faql.htm
Celano, M., Geller, R.J., Phillips, K.M., & Ziman, R. (1998). Treatment
adherence among low-income children with asthma. Journal of Pediatric Psychology,
23(6), 345-349.
Chesney, M. (1997, June). New antiretroviral therapies: Adherence challenges
and strategies. Presented at the Adherence in HIV Therapy: Implications for Nursing
Practice Conference, San Francisco.
Conley, S.B., & Salvatierra, O. (1996). Noncompliance among adolescents: Does
it impact the success of transplantation? Nephrology News & Issues, 10, 18-19.
Delamater, A.M. (2000). Critical issues in the assessment of regimen adherence
in children with diabetes. In D. Drotar (Ed.), Promoting Adherence to Medical
Treatment in Chronic Childhood Illness (pp. 173-195). Mahwah, NJ: Lawrence
Earlbaum Associates.
DiGirolamo, A. M. Quittner, A. L., Ackerman, V, & Stevens, J. (1997).
Identification and assessment of ongoing stressors in adolescents with a chronic illnes: an
application of the behavior-analytic model. Journal of Clinical Child Psychology, 26(1),
53-66.
DiMatteo, M.R., & DiNicola, D.D. (1982). Achieving Patient Compliance: The
Psychology of the Medical Practitioner's Role. New York, NY: Pergamon Press.
Duong, M., Pirothe, L., Peytavin, G., Forte, F., Kohli, E., Grappin, M., Buisson,
M., Chavanet, & Portier, H. (2001). Value of patient self-report and plasma human
immunodeficiency virus protease inhibitor level as markers of adherence to antiretroviral
therapy: Relationship to virologic response. Clinical Infectious Diseases. 33. 386-392.
Ettenger, R.B., Rosenthal, J.T., Marik, J.L., Malekzadeh, M., Forsythe, S.B.,
Kamile, E.S., Salusky, I.B., & Fine, R.N. (1991). Improved cadaveric renal transplant
outcome in children. Pediatric Nephrology. 5. 137-142.


53
caregivers perceptions of the prescribed intervals, as caregiver reports may have been
highly influenced by social desirability. For the purpose of analysis, the data were sorted
into interval buckets indicating the amount of deviation from the prescribed interval.
For example, if medication was taken less than 45 minutes before or after the prescribed
interval, the interval was scored as 0; if the medication was taken one hour before or
after the prescribed interval had passed, it was scored 1. Intervals were not computed
for once daily dosing, as the interview provided only one data point for dosing times of
once daily medications.
Descriptive statistics regarding interval deviance are provided in Table 5. Results
showed that the majority of doses were reportedly given within the prescribed intervals:
79.3% of the reported typical dosing intervals were less than 2 hours deviant from the
prescribed intervals, 88.8% were less than 3 hours deviant from the prescribed intervals,
and 96.3% were less than 4 hours deviant from the prescribed intervals. When only the
intervals for thrice daily medications were examined, only 40.7% of the reported typical
intervals were less than 2 hours deviant, 63% were less than 3 hours deviant, and 88.9%
were less than 4 hours deviant.
Relationship between Knowledge and Typical Adherence Behavior
In order to determine whether or not any differences existed between the
medications the caregivers believed were prescribed for the children and those the
caregivers reported that the children typically received, during the AI-HIV the caregivers
were asked to report the medications their children typically had taken in the past two
weeks, and the corresponding dosing amounts and dosing frequencies. In all 63
interviews, concordance was 100% between the two reports. Therefore, typical adherence
behavior data from the AI-HIV were not included in further analyses.


32
concerns, and depressed. Although the study identified which of those 7 reasons was
most widely endorsed by the teens, the study was limited by the pre-selected list of
barriers examined and a small sample size. Further, no information was provided about
how those barriers were chosen for examination. Thus, although the study examined the
extent to which the 7 barriers were endorsed by the teens, the study failed to identify the
variety of other barriers that may have impacted adherence.
Two qualitative studies have assessed adherence to ART regimens among adults.
In a pilot study of HIV-infected adults enrolled in a clinical drug trial (Chesney, 1997),
patients who admitted missing doses were asked to list as many reasons as possible for
why they missed their medication. Common barriers listed were: forgetting (40%),
sleeping through doses (37%), being away from home (34%), changing the therapy
routine (27%), being busy (22%), being sick (13%), experiencing side effects (10%), and
feeling depressed (9%). In a study of 20 HIV-infected women who were asked to write
about a future in which you only have to take one pill a day, multiple barriers to
adherence were revealed. Many women reported that the regimens were difficult to
follow, pointing to the dosing intervals, food guidelines, number of pills, and pill size, as
particularly problematic. Side effects were also identified as significant barriers,
specifically excessive perspiration, nausea and vomiting, diarrhea, and weight loss or
gain associated with medication-specific food requirements. Lack of confidence in pill
effectiveness was also discussed, as well as the way in which the medications serve as a
regular reminder of disease status. Finally, scheduling problems and high activity were
also reported as barriers to adherence. Jointly, these qualitative studies demonstrate the
value of open-ended assessment for the identification of barriers to adherence.


26
within the past week and then were asked several related questions, including information
about how many times per day they took each medication, and how many pills they took
each time. The Regimen Screen also consisted of questions about how many doses were
missed in the previous week. Four out of 20 patients reported some nonadherence or lack
of knowledge on the Regimen Screen. Results showed that the Regimen Screen was
associated with repeated nonadherence, measured with MEMS. However, the Regimen
Screen results were not associated with sporadic nonadherence.
Bangsberg and colleagues (2001) included a brief regimen knowledge assessment
in another small study. Forty-six mostly male patients on HAART were asked about
their medication regimen during an interview at patients typical places of residence. Six
patients reported that they were taking either a different number of doses per day or a
different number of pills per dose than the medication container instructed. Four of those
patients attributed this difference to misunderstanding, while two reported that they had
decided to take their medicine differently than it was prescribed.
In a larger study, Parietnti and colleagues (2001) used a pill identification test to
determine regimen knowledge among 223 HIV-infected adults in France. Participants
were given a board that contained pictures of 23 ART medications, with 2 similar
looking (twin pills) for each ART medication. A score was calculated based on the
number of misidentifications weighted according to the degree of pill resemblance. No
information was provided about the total percent of patients with inadequate knowledge,
though results showed that 38% of patients with poor adherence based on an investigator-
completed scale also had poor regimen knowledge.


