Citation
Barriers to Human Immunodeficiency Virus (HIV) Treatment Participation during the Perinatal and Postpartum Periods: A Comparison of Maternal and Infant Health Outcomes

Material Information

Title:
Barriers to Human Immunodeficiency Virus (HIV) Treatment Participation during the Perinatal and Postpartum Periods: A Comparison of Maternal and Infant Health Outcomes
Creator:
ROBINSON, PATRICIA STEARNES
Copyright Date:
2008

Subjects

Subjects / Keywords:
Disease risks ( jstor )
Dyadic relations ( jstor )
HIV ( jstor )
Infants ( jstor )
Medications ( jstor )
Mothers ( jstor )
Pills ( jstor )
Stigma ( jstor )
Substance abuse ( jstor )
Women ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Patricia Stearnes Robinson. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
7/24/2006

Downloads

This item has the following downloads:


Full Text











BARRIERS TO HUMAN IMMUNODEFICIENCY VIRUS (HIV) TREATMENT
PARTICIPATION DURING THE PERINATAL AND POSTPARTUM PERIODS: A
COMPARISON OF MATERNAL AND INFANT HEALTH OUTCOMES














By

PATRICIA STEARNES ROBINSON


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


2006
































Copyright 2006

by

Patricia Steames Robinson

































This document is dedicated to strong, healthy mothers, children, and families.















ACKNOWLEDGMENTS

I thank my committee, Dr. Sharleen Simpson, Dr. Donna Treloar, Dr. Sandra

Seymour, and Dr. Nabih Asal, for all of their patient work with me. I am also thankful

for the support and encouragement of the faculty and the students of the FAM

Department in the College of Nursing. They demonstrated an endless ability to listen to

complaints about the rigors of doctoral work. I am thankful for the examples set by the

doctoral faculty, and especially the professionalism of Dr. Ann Horgas and Dr. Shawn

Kneipp. I am also thankful to Dr. Hossein Yarandi for his significant statistical help

(p<0.0001). I thank Dr. Stacey Langwick for the lessons in language, views, humility,

and mostly for "getting me." This research could not have been completed with out the

support of Catherine Lamprecht, MD and Alelia Munroe, MPH.

Most of all, I thank Harriet Miller, Dr. Edie Devers, and Sr. Louise McEachern

for being both my family and my strongest supporters. I must also acknowledge my

career mentors and personal inspirations, Dr. Jo Snider and Dr. Linda Hennig. All of

these people saw more in me than I saw in myself. For this reason, I thank them

absolutely.















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ................................ .......... iv

LIST OF TABLES ............. ................................ viii

LIST OF FIGURES .......... .. ...................... .. .......... ix

ABSTRACT ............ ............................................. x

CHAPTER

1 INTRODUCTION .......................................... 1

Problem Statement ........... ....................................... 2
Purpose....................................................................... 3
Hypothesis ............ ............................................. 3
Variables ........... ..............................................4
Terminology........................................ 4
Assumptions .............................. ................... 6
Limitations ............ ............................................. 6
Background and Significance ................ ........................ 6
Theoretical Framework ............................................. 10
Concept Definitions ............. .. .. ......... .......... 13
Operationalized Empirical Indicators ................................. 13

2 REVIEW OF THE LITERATURE ..................................... 15

Purpose............................................................15
Perinatal Transmission .............. ..................................15
Highly Active Anti-Retroviral Therapy ................ ................ 16
Risk and Prevention ................ .................... ......... 17
Adolescent Adherence ............................................... 18
Adherence in Other Marginalized Groups .............................. 22
Stigma Theory and HIV .............................. ............. 24
Psychological Distress and HIV ........................................28
Minorities and HIV ................. ............................ 29
Positive Adaptation ................. ............................ 33
Mothering by HIV-Infected Women ................ .................. 34


v











3 M ETHODS ............. ................... ....... ............. 37

Purpose............................................................37
Design ........... ...............................................37
Setting ........... ...............................................38
Subjects ........... ..............................................39
Sample Selection ............. ....................................39
Inclusion Criteria ................. ........................... 39
Exclusion Criteria ................................ .. ......... 39
Data Analysis ................. ............................... 40
Procedure ............... ............................. 41
Data Collection .......... ....... ................ .............. 41
Potential Health Risks ................ ................ .......... 42
Potential Health Benefits ............................................ 42
Potential Financial Risk ................................ .............42
Potential Financial Benefits .......................................... 42
Conflicts of Interest ............ ...............................42