90
Janz, N. K. B., M.H. (1984). The health belief model: A decade later. Health
Education Quarterly, 11, 1-47.
Johnson, S.B. (1995). Managing insulin-dependent diabetes mellitus in
adolescence: A developmental perspective. In L.J. Wallander & J. L. Siegel (Eds.),
Adolescent Health Problems: Behavioral perspectives, (pp. 265-288). New York, NY:
Guilford Press.
Johnson, S.B., Kelly, M., Henreta, J.C., Cunningham, W.R. & Tomer, A.,
Silverstein, J.H. (1992). A longitudinal analysis of adherence and health status in
childhood diabetes. Journal of Pediatric Psychology, 17(5), 537-553.
Johnson, S.B., Poliak, R.T., Silverstein, J.H., Rosenbloom, A.L., Spillar, R.,
McCallum, M., & Harkavy, J. (1982). Cognitive and behavioral knowledge about
insulin-dependent diabetes among children and parents. Pediatrics, 69(6), 708-713.
Johnson, S.B., Silverstein, J., Rosenbloom, A., Carter, R. & Cunningham, W.
(1986). Assessing daily management in childhood diabetes. Health Psychology, 5(6),
545-564. ~ ~ ~~
Kaplan, R.M., & Simon, H.J. (1990). Compliance in medical care:
Reconstruction of self-predictions. Annals of Behavioral Medicine. 12. 66-71.
Kastrissios, H., Suarez, J.R., Katzenstein, D., Girard, P., Sheiner, L.B., &
Blaschke, T.F. (1998). Characterizing patterns of drug-taking behavior with multiple drug
regimen in an AIDS clinical trial. AIDS, 12(17). 2295-2303.
Katko, E., Johnson, G.M., Fowler, S.L., & Turner, R.B. (2001). Assessment of
adherence with medications in human immunodeficiency virus-infected children. The
Pediatric Infectious Disease Journal. 20(121. 174-176.
Kelly, J.A., Otto-Salaj, L.L., Sikkema, K.J., Pinkerton, S.D., & Bloom, F.R.
(1998). Implications of HIV treatment advances for behavioral research on AIDS:
protease inhibitors and new challenges in HIV secondary prevention. Health Psychology,
17(4), 310-319. '
Klinnert, M.D., McQuaid, E.L., & Gavin, L.A. (1997). Assessing the family
asthma management system. Journal of Asthma. 34. 77-88.
Knobel, H., Alonso, J., Casado, J.L., Collazos, J., Gongalez, J., Ruiz, I., Kindelan,
J.M., Carmona, A., Juega, J., Ocampo, A, & the GEEMA Study Group (2002).
Validation of a simplified medication adherence questionnaire in a large cohort of HIV-
infected patients: the GEEMA study. AIDS, 16. 605-613.


REVIEW OF THE LITERATURE
According to the American Academy of Pediatrics (AAP), childhood chronic
illnesses are defined as conditions that last at least 3 months, and require medical
attention and care that is above and beyond that which is normally expected for a child or
adolescent of the same age (AAP, 1993). Estimates suggest that 18% of children in the
United States have a chronic illness (Newacheck et al., 2000). The most common
childhood chronic illnesses include rheumatoid arthritis, asthma, leukemia and other
malignancies, spina bifida, seizure disorders, neuromuscular diseases, acquired
immunodeficiency syndrome, and diabetes (AAP, 1993). Conditions vary widely in their
onset, course, duration, and severity, and in the daily demands required for disease
management.
Prevalence of Poor Adherence among Children with Chronic Illness
Among children with chronic illnesses, the prevalence of poor adherence to
medical regimens varies according to the study sample, the specifics of regimen
requirements, the assessment of adherence, and the criteria used to classify children as
adherent or poorly adherent (Rapoff, 1999). Across studies of children with chronic
illnesses, nonadherence has been estimated at 21-52% (Alessandro, Vincenzo, Marco,
Marcello, & Enrica, 1994; Conley & Salvatierra, 1996; Ettenger et al., 199; Festa,
Tamaroff, Chasalow, & Lanzowsky, 1992; Meyers, Thompson, & Weiland, 1996; Schoni
et al., 1995; Weisberg-Benchell et al., 1995). However, these estimates were taken from
a review of studies of adherence to a small set of chronic illness regimens, in which
4


14
Numerous self-report measures have been developed and used to assess adherence
in children with chronic illnesses, though few measures have gained widespread
popularity and use. One common method of self-report that has demonstrated good
reliability is the recall interview or diary method. Specifically, the 24-hour recall
interview (Freund, Johnson, Silverstein, & Thomas, 1991; Johnson, Silverstein,
Rosenbloom, Carter, & Cunningham, 1986; Johnson et al., 1992) has been used
successfully in the study of adherence in diabetes. The 24-hour recall interview is a
diabetes-specific assessment of adherence. The assessment procedure includes three
phone interviews (two during weekdays, one on weekend) within a two-week period.
Patients and parents are interviewed separately and asked to recall the events of the
previous 24-hours. If they fail to mention diabetes adherence behaviors, the interviewer
cues them. Quantified measures correspond to each aspect of the regimen behavior
(insulin injections, blood-glucose testing, diet, and exercise).
Generally, studies have supported the psychometric properties of the 24-hour
recall interview. Correlations between child and parent reports have ranged from r = .42
to r = .78 (M = -62), suggesting moderate inter-rater reliability (Johnson et al., 1986).
Lack of complete concordance is not surprising, given that parents do not observe all of
their childrens behavior, and the likelihood of errors of memory. Agreement between
child-reported adherence behaviors and observed behaviors supports the construct
validity of the measure (Reynolds, Johnson, & Silverstein, 1990). Greatest concordance
was found for the occurrence or nonoccurrence of behaviors regarding insulin injection,
exercise, and blood-glucose testing. Poorer agreement was found for the time during
which the behaviors occurred. The poorest agreement was found for specific measures of


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EXAMINING THE EXTENT OF POOR ADHERENCE AND THE BARRIERS TO
ADHERENCE AMONG HIV-INFECTED CHILDREN ON ANTIRETROVIRAL
THERAPY: A COMBINED QUALITATIVE AND QUANTITATIVE APPROACH
By
Stephanie L. Marhefka
December 2002
Chairperson: James R. Rodrigue, Ph.D.
Major Department: Clinical and Health Psychology
This study assessed adherence to antiretroviral medications among 63 children
with HIV infection. Trained interviewers administered the Adherence Interview-HIV
(AI-HIV) to caregivers of infected children to assess typical medication regimen
behaviors and knowledge of the target childs prescribed regimen. Next, demographic
information was collected via interview and a medical record review was conducted.
Two weeks following the initial interview, trained interviews began the 24-Hour Recall
Interview (24RI) procedure with each caregiver; caregivers were then interviewed over
the telephone two additional times within the next two weeks. Three months after the
initial interview, pharmacy refill data were collected.
Results show that 33% of caregivers failed to correctly identify at least one of
their childs medication names, 31% of caregivers failed to correctly identify the dosage
for at least one medication, and half of the caregivers failed to correctly identify the
vii