4 RESULTS ..................................................... 43

Purpose............................................................43
Description of Sample ................................ .......... 43
Variables ............ ...................................... .44
Hypothesis ......... ......................... .... .......... 44
Research Question One .......... ............. ....... 45
Research Question Two ................. .................. 46
Research Question Three .......................................... 48
Research Question Four ........................................... 50
Support for Hypothesis .............. .................... ......... 51

5 DISCUSSION AND IMPLICATIONS ................................. 53

Purpose............................................................53
Descriptive Statistics ............. ....................................55
Laboratory Results ................................. ............. 55
M education Adherence ............................................ 56
Appointment Adherence .......................................... 58
Summary of Findings ............................................. 58
M missing and Excluded Data ........................................... 59
Ethical Considerations .............................................. 60
Significance .................. ...................... ............ 64
Implications for Future Research ................ ................... .. 67
Suggestions for Further Research ................ ................... 70




vi









APPENDIX

A DATA COLLECTION INSTRUMENT ...............................72

B EXPECTED VALUES FOR DATA COLLECTION ........................ 73

REFERENCES .............. ............................ .75

BIOGRAPHICAL SKETCH ............................................. 79















LIST OF TABLES


Table page

4-1 Number and percent distribution of maternal sample by age and race ........ 44

4-2 Number and percent distribution of maternal sample by dichotomous
variables ........... ........................................... 44

4-3 Distribution of cases and comparison groups by independent variables of
exposure to barriers and exposure to facilitators, and by outcome measures of
favorable laboratory results, medication adherence, and appointment
adherence ........... ..........................................46

4-4 Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure of
laboratory results .................................. .......... 47

4-5 Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure of
medication adherence ................................ ......... 48

4-6 Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure of
appointment adherence .................. .............. ....... 48

4-7 Relationship between structural barriers and structural facilities and the
outcome measures of laboratory results, medication adherence, and
appointment adherence ........................................... 49

4-8 Summary of logistic regression analysis predicting favorable laboratory
results ............... ............................. 50

4-10 Summary of logistic regression analysis predicting favorable appointment
adherence ............. .........................................52















LIST OF FIGURES

Figure page

1-1 Routine prenatal screening and care of HIV-positive women ................ 7

1-2 Stigma theory ........... ....................................... 11

1-3 Proposed revision to stigma theory .......................... ...... .12

3-1 Subject selection ................................... ......... 38

5-1 Stigma theory in this study ......................................... 54

5-2 Florida Department of Health announcement, circa 1995 .................. 66

5-3 Florida Department of Health announcement, circa 2005 .................. 69















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

BARRIERS TO HUMAN IMMUNODEFICIENCY VIRUS (HIV) TREATMENT
PARTICIPATION DURING THE PERINATAL AND POSTPARTUM PERIODS: A
COMPARISON OF MATERNAL AND INFANT HEALTH OUTCOMES

By

Patricia Stearnes Robinson

May 2006

Chair: Sharleen Simpson
Major Department: Nursing

Public health efforts have yielded a dramatic decrease in the number of infants

born Human Immunodeficiency Virus (HIV)-positive through early identification of

maternal HIV status during pregnancy. Despite this overwhelming success with fetal and

infant outcomes, maternal outcomes remain dismal. The purpose of this study was to

determine if previously published barriers to HIV treatment adherence affect the perinatal

dyad, the postpartum mother, or the exposed infant equally; and to determine if the

proposed constructs of structural facilitators or structural barriers affect adherence in the

perinatal dyad, the postpartum mother, or the exposed infant. A descriptive and analytic

case comparison study was conducted on 100 pregnant women, and they and their infants

were followed from entry into the high risk obstetrical clinic with a diagnosis of HIV-

infection to 18 months postpartum. All data collected were collected through

retrospective chart reviews. Data were analyzed through both descriptive statistics and









logistic regression analysis. Structural facilitators (transportation to clinic, Targeted

Outreach to Pregnant Women Act involvement [to provide transportation, emotional

support, and/or expedited paperwork], HIV nurse and social worker involvement,

medication dispensed [instead of prescriptions given], primary care available onsite,

intake paperwork not necessary, and/or previously established and ongoing relationship

with case manager) were associated with infant adherence to HIV-related care, and

structural barriers (insurance companies, including Medicaid HMOs, that limit access to

comprehensive care in the subspecialty HIV clinic or requirement of an individual to link

herself with follow-up care, including the requirement of self-initiated case management

before clinic appointment is given) were associated with decreased maternal adherence in

the postpartum period. Suggestions were made for further research.