57
refill adherence cut-off of 80%, 46% of participants were considered adherent. When a
cut-off of 90% was used, 32% were considered adherent.
Nurses rated each childs adherence, drawing from information in the medical
record and information obtained during interactions with the children and their
caregivers. The mean nurse report rating was 5.92 (SD= 1.22) on a 7-point Likert scale.
The median rating was 6 and the modal rating was 7. No children received a rating of 1-
Not at all Adherent.
Relationship between Adherence Measures
Correlations were performed to determine relationships between adherence
measures (Table 6). 24RI frequency was not included in the correlation matrix, due to
kurtosis. 24RI interval and dietary adherence ratings did not significantly correlate with
pharmacy refill adherence ratings or nurse-reported adherence levels. However,
pharmacy refill adherence ratings significantly and positively correlated with nurse-
reported adherence levels (r= .26, p < .05), such that higher pharmacy refill ratings
related to higher nurse report ratings of adherence.
Relationship between Adherence and Disease Severity
Chi Square analyses were performed in order to test for relationships between
adherence measures and disease severity measures (viral load, CDC letter, or CDC
number). When participants were categorized as adherent (> 80% adherence rating)
and nonadherent (< 80% adherence rating) none of the disease severity measures was
significantly related with 24RI frequency adherence or refill adherence ratings. When a
more stringent criterion (90% adherence) was used, 24RI frequency was significantly and
positively related to CDC number (X2= 6.78, p< .05), such that those with higher 24RI


70
clinicians may be able to gain valuable information about adherence simply by asking
caregivers to state the medications that their children are currently taking. Nevertheless,
it remains unclear whether caregivers are more adherent because they know the
medication names, or they know the medication names because they are adherent.
Further investigation is necessary in order to determine to what extent caregivers lack of
knowledge is due to lack of effective regimen-specific education, lack of regimen
behaviors, and lack of caregivers involvement in the regimen behaviors.
AI-HIV interval knowledge was hypothesized to correlate significantly and
positively with 24RI interval. The present study supports this relationship, indicating that
caregiver reports of the typical intervals between dosing over a 2-week period tend to
increase as caregiver reports of the typical intervals between dosing over (3) one-day
periods increase. However, it is important to recognize that AI-HIV interval knowledge
is a proxy measure of interval knowledge, since caregivers are not directly asked to state
how many hours that they believe should pass between each dose.
It was also expected that 24RI dietary adherence would relate to knowledge of
dietary requirements, though this hypothesis is not supported by the current study. This
finding suggests that even when dietary requirements are known, children may not adhere
to the dietary component of their medication regimens. This is consistent with findings
from a review of dietary adherence among children with chronic illness (Mackner,
McGrath, & Stark, 2001), which showed that dietary adherence among chronically ill
children tends to hover at or below 50%, and knowledge of the dietary regimen is
insufficient for dietary adherence.


98
Oftentimes children have special ways in which they take their medicine. For
example, some children take their medicine right before they eat, right after, or
take it with a special drink. In the last two weeks, how has your child taken
? Record the response, and say:
And who has been involved in that? Record the response and provide the
caregiver with encouragement that his/her responses are important and helpful.
Then, go to the next medication on the list, and repeat this set of questions for each
medication on the list.
When you have queried all medications for typical medication-taking behaviors, you may
want to provide a summary statement that lets the caregiver know that you realize that
managing the medication-taking must be difficult, and it sounds as if he/she is really
trying hard (if appropriate).
Section 2.
Next, say: Now that I have learned all about the medications that
typically takes, Id like you to tell me about the things your doctor, nurse, or
pharmacist told you to do. During your last clinic visit (before today), what
medications did your doctor, nurse, or pharmacist recommend that you give to your
child?
Mark the medications on the recording sheet. If the caregiver describes the
medication (s), say: Do you know the name of it?
If the caregiver cannot recall the name, first say: Well come back to that one.
What other medications has your doctor, nurse, or pharmacist recommended?
When the caregiver can no longer name any medications, if some are left
unnamed, then SHOW THE DISPLAY CARD. Otherwise, begin with the first
medication, and say:
Lets start with How often did your doctor,
nurse, or pharmacist recommend that you give your child
? Record the response, and say:
And how much did your doctor, nurse, or pharmacist recommend that
your child take each time? Record the response, and say:
Sometimes doctors, nurses, and pharmacists give special instructions
about how to take medications. For example, they might say take the
medication with food, take it on an empty stomach, do not eat fatty foods with
this medication, or give other instructions. What special instructions did you
receive about giving your child ? Record the response, and
say:


24
mutations from replicating. The result is a high presence of the mutant virus in the blood
that is resistant to those drugs and similar drugs, as well. This is particularly problematic
since resistance may occur most rapidly among those whose adherence falls between
50% and 90% (Bangsberg et al., 2000), a range in which many children and adults on
HAART may fall (Belzer, Fuchs, Luftman, & Tucker, 1999; Kastrissios et al., 1998;
Watson & Farley, 1999).
For both the individual and society, the impact of poor adherence and subsequent
drug resistance may be severe. First, with the limited number of antiretroviral agents and
the ability of resistance to affect entire drug classes, resistance may leave an individual
with few options for additional therapy (Gallant, 2000). As was seen in the early 1980s,
without therapy the virus replicates and completely suppresses the immune system.
When the immune system is suppressed, the body cannot fight even a common cold.
Unfortunately, without therapy, rapid physical decline is imminent. Second, research has
demonstrated that drug resistance leads to increased viral replication in the blood, semen,
and vaginal fluid (Gupta et al., 1997; Vemazza, Gillem, & Flepp, 1997). The potency of
the virus in these fluids increases the likelihood that the virus may be transmitted to
exposed individuals (Wainberg & Friedland, 1998). If the drug-resistant virus is
transmitted to others, those infected will have a drug-resistant virus, as well (Friedland &
Williams, 1999). Like the poorly adherent individuals with drug resistance, newly
infected individuals will have little or no options for therapy. As a result, future
generations of HIV-positive individuals may suffer dramatically from the poor adherence
of their predecessors. Clearly, among HIV-infected individuals, the potential public
health implications of poor adherence and subsequent drug resistance are grave.


80
measures may become important components of multidimensional HAART adherence
assessment in both the clinical and research domains.
Finally, this study adds to the existing literature regarding childhood chronic
illnesses, as the results, which are somewhat disparate from previous findings related to
adherence in other chronic illnesses, help to demonstrate the variability of disease
management behaviors across different illnesses. Findings also demonstrate that despite
variability in disease management behaviors, instruments that have been developed for
use in measuring adherence to one chronic illness may be useful in assessing adherence
to other chronic illnesses with other target behaviors; nevertheless, such instruments may
require modification in order to fit with specific populations and regimens.
Study Limitations
Despite the many strengths of the current study, several weaknesses should be
considered, as they may limit the interpretation of the findings. The first group of
considerations pertains to selection bias. First, several groups were excluded from the
study by not meeting inclusion criteria. Children concurrently enrolled in adherence
intervention studies and those with unstable home environments and unstable medication
regimens, foster caregivers, and caregivers for whom English was a second or other
language were all excluded from the study. Therefore, the results presented may not
generalize to these groups. Second, participants were recruited from 3 urban southeastern
medical centers in the United States and thus, results may not generalize to other regions
nationally or internationally. Also, 14% of eligible participants approached refused to
participate in the study, several subjects failed to complete the 24RI, and pharmacy refill
data were not retrievable for one subject. Self-selection bias may have occurred. It may


APPENDIX F
24-HOUR RECALL INTERVIEW
The 24-Hour recall interview is intended to be administered in a conversational format.
Although participants will be aware that they will be called during a particular two-week
period, they should not know in advance on which days the calls will occur.
The interviewer should always try to first establish a rapport with the respondent.
Respondents are asked to recall the events of the previous day in temporal sequence,
beginning with the time they awoke and ending with the time they went to bed. The
interviewer should begin by asking questions such as Do you remember what time your
child got up yesterday? to promote thinking about the events of the previous day in an
unstructured, free-flowing manner. All HIV regimen-specific behaviors are recorded. If
the respondent mentions that an activity occurred, the interviewer may ask for details
regarding the behavior. If the necessary information is not voluntarily offered, the
interviewer may prompt with questions.
The key to the interview is specificity. Interviewers will want to determine, did the
behavior occur? Did the parent observe the child taking medicine/eating/ etc..? What time
was it? When obtaining meal information, be as specific as possible.
Introductory Phone Script
Hello, this is calling from Project FRAME. May I speak with
?
If different than person who answered the phone, say: Hello, this is
calling from Project FRAME.
How are you doing today? Use this casual conversation to build rapport. Express
empathy for any hardships.
I am going to ask you some questions about you childs day yesterday. This should
take about 20 minutes. Lets get started. Do you remember what time your child woke
up yesterday? The remainder of the questions should flow from here.
When the informant mentions that medication-taking occurred, record information about
the medication names and dosing. If the informant eludes to medication-taking but does
not provide the specific names and dosing, ask the informant which medications the child
took at a given time, and how much the child received of each medication.
112


63
Ran out
1(1)


Pharmacy doesnt always have
medicine in stock
1(1)


Difficulty remembering to take pills
along when going out
1(1)


Out of routine
1(1)


Doesnt drink all of the medicine
1(0)


Total number of barrier reports
69 (32)


Note. Numbers in parentheses represent the number of barriers reported during 24 RI.
Barriers provided during the 24RI were not rated according to frequency and strength.


95
Walsh, J.C., Home, R., Dalton, M., Burgess, A.P., & Gazzard, B.G. (2001).
Reasons for non-adherence to antiretroviral therapy: Patients' perspectives provide
evidence of multiple causes. AIDS Care, 13(6), 709-720.
Walsh, J.C., Mandalia, S, & Gazzard, B.G. (2002). Responses to a 1-month self-
report on adherence to antiretroviral therapy are consistent with electronic data and
virological treatment outcome. AIDS, 16, 269-277.
Watson, D. C. & Farley, J.J. (1999). Efficacy of and adherence to highly active
antiretroviral therapy in children infected with human immunodeficiency vims type 1.
Pediatric Infectious Disease Journal, 18(8), 682-689.
Weisberg-Benchell, J., Glasgow, A.M., Tynan, W.D., Wirtz, P., Turek, J., &
Ward, J. (1995). Adolescent diabetes management and mismanagement. Diabetes Care,
18, 77-82.


25
Assessment of Adherence among Children with HIV
Despite the implications of poor adherence among children on ART, few studies
have examined adherence levels in this population. Although many studies have assessed
adherence as one of the inclusion criteria for clinical trials (J. W. Sleasman, personal
communication, July 27, 2000), and as part of the Pediatric Aids Clinical Trials Group
assessment battery, few known studies have reported systematic assessments of
adherence, and no known studies have assessed knowledge of the prescribed medication
regimen among HIV-infected children.
Regimen knowledge
Only one study has assessed regimen knowledge among caregivers of HIV-
infected children or among the children themselves. Katko and colleagues (2001) asked
35 caregivers to name or describe their childrens ART medications and corresponding
dosages and dosing frequencies. Nineteen (54%) of the caregivers provided accurate
medication information, and of those, 12 had pharmacy refill ratings of 90% adherence or
better, while none of those who lacked knowledge of the regimen had 90% adherence
ratings or higher. This study was limited by a small sample size and only one method of
adherence assessment, yet the study is important, as it supports the use of knowledge
assessment as an indicator of possible poor adherence.
Although only one study has included regimen knowledge as part of the
adherence assessment of HIV-infected pediatric populations, several studies have
assessed regimen knowledge among HIV-infected adults. In a small study, Svarstad,
Chewning, Sleath, and Claesson (1999) assessed regimen knowledge among 20 HIV-
infected men and women. As part of the Brief Medication Questionnaire (BMQ) they
conducted a Regimen Screen. Patients were asked to list all of the medications taken


21
often the timing of medication taking is crucial (Kelly et al., 1998). With such complex
instructions, it is not surprising that adherence to ART is of great concern (Friedland &
Williams, 1999; Kelly et al, 1998).
Implications of Poor Adherence to ART
As with other chronic illnesses, poor adherence to ART affects healthcare costs,
clinical decision-making, conclusions drawn from clinical trials, and health outcomes.
The widespread use of HAART has reduced morbidity and mortality, hospitalizations,
the development of opportunistic infections, and the use of home-care services and
hospices (Brettle et al., 1998). In reducing morbidity and mortality, HAART has also led
to reduced costs; however, when patients are nonadherent to HAART regimens, HAART
may not be cost-effective, because the expensive medicines are not taken at the
appropriate intervals and are likely less effective. As a result, resource utilization may be
higher and costs may approach those incurred prior to the availability of ART.
Poor adherence to ART can also adversely affect clinical decision-making. For
example, consider an HIV-infected patient who misses doses of ART often, but is
perceived to be adherent. If viral suppression is not achieved with the current medication
regimen, then a physician may conduct extensive laboratory tests in hopes of determining
the reason for failure. Orders for these tests would not have been made if the physician
knew the patient was poorly adherent. The physician might have instructed the patient to
discontinue the medication altogether, since intermittent adherence may be more harmful
than effective (Friedland & Williams, 1999). The physician may have prescribed another
medication with fewer side effects or with simpler dosing, with which the patient might