CHAPTER 1
INTRODUCTION

Human immunodeficiency virus (HIV) infections continue to increase in

incidence and prevalence across minority populations in the United States (Centers for

Disease Control [CDC], 2005). In minorities, 64% of transmissions occur in women,

90% of whom were infected through heterosexual contact, with 74% of those infections

occurring in non-Hispanic blacks (CDC, 2005). Once infected, the death rate from

acquired immune deficiency syndrome (AIDS) for black women in the United States is

nine times higher than for Caucasian women (CDC, 2004). Minority adolescent females

demonstrate the most rapidly increasing incidence of new infections (Rogers, 2001).

The HIV diagnosis usually comes as the result of routine prenatal testing. It is not linked

to symptoms experienced or any perceived high-risk behaviors; rather the tests are

mandated by public health policy to reduce rates of transmission to fetuses. When a new

HIV diagnosis is made, pregnancy is turned into pathology for many minority women. It

is particularly hard to engage and retain young postpartum mothers in care, because they

do not link well to the current network of adult HIV treatment centers designed to treat

men.

HIV-positive women are disenfranchised from both society and treatment

facilities due to multiple inequalities. Stigma related to race, income, age, gender,

education level, early-onset parenting, sexuality, substance abuse, and HIV infection are

present, and a perceived indifference by the medical community is often reported

(Ingram & Hutchinson, 1999a). Conventional wisdom within the treatment community







2

reiterates that adolescents are likely to fail to adhere to treatment because of these

barriers (Rogers, Miller, Murphy, Tanney, & Fortune, 2001). In addition, denial of the

HIV diagnosis and refusal to act in their own best interest are considered important

reasons for continued risky behaviors (Rogers et al., 2001). Yet, at the same points in

time, young mothers often care for their exposed infants with vigor and success that is

not evident in their own care. This raises a number of questions. Specifically, what are

the conditions and attributes that allow young mothers to provide this specialty care

successfully for their infants, and what are the barriers that prevent them from

participating in self-care? Why are treatment plans that are effective for other

populations not effective for minority adolescent mothers?

Problem Statement

The health care system has strategies in place to identify HIV-positive women

during pregnancy for the purpose of preventing viral transmission to exposed fetuses.

Structural facilitators are in place to increase the ability of HIV-positive pregnant women

to comply with prenatal and infant care. These facilitators include automatic referral to

the HIV treatment team in the High Risk Obstetrical Clinic (HROB) upon diagnosis,

immediate access to antiretroviral medication (ART), fast track paperwork to gain

emergency Medicaid coverage, field workers from the Targeted Outreach to Pregnant

Women Act (TOPWA) to provide support and transportation to clinic and financial

appointments, and a registered nurse and a social worker who both attend HROB.

Once born, neonates receive their initial HIV blood test, a two week follow-up

appointment at the HIV specialty clinic for results, and ART is dispensed for use at

home. Unlike their neonates, postpartum mothers have no automatic referral to the HIV

specialty clinic, and have no mechanism in place to ensure they obtain ART prior to







3

discharge after delivery. Postpartum mothers are often lost to follow-up after delivery

until they are pregnant again or until they present for care after the onset of HIV-related

symptoms. The interruption of care between delivery and symptom onset has the

potential to create the burdens of increased utilization of tertiary health care resources,

infant/child abandonment or orphanage, and personal suffering for the mothers. The lack

of maternal self-care is attributed in the literature to poverty, education, youth, and

clinical depression (Murphy, Wilson, Durako, Muenz, & Belzer, 2001). These maternal

characteristics are shared by both the mother and the baby postpartum, but may not

always serve as barriers to health care for them both.

Purpose

The purpose of this study was to determine if previously identified barriers to

HIV treatment adherence affect the perinatal dyad, the postpartum mother, or the

exposed infant equally; and to determine if the proposed constructs of structural

facilitators or structural barriers are associated with HIV treatment adherence in the

perinatal dyad, the postpartum mother, or the exposed infant.

Hypothesis

The following two-tailed hypothesis was tested: Maternal depression or other

mental illness, proximity of HIV diagnosis, race, income, adolescence, high school

education, substance abuse, pill burden, structural barriers to access, and structural

facilitators to care are associated with a significant difference in adherence as measured

by the laboratory results, missed medication doses, and delayed or missed clinic

appointments of HIV-positive pregnant women, postpartum mothers, or their infants.