94
Scott, G. B. & Sleasman, J.W. (1999). Pediatric HIV Infection.: Florida
Department of Health Children's Medicals Services.
Singh, N., Berman, S.M., Swindell, S., Justis, J.C., Mohr, J.A., Squier, C., &
Wagener, M.M. (1999). Adherence of human immunodeficiency virus-infected patients
to antiretroviral therapy. Clinical Infectious Diseases, 29(4), 824-830.
Singh, N., Squier, C., Sivek, C., Wagmer, M., Nguyen, M.H., & Yu, V.L. (1996).
Determinants of compliance with antiretroviral therapy in patients with human
immunodeficiency virus: prospective assessment with implications for enhancing
compliance. AIDS Care, 8(3), 261-269.
Stone, V.E., Hogan, J.W., Schuman, P., Rmpalo, A.M., Howard, A.A.,
Korkontzelou, C., Smith, D.K. (2001). Journal of Acquired Immune Deficiency
Syndromes, 28, 124-131.
Svarstad, B.L., Chewning, B.A., Sleath, B.L., & Claesson, C. (1999). The brief
medication questionnaire: A tool for screening patient adherence and barriers to
adherence. Patient Education and Counseling, 37, 113-124.
Tebbi, C.K., Cummings, K.M., Zevon, M.A., Smith, L., Richards, M., & Mallon,
J.C. (1986). Compliance of pediatric and adolescent cancer patients. Cancer, 58, 1179-
1184.
UNAIDS- The Joint United Nations Programme on HIV/AIDS (2000). Report on
the global HIV/AIDS epidemic. Retrieved July 6, 2000, from http://www.unaids.org/
epidemic_update/report/Epi_report.htm.
Urquhart, J. (1989). Noncompliance: The ultimate absorption barrier. In L. F.
Prescott & W. S. Mimmo (Eds.), Novel Drug Delivery and its Therapeutic Application.
New York, NY: Wiley.
Ventura, S. J., Peters, K. T., Martin, J. A., & Maurer, J. D. (1996). Births and
deaths: United States. Monthly Vital Statistics Report, 46tSupplement 2), 32.
Vemazza, P., Gillem, B., & Flepp, M. (1997). Effect of antiviral treatment on the
shedding of HIV-1 in semen. AIDS, 11. 1249-12454.
Wagner, J.H., Justice, A.C., Chesney, M., Sinclair, G., Weissman, S., Rodrguez-
Barradas, M., & the VACS 3 Project Team (2001). Patient- and provider-reported
adherence: Toward a clinically useful approach to measuring antiretroviral adherence.
Journal of Clinical Epidemiology, 54. S91-S98.
Wainberg, M. A., & Friedland, G. (1998). Public health implications of
antiretroviral therapy and HIV drug resistance. Journal of the American Medical
Association. 279(24). 1977-1983.


72
formulations. These findings may be due to caregivers giving greater responsibility to
children taking pills, children then forgetting to take the pills or deliberately returning the
pills to the container, and children then forgetting or neglecting to inform the caregiver of
the need for refills. Caregivers may be more directly involved in the administration and
refill of liquid medications, therefore, their reports regarding adherence to liquid
medications may be more accurate.
Barriers to Adherence
Another goal of the present study was to document the perceived barriers to
adherence among caregivers of children on ART. Caregivers were asked to consider the
past two weeks and report things that had made it difficult for the child to get their
medication exactly as prescribed. On average, caregivers reported about one barrier
each, which is far lower than might be expected, given the large percentage of caregivers
defined as nonadherent based on pharmacy refill rates of less than 80%. However, an
examination of the self-report data serves as a reminder that barrier reporting may be
strongly impacted by social desirability. Both adherent and nonadherent caregivers may
be concerned that reporting hardship or problems with the regimen would lead the
researcher to conclude that their child is not getting the medicine. Due to unawareness,
other caregivers may be unable to articulate the barriers they experience, particularly in
cases in which anger, guilt, depression, or denial prevents adherence, or in other cases, in
which children lie about taking their medication. Other caregivers may experience
difficulty remembering the problems that occur. It also remains possible that some
families do not experience barriers to ART adherence, a phenomenon that was suggested
in the present study by caregivers who reported that they experienced many problems


REFERENCES
Abbott Laboratories (2001). My Medication Schedule. North Chicago, IL:
Abbott Laboratories, Inc.
Alessandro, F., Vincenzo, Z.G., Marco, S., Marcello, G., & Enrica, R. (1994).
Compliance with pharmacologic prophylaxis and therapy in bronchial asthma. Annals of
Allergy. 73(2). 135-40.
American Academy of Pediatrics (1993). Psychosocial risks of child health
conditions in childhood and adolescence. Pediatrics, 92(6), 876-878.
Ammassari, A., Murri, R., Pezzotti, P., Trotta, M.P., Ravasio, L., DeLongis, P.,
LoCaputo, S., Narciso, P., Pauluzzi, S., Carosi, G., Nappa, S., Piano, P., Izzo, C.M.,
Lictner, M., Rezza, G., Montforte, A.A., Ippolito, G., Moroni, M., Wu, A.W., & Antinori,
A. (2001). Self-reported symptoms and medication side effects influence adherence to
highly active antiretroviral therapy in persons with HIV infection. Journal of Acquired
Immune Deficiency Syndromes, 28, 445-449.
Anthony, H., Paxton, S., Bines, J., & Phelan, P. (1999). Psychosocial predictors
of adherence to nutritional recommendations and growth outcomes in children with cystic
fibrosis. Journal of Psychosomatic Research, 47(6), 623-634.
Bangsberg, D.R., Bronstone, A., & Hofman, R. (2002). A computer-based
assessment detects regimen misunderstanding and nonadherence for patients on HIV
antiretroviral therapy. AIDS Care, 14(1). 3-15.
Bangsberg, D.R., Hecht, F.M., Charlebois, E.D., Zolopa, A.R., Holodniy, M.,
Sheiner, L., Bamburger, J.D., Chesney, M.A., & Moss, A. (2000). Adherence to protease
inhibitors, HIV-1 viral load, and development of drug resistance in an indigent
population. AIDS, 14, 357-366.
Bangsberg, D.R., Hecht, F.M., Clague, H., Charlebois, E.D., Ciccarone, D.,
Chesney, M., & Moss, A. (2001). Provider assessment of adherence to HIV antiretroviral
therapy. Journal of Acquired Immune Deficiency Syndromes, 26, 435-442.
Beck, D. E., Fennell, R. S., Yost, R. L., Robinson, J. D., Geary, D. & Richards, G.
A. (1980). Evaluation of an educational program on compliance with medication
regimens in pediatric patients with renal transplants. The Journal of Pediatrics, 96(6),
1094-1097.
85


48
80% agreement with the authors coding of a practice interview before conducting
interviews for the study.
Pharmacy Refill History
The validity of pharmacy refills history has been supported by previous studies
(Farley et ah, 2002), though the reliability of pharmacy refill over time has not been well
established. Therefore, like the other assessment methods, this method is somewhat
exploratory in nature.
Written informed consent to contact the pharmacies, as well as the list of
pharmacies used by each family was obtained at the initial clinic visit. Pharmacies for
each family were contacted and pharmacy refill histories for the three-month period were
obtained. Scores were calculated by dividing the intended number of days to one refill
(typically 30 days) by the actual number of days to one refill. Values greater than one
were scored as 100% adherent, as the families had refilled their prescriptions prior to the
date on which the medications would be needed if the family were 100% adherent. If no
refill was detected within the 3 month period, the family was scored as 0% adherent.