In order to determine the answer, the following research questions were posed.









1. Are the three groups (postpartum mothers, infants, and dyads) homogeneous with
respect to the three outcome measures (laboratory results, medication adherence, and
appointment adherence)?

2. Is there a significant relationship between the outcome variables (laboratory results,
appointment adherence, and medication adherence) and demographic variables (age,
race, mental illness, poverty, new diagnosis, pill burden, high school education,
substance abuse) among the three groups?

3. Is there a significant association between the outcome variables (medication
adherence, laboratory results, and appointment adherence) and structural facilitators
and structural barriers?

4. Is maternal depression or other mental illness, proximity of HIV diagnosis, race,
income, adolescence, high school education, poverty, substance abuse, pill burden,
structural barriers to access, and structural facilitators to care significantly different
as measured by the laboratory results, medication adherence, and appointment
adherence in HIV-positive pregnant women, postpartum mothers, or their infants?

Variables

The independent variables are maternal depression or other mental illness,

proximity (newness) of HIV diagnosis, race, income, adolescence (age), high school

education, substance abuse, pill burden, structural barriers to access, and structural

facilitators to access. The dependent variables are consistent, but defined differently for

each group because the standard of care is different during each period. These outcome

measures include laboratory values, percentage of missed doses, and delayed or missed

appointments for care.

Terminology

For the purpose of this study, the study variables are operationalized as follows.

* Adolescence-maternal age less than 24 years at time of conception

* Delayed care-HIV-related care not provided during the recommended window of
time, or postpartum maternal care or infant immunizations not provided during the
recommended window of time

* Dyad-unit that includes the pregnant woman and her fetus during the prenatal and
antenatal periods









* Education-attainment of a high school diploma, GED, or current enrollment in an
equivalent program

* Income-household income less than poverty threshold for the year under
investigation

* Laboratory results (dyad)-CD4 and HIV viral loads within expected range based on
CDC information for the year under observation

* Laboratory results (infant)-HIV DNA PCRs remain negative at the three
recommended intervals, and HIV ELISA /Western Blot is negative at 18 months

* Laboratory results (maternal)-CD4 and HIV viral loads remain within expected
range based on CDC recommendations for the period under observation

* Maternal depression or other mental illness-Any psychological diagnosis in the
prenatal medical chart that is found in DSM-IV, whether or not treated by medication
or therapy

* Missed appointments-appointments missed and not rescheduled >10% of the time

* Missed doses->10% of ART doses missed by self report at clinic visit

* Pill burden-number of pills/doses per day greater than established norm during the
year prescribed (i.e., 5 Viracept and 1 Combivir tabs twice daily = pill burden of 12
tabs/day = average prenatal pill burden from 2000 to 2004; and AZT syrup every 6
hours = 4 doses/day = average neonatal pill burden during entire study period)

* Proximity of HIV diagnosis-determination of whether HIV seropositivity was
established during the current pregnancy; a measure of the newness of the diagnosis

* Race-self-reported race (in accordance with federal guidelines)

* Structural barriers to access-Insurance companies, including Medicaid HMOs that
limit access to comprehensive care in the subspecialty HIV clinic or requirement of
an individual to link herself with follow-up care, including the requirement of self-
initiated case management before clinic appointment is given

* Structural facilitators -transportation available to clinic, Targeted Outreach to
Pregnant Women Act (TOPWA) involvement to provide transportation, emotional
support, and/or expedited paperwork, HIV nurse and social worker involvement,
medication dispensed (instead of prescriptions given), primary care available onsite,
intake paperwork not necessary, or previously established ongoing relationship with
case manager


* Substance abuse-illicit drug or alcohol use during pregnancy







6

Assumptions

As with any retrospective chart analysis, it was assumed that information

contained in the charts was accurate and complete in both reporting by clients and

recording by staff.

Limitations

The generalizability of this study is limited to women in social and economic

circumstances similar to those of the subjects. Incarcerated and/or toxicology screen-

positive mothers at delivery were excluded from this study, as were neonates born with

prematurity significant enough to prevent mothers from providing care for their infants.

Though foster parents or extended family and friends care for many HIV-exposed

infants, this study sought to examine mothers who were primary care providers so that

infants and mothers shared the same maternal risk factors at the same times. This limits

this study's ability to describe the circumstances of broken families. Also excluded were

the sparse subjects that elected to seek their specialty care from community providers.