16
The clinical interview is a relatively inexpensive, unobtrusive methodology that allows
for ideographic assessment of adherence behaviors and barriers. It is practical for use in
both clinical and research domains. However, few studies have used the clinical
interview for the assessment of adherence behaviors in children with chronic illnesses
(Hanson et al., 1996; La Greca, 1995), and few known studies have used the clinical
interview to assess the extent and nature of barriers to adherence.
Assessment of Barriers
Barriers to medical regimen adherence have rarely been systematically studied in
the child and adult literature. Although many studies have examined barriers to treatment
among children with chronic illnesses, most studies have examined barriers as individual
specific constructs that may be predictive of adherence, such as perceived severity,
perceived self-efficacy, and self-esteem, among others. Few studies have asked patients
to identify their perceived barriers to treatment, neglecting a great resource that may help
to elucidate appropriate areas for intervention.
One study assessed barriers to adherence among teenagers and adults with
diabetes (Glasgow, McCaul, & Schafer, 1986). In this study, the Barriers to Adherence
Questionnaire (BAQ) was developed, based on the Behavior Analytic Model (Goldfried
and DZurilla, 1969). Six individuals with insulin-dependent diabetes mellitus (LDDM),
and two diabetes nurse-educators generated a list of barriers to adherence to various
aspects of the diabetes regimen (diet, exercise, glucose testing, and insulin injections).
Next, the generated items were empirically tested, and a scale with 15 items was derived.
Frequency and severity ratings were provided, although the severity ratings were
eliminated due to high intercorrelation with the frequency ratings. No data are available


Section 1.
APPENDIX B
ADHERENCE INTERVIEW-HIV
Thank you for agreeing to participate in this study. As we discussed during the
informed consent process, we are trying to learn more about families with children
who have illnesses like your child has. Today, I want to learn about what things are
like in your family. Doctors often give children like yours many medications to take.
But making sure that your child gets all the medications can be really hard, so families
do the best that they can. Tell me, what things has your family typically done to
manage or treat s illness in the past two weeks? Has
take any medications in the past two weeks?
If the caregiver answers yes, What medications has your child taken?
If the caregiver answers no, proceed to Section 2.
Mark the medications on the recording sheet. If the caregiver describes the
medication (s), say: Do you know the name of it?
If the caregiver cannot recall the name, first say: Well come back to that one.
What other medications has your child taken?
When the caregiver can no longer name any medications, if some are left
unnamed, then SHOW THE DISPLAY CARD. If display card is used, mark DC on
recording sheet. If subject uses notes to name medication at any time, mark CS (cheat
sheet) for that medication. Otherwise, begin with the first medication, and say:
You said your child has taken does that come in
capsules, tablets, powder, or liquid? Record the response, then say:
In the past two weeks, how many times a day has your child taken
? Record the response, then say:
And at what times during the day has your child taken
? Record the response, then say:
When your child has taken how much has he/she
taken? Here, you want to know how many tablets or capsules, or how much
powder or liquid.
97


APPENDIX A
ELIGIBILITY CHECKLIST
Research Personnel: Ask yourself the following questions about the patient. Place a
checbnark next to the best answer. Please complete ALL items.
1. Is the child currently prescribed anti-retroviral
medication?
yes
no
2.
Has the child been prescribed anti-retroviral
medication for at least 3 months?
yes
no
3.
Have the childs medications been the same in the
last 3 months?
ves
no
4.
Is the child between the ages of 2 and 12 years?
yes
no
5.
Is the child currently enrolled in another adherence
yes
no
study?
ALL items must be marked YES to proceed to below.
Preliminary Questions (Inclusion Criteria)
Research Personnel: Ask the caresiver (who has accompanied the child to clinic) the
following questions. Place a checkmark next to the best answer. Please complete ALL of
the items.
1.
Are you one of the persons responsible for your child
getting HIV medications?
yes
no
2.
In the last month, has your child lived with you?
yes
no
3.
In the next 3 months, do you expect that your child
will continue to live with you?
yes
no
4.
Is English your first language?
yes
no
96


22
have been more adherent. Thus, non-detection of adherence problems may result in
inefficient, ineffective, and perhaps harmful medical care.
As discussed previously, poor adherence also affects conclusions drawn from
clinical trials. Undetected poor adherence in clinical trials can lead researchers to
erroneously conclude that drugs are not effective. With HIV, clinical trials are a critical
component of the development of new medications. Since few drugs are currently
available and drugs may become less effective for an individual over time (Gallant,
2000), the consequences of poor adherence in HIV clinical trials may be widespread and
significant.
As antiretroviral therapy has begun to slow disease progression and increase life
span, the issue of adherence to medication treatment has become more salient. Friedland
and Williams (1999) stated that: the key to success of the new regimens is the ability
and willingness of HIV-positive individuals to adhere to complex antiretroviral regimens,
perhaps for life. Adherence is essential to suppressing the viral load, the amount of HIV
in blood. Although multiple factors may contribute to inadequate viral suppression, poor
adherence to ART regimens is among the most important (Melvin, 1999).
Researchers have attempted to identify the adherence rates necessary to achieve
optimal health. Among adults, one prospective study assessed adherence levels among
81 mostly male patients with MEMS for 3 to 15 months (Paterson et al., 2000). Results
showed that patients with > 95% adherence to ART regimens have lower viral loads,
greater increases in CD4 counts (amount of T-lymphocyte white blood cells), and lower
hospitalization rates than those with poorer adherence. Another study assessed adherence
among 24 HIV-infected adults on ART (Bangsberg et ah, 2000). Researchers found that


17
regarding the reliability of the measure. Construct validity was supported by moderate
correlations between the BAQ and self-reported adherence. Although the BAQ may be
an effective tool in identifying barriers to adherence among individuals with diabetes, no
additional research has used the BAQ to assess barriers to adherence. Perhaps the
reliability of the measure was poor or the predictive validity of the measure was
inadequate to warrant further use, although these are simply speculations. Nevertheless,
a questionnaire that assesses regimen-specific barriers may be important in identifying
targets for interventions.
Rapoff (1999) reports that he often assesses barriers to adherence during clinical
interview by asking patients: What gets in the way of you taking your medicine? He
reports that the responses can be helpful in identifying targets for intervention. However,
no known published study has systematically assessed barriers in this manner.
Other literature on adherence in childhood chronic illness populations reveals
common developmental and behavioral reactions to managing complex medical
regimens. Several important trends have been identified. For example, the tendency for
older children and adolescents to exhibit poorer adherence than younger children has
been supported by this literature (Hanson, Henggeler, & Burghen, 1987b; Johnson,
Silverstein, Rosenbloom, Carter, & Cunningham, 1986; Ricker, Delamater, & Hsu,
1998). This may be, in part, because parents may tend to discontinue or decrease
supervision of disease management behaviors as children become older (Johnson, 1995).
Moreover, adolescence, in particular, is a developmental period strongly related to
decreased disease management behaviors. Adolescence is a time of striving for
independence, and one way in which adolescents sometimes assert their independence is
by choosing to not perform disease management behaviors.