Background and Significance

The public health and medical communities have given extensive effort to the

protection of fetuses exposed to HIV. The gold standard of care in the United States is

the public health initiative to engage the mother-baby dyad in the prevention efforts

detailed in the Pediatric AIDS Clinical Trial Group Study Protocol 076 (PACTG 076).

This effort focuses on the outcome measure of prevention of HIV transmission to

exposed babies. The protocol is overwhelmingly successful in its intent to reduce

transmission within the United States, reducing the rate from a natural history of 25-28%

(NIH/NIAID, 1999) to about 2% (NIH/NIAID, 1999). The follow-up study, PACTG

219C, continues to look at the health of the exposed infant as it grows into adulthood






7
(NIH/NIAID, 1999). Neither study considers the outcome of maternal health. Since this
research produces such convincing evidence of its efficacy, funding and policy decisions
are made based on adherence of clinics to this protocol. In the era of evidence-based
practice, the separation of maternal and child health at the time of delivery may
unintentionally inform providers' decisions to cease to address maternal health after
delivery. The treatment guidelines continue in a linear fashion along the prevention of
transmission, stating guidelines for medicating the neonate with antiretroviral
medication, with no recommendations for maternal care after delivery (Figure 1).
Mothers cease to be treated by the protocol for their own infections, because they cease
to be regarded once they are not carrying babies, and thus not a threat of transmission to
children.



--


Figure 1-1. Routine prenatal screening and care of HIV-positive women







8

Yoder (2002) contends that participation in the social institutions of medicine and

public health can shape values, assumptions, explanatory models, and problem-solving

approaches of providers. Altruistic, novice providers who enter the field of HIV

treatment and prevention are quickly influenced by the existing treatment culture and

explanatory models. Mothers that do not follow-up with their own HIV-related care after

delivery are often viewed by providers as being intrinsically unmotivated or self-

destructive. Providers often feel that the amount of attention and effort that has been

given to mothers during the prenatal period should be enough to retain them in care after

delivery. So, their failure is explained away in the literature by describing them as likely

to fail based on their intrinsic characteristics (Andersen, 1995) without looking at why

that is so. Often, intrinsic factors such as race, income, and age are used to describe

away failure, without examination of the context in which the failure occurs. The

healthcare community creates this context. Thus, it is important to challenge the existing

beliefs and actions of the public health community and health providers who dismiss the

mothers as unreachable by looking at the specific circumstances that contribute to their

health outcomes. This requires that research take into account structural components that

may inform maternal choices.

The public health community determines when treatment occurs, when it does

not, and on what priorities research funds are spent. The prevalent themes in the

literature that underlie beliefs about HIV prevention, infection, and adherence should be

compared to the life circumstances of minority, young females infected with HIV.

Before a vulnerable population is dismissed as refractory to help, an earnest scientific







9

effort should be employed to understand the fit between the system and these mothers. It

is imperative to stop dismissing as unreachable those who may simply be misunderstood.

Because of the repetitive claims that minorities cannot or do not adhere to

therapy, studies show that minorities are less likely to receive anti-retroviral medications

(Cargill, Stone, & Robinson, 2004), though these are the people most likely in need of

therapeutic efforts (Basu, 2004). Outside of the United States, populations previously

dismissed as unreachable are showing significant adherence through the removal of

structural barriers to care. In the central plain of Haiti, adherence to ART heightened in

the presence of the concomitant provision of food, opportunity to generate income, and a

continuity and parity of health care distribution (Farmer, 2003). To understand the

dismissal of adherence potential in HIV-infected young mothers, a look at the history of

HIV in the United States is helpful.

HIV first appeared in the United States in the early 1980s as a disease that

infected and killed males who were identified as having sex with other males (MSM)

(CDC, 2004). Soon after, HIV and the resultant AIDS started infecting and killing

intravenous drug users (IVDUs), as well as the female sexual or needle-sharing contacts

of both groups (CDC, 2004). HIV in the United States is well established through this

history as a disease of stigmatized and vulnerable groups. In 1981, two physicians in

New York requested funding from the National Institutes of Health and the American

Foundation for AIDS Research to investigate the particular effects of HIV on women.