METHOD
Participants
Study participants were primary caregivers of HIV-seropositive children ages 2-
12, contacted at pediatric HIV specialty clinics in Gainesville and Jacksonville, Florida,
and Baltimore, Maryland. Seventy-three primary caregivers met eligibility criteria and
were asked to participate in the study. Sixty-three primary caregivers (86%) agreed to
participate in the study, including 18 from the Gainesville site, 25 from the Jacksonville
site, and 20 from the Baltimore site. Chi square tests and one-way ANOVAs were
conducted to test for demographic differences among participants from the three sites.
No significant differences were found for child or caregiver age or gender, caregiver
marital status, caregiver education, caregiver employment status, household yearly
income, child viral load, whether or not the children were aware of their HIV status,
whether or not the caregiver was HIV-infected, and the frequency at which the caregiver
reported helping the child with medication-taking. Differences were found for both child
and caregiver ethnicity (X2= 15.55, p< .05; X2= 12.59, p= .05, respectively), with fewer
African-American participants at the Gainesville site.
Full descriptive statistics on participant demographic information can be found in
Tables 1 and 2. Child participants were primarily African-American (79.4%), male
(57.1%), and aware of their HIV diagnosis (50.8%). The average child participant was
8.8 years old (SD=2.97) and had been prescribed ART for an average of 6.8 years. The
modal viral load was less than 50 (undetectable). Caregiver participants were primarily
40


73
when they initially began the regimen, but that often with the help of healthcare
providers, solutions were found and problems were eliminated. Finally, because children
with the greatest adherence barriers may quickly develop drug resistance, and thus,
frequently change medication regimens, children with the greatest adherence barriers
may have been excluded from this study based on the inclusion criterion that children
must have been prescribed the same medication for the previous 3 months.
Twenty-two different barriers were reported between the AI-HIV and 24RI.
Reported barriers generally related to medication-specific attributes, such as the size or
taste of pills, problems with scheduling or routine, or problems with the child resisting,
refusing, or hiding the medication. With the exception of the child-specific problems, the
barriers reported in the current study have been reported in the adult ART literature, as
well (Ammassari et al., 2001; Catz et ah, 2000; Chesney, 1997; Mannheimer et ah, 2002;
Murphy et ah, 2000; Walsh et ah, 2001). Caregivers in the present study did fail to
mention some barriers that have been highly endorsed by HIV-infected adults; in
particular, emotional/psychological barriers such as being angry (Murphy et ah, 2000) or
depressed (Chesney, 1997; Murphy et ah, 2000), or not wanting to be reminded of HIV
(Catz et ah, 2000) were not mentioned. Also, difficulty with the dietary restrictions and
timing of doses (Catz et ah, 2000; Mannheimer et ah, 2002; Murphy et ah, 2000; Roberts
& Mann, 2000; Walsh et ah, 2001) were not mentioned as barriers.
When caregivers reported barriers, they were asked to rate the frequency of
occurrence and the strength of barriers in preventing medication-taking. Many caregivers
reported high barrier frequency ratings, and low barrier strength ratings were almost
always given. These results imply that although families may experience many things


93
Quittner, A.L., Drotar, D., Ievers-Landis, C., Slocum, N., Seidner, D., &
Jacobsen, J. (2000c). In D. Drotar (Ed.), Promoting Adherence to Medical Treatment in
Chronic Childhood Illness (pp. 383-407). Mahwah, NJ: Lawrence Earlbaum Associates.
Quittner, A.L. & Espelage, D.L. (1999). Reliability and validity of a daily phone
diary measure to assess daily activities and family interactions. Unpublished Manuscript.
Quittner, A.L., Espelage, D.L., Levers-Landis, C., & Drotar, D. (2000b).
Measuring adherence to medical treatments in childhood chronic illness: Considering
multiple methods and sources of information. Journal of Clinical Psychology in Medical
Settings, 7(1), 41-53.
Rand, C.S., & Weeks, K. (1998). Measuring adherence with medication regimens
in clinical care and research. The Handbook of Health Behavior Change (2nd Ed.), 114-
132. New York, NY: Springer Publishing.
Rapoff, M.A. (1999). Adherence to pediatric medical regimens. In M. C. Roberts
and A.M. LaGreca (Eds.), Clinical Child Psychology Library (pp. 1-75).
Rapoff, M.A., Lindsley, C.B., & Christopherson, E.R. (1985). Improving
compliance with medical regimens: Case study with juvenile rheumatoid arthritis.
Archives of Physical Medicine and Rehabilitation. 65. 267-269.
Reid, P. & Appleton, P. (1991). Insulin dependent diabetes mellitus: Regimen
adherence in children and young people. Irish Journal of Psychology, 12(1), 17-32.
Reynolds, L.A., Johnson, S.B., & Silverstein, J. (1990). Assessing daily diabetes
management by 24-hour recall interview: The validity of children's reports. Journal of
Pediatric Psychology, 15, 493-509.
Ricker, J.H., Delamater, A.M., & Hsu, J. (1998). Correlates of regimen adherence
in cystic fibrosis. Journal of Clinical Psychology in Medical Settings. 5(2). 159-172.
Roberts, K.J., & Mann, T. (2000). Barriers to antiretroviral medication adherence
in HIV-infected women. AIDS Care, 12(4). 377-386.
Rosenbach, K.A., Allison, R., & Nadler, J.P. (2002). Daily dosing of highly
active antiretroviral therapy. Clinical Infectious Diseases. 34, 686-692.
Rubin, D.H., Bauman, L. J., & Lauby, J. L. (1989). The relationship between
knowledge and reported behavior in childhood asthma. Developmental and Behavioral
Pediatrics. 10(61,307-312.
Schoni, M.H., Horak, E., & Nikolaizik, W.H. (1995). Compliance with therapy in
children with respiratory diseases. European Journal of Pediatrics. 154. S77-S81.


8
Conversely, multiple indirect assessments have been found to be effective, relatively easy
to obtain, and inexpensive.
Regimen Knowledge
One important aspect of measuring adherence indirectly is determining the extent
to which the persons responsible for a childs medication-taking are knowledgeable about
the prescribed regimen. If these individuals do not know and understand the specifics
regarding the medication regimen, complete adherence is unlikely. Among children with
a variety of chronic illnesses, lack of child and caregiver regimen knowledge has been
associated with poor adherence (Gudas, Koocher, & Wypij, 1991; Hanson, Henggeler, &
Burghen, 1987a; levers et al., 1999; Rubin, Bauman, & Lauby, 1989). However, not all
studies have found a relationship between knowledge and adherence (Beck et al., 1980).
These variable findings may be explained because although knowledge of the regimen is
important for adherence, it may not be necessary for children and caregivers to have
complete knowledge while in the clinic. For example, knowledge of exact dosages may
not relate to adherence, as that information is likely displayed on the medication
containers, and is, therefore, available to children and caregivers when it is most needed.
These findings may also be explained because although knowledge is important,
knowledge alone is not sufficient for adherence. Thus, accurate knowledge appears to be
an important component to adherence, though it should not be assumed that improving
knowledge would necessarily improve adherence.
Several questionnaires have been developed to measure diabetes-related
knowledge. The Test of Diabetes Knowledge (TDK; Johnson et al., 1982) was developed
to assess general knowledge of diabetes, diabetes-related problem-solving, and observed


36
problem more thoroughly and provide the information needed to effectively apply
quantitative methods and specific, focused interventions to the problem.


44
one weekend day. Three months following the clinic visit pharmacies were contacted to
obtain pharmacy recall histories for the prior three months.
Measures
Adherence Interview-HIV (AI-HIV; Appendix B)
This structured interview was designed by the author to assess typical adherence
to prescribed regimens, understanding and knowledge of the prescribed regimen, and
barriers to adhering to each aspect of the pediatric HIV regimen. The interview is
conducted with caregivers. It was adapted from the Treatment Adherence Questionnaire,
which is used to assess adherence to pediatric CF regimens (Quittner et al., 2000a). Due
to the novelty of this measure, reliability and validity have not been established.
The interviewer begins by asking parents/caregivers to inform the interviewer of
what the family has typically done to treat the childs illness in the last two weeks.
Difficulty managing the childs regimen is normalized, in order to provide a safe,
nonjudgmental context in which honest responding is encouraged (Murphy & Dillon,
1998). A display card with pictures of each medication and corresponding medication
names is used to aid those who may not know the medication names. Each medication is
discussed individually, in terms of how often each was taken, when the medication was
taken, how much was taken, and who was involved in the medication-taking. After the
informant has discussed all of the medications the child has typically taken in the last two
weeks, the interviewer asks the informant to report what medications the doctor
prescribed, how often they were prescribed, when the medications were prescribed to be
taken, how much was prescribed to be taken at a time, and special instructions about food
and drink intake. First, the actual disease management behaviors are queried in a non-


76
the behavior was observed, and at what time the behavior occurred. Eliminating queries
about this specific information may help to decrease the face validity and associated
social desirability of the measure. Future studies should aim to test the utility of the
measure once these modifications are made, and shouldexamine the reliability and
validity of the 24RI using both child and caregiver reports. Once the reliability and
validity of the measure are supported, clinicians may consider administering the 24RI to
children who are prescribed ART and their caregivers. Clinicians may wish to administer
the 24RI in the clinical setting or over the telephone at random intervals.
The present study also adds to the existing literature by confirming the value of
the pharmacy refill methodology for assessing adherence to ART regimens among HIV-
infected children. Pharmacy refill histories appear to provide more conservative, and
seemingly more accurate adherence estimates than self-report. Thus, both clinicians and
researchers should continue to consider using pharmacy refill histories as an inexpensive
methodology for obtaining adherence estimates to supplement self-report data.
The present study also has implications for interventions designed to improve
adherence among HIV-infected children on ART. First, by demonstrating that between
57% and 68% of children may receive suboptimal levels of ART, this study supports the
widespread belief that many children are not adherent to ART regimens, and thus,
supports the need for adherence interventions in this population. Next, this study points
to possible targets for intervention. Results of the AI-HIV and 24RI suggest that
improving knowledge about the prescribed regimen, including information about
appropriate dosing intervals and dietary requirements, is one possible intervention target
implicated in this study. Results of the barrier assessments suggest at least two additional


59
Table 6. Correlations Between Adherence Measures
24RI interval
24RI Dietary
Pharmacy Refill
24RI Interval


24RI Dietary
.04
(n= 38)


Pharmacy Refill
-.05
(n= 47)
-.29
(n= 38)

Nurse-reported
-.12
(n= 51)
-.09
(n= 42)
.26*
(n=59)
Note: *= g< .05


APPENDIX E
MEDICAL RECORD REPORT FORM
1.What is the childs current CDC status (clinical status)?
Check one
Check one
A
1
B
2
C
3
2.Childs most recent CD4 count: cells %
3.Date of most recent CD4 count: / /
month / day / year
4.Childs most recent viral load:
copies/ mL
5.Date of most recent viral load: / /
month / day / year
6. To the best of your ability, please rate the childs adherence to his/her
antiretroviral therapy regimen. Circle the number that best represents the extent to
which the child has followed his/her medical regimen within the past 6 months:
1 2 3 4 5 6 7
Not at all Sometimes Always
Adherent Adherent Adherent
7. Prescribed Regimen (please complete the form on the next page).
108


38
3)To document caregivers perceived barriers to adherence to HIV medication
regimens.
Secondary Aims:
1) To document the utility of the Adherence Interview-HIV (AI-HIV) for
obtaining information about caregiver knowledge and adherence.
2) To document the utility of a medication display card in helping caregivers
identify medication names.
3) To document the utility of the 24-hour recall interview (24RI) for assessing
adherence to pediatric HIV regimens.
4) To document the relationship between regimen knowledge and adherence to
pediatric HIV regimens.
5) To document the relationship between disease severity and adherence to
pediatric HIV regimens.
Hypotheses:
1. Caregiver knowledge of the prescribed regimen will differ significantly from
reported typical adherence behavior (AI-HIV) as some caregivers will actively
decide not to give their children some medications that may have poor taste or
significant side effects.
2. Caregiver knowledge of medication names will not differ significantly for
those who used the medication display card to identify the names versus those
who did not.


DISCUSSION
Caregiver Knowledge of the Prescribed Regimen
The first goal of the current study was to document caregivers knowledge of
medication names, dosage, dosing frequencies, dosing intervals, and medication-specific
dietary requirements. Although all but one caregiver correctly identified the prescribed
medication-taking frequency of their childs medication, 33% of caregivers were unable
to correctly identify at least one of their childs medication names, 31% of caregivers
failed to correctly identify the dosage for at least one medication, and half of the
caregivers failed to correctly identify the specific dietary requirements for at least one
medication. These results suggest that caregivers of children on ART may have more
difficulty remembering or identifying medication names and dietary requirements, and
less difficulty remembering or identifying medication-taking frequencies than adults on
ART (Bangsberg, Bronstone, & Hofman, 2002; Stone et al., 2001). Caregivers may have
more difficulty remembering or identifying specific information about medications in
their childs regimen because they are less involved in their childs regimen behaviors
than adults are in their own regimen behaviors. In the future, it may be worthwhile to
examine the relationship between caregiver knowledge and caregiver involvement.
While almost all caregivers exhibited accurate knowledge of dosing frequencies,
results indicate that some caregivers may be unaware of the exact intervals at which the
doses are to be given. When caregivers were asked to state the times of day at which the
children had typically taken their medicine during the previous two weeks, 21% of the
64