Both grantors rejected the proposals, stating on the return critiques that the effect of HIV

infection on women was not an urgent topic for study (Bunting, 1996). Research money

and effort in the United States and the world have stayed focused on the treatment of

infected men and the prevention of transmission to fetuses, while the epidemic itself has







10

moved disproportionately to killing young minority women (CDC, 2005). Researchers

are now faced with the majority of new infections occurring in minority populations

within the United States-populations that likely have different priorities and

circumstances than are addressed by the current treatment guidelines. The first step to

examining the fit between treatment centers, policy, and young mothers is to explore and

describe any differences that may exist in the way the HIV treatment network treats

young mothers, and any possible subsequent differences in outcomes.

Theoretical Framework

Stigma theory is a middle range theory that was first published in sociology

literature by Goffman (1963). Stigma was defined by the Greeks as a bodily sign of a

moral flaw Goffman (1963), and was a common concept in the field of social psychology

in the 1960s. It was developed in to a sociological theory to explain the causes of

discrimination in society (Goffman, 1963). Concepts in the theory are the stigmatized

attribute, discredited state, discreditable state, stigma, and discrimination (Figure 2).

These concepts are incorporated into the theoretical framework of this study, along with

four additional concepts that begin to allow for an individual's response to being

stigmatized. Goffman's (1963) stigma theory is comprised of the following relational

statements:

* A stigmatized attribute may lead to discredit.

* Discredit causes stigma.

* Stigma justifies discrimination.

* Discrimination then occurs.

* Discrimination becomes part of an individual's or group's environment, the effect of
which may increase the stigma experienced.





































Figure 1-2. Stigma theory (Goffman, 1963)

What happens if a decision is made not to discredit someone with the potential for

a stigmatized attribute? It is possible to confer on them protection from the already

stigmatized source of this potential stigma. One of the purposes of stigma is to separate

normals from the undesirable outcome, or the risk of becoming undesirable One way that

normals distance themselves from perceived risk is to blame stigmatized individuals and

groups for having the very trait that renders them stigmatized (Goffman, 1963). Because

groups are already blamed for being HIV-positive, other groups that are at risk of

becoming infected by these flawed individuals would logically be viewed (intentionally

or unintentionally) as victims.







12

It is plausible that there is an actual protective effect of being perceived as a

victim, as much as there is a harmful effect of being perceived as stigmatized or flawed.

In this study, a model is proposed (Figure 3) that is reflective of both sides of a potential

stigmatizing attribute-one that reflects possible positive and negative outcomes, simply

dependent upon how the health care community and health care policy makers regard the

individual or group at entry into care.


Figure 1-3. Proposed revision to stigma theory, as adapted from Goffman (1963)









Concept Definitions

* Stigmatized attribute-Characteristic that reduces a person from "whole and usual to
discounted and tainted"; and that may be a perceived physical flaw, character flaw, or
tribal membership (Goffman, 1963). In his original work, examples of physical flaws
were disfigurement or blindness; examples of character flaws would be
homosexuality or being a convicted criminal; and examples of tribal flaws would be a
trait passed through family or tradition, such as skin color or religion.

* Discreditable state-State in which an individual with a stigmatized attribute that can
be hidden exists. Hiding an attribute such as HIV infection prevents being
discredited and possible stigma. The state involves increased stress and worry of
being discovered.

* Discredited state-State in which an individual exists when a stigmatized attribute is
known to normals.

* Stigma-A belief or an attitude that a person is less than human that is constructed
and held by normals as justification for discrimination, or a situation in which an
individual is denied full social acceptance (Goffman, 1963).

* Discrimination-The act of prejudicial treatment.

* Protected class state-Membership in a group that has a stigmatized attribute, yet is
viewed as a victim of stigmatized individuals instead of as a stigmatized individual.

* Diminished health outcome-Final health status resulting from failure to receive
health care and support to the extent they are available for individuals with similar
medical conditions.

* Increased protective effort-Structural facilitators outnumbering those available to
other individuals and groups with the same stigmatized attribute; structural barriers
are less than those encountered by other individuals and groups with the same
stigmatized attribute.

* Optimized health outcome-Final health status resulting from best possible access to
health resources and support as compared to individuals with similar medical
conditions.

Operationalized Empirical Indicators

* Stigmatized attributes-Racial minority, aged 15-23 years, pregnant, HIV-positive,
income below poverty level, education below high school level, documented history
of drug or alcohol use during pregnancy, or a documented history of mental illness.







14

* Discreditable state-An HIV-positive mother who does not disclose her diagnosis to
providers and/or to her community, or an HIV-exposed fetus or infant whose status is
either unknown or is newly known upon disclosure of mother's HIV status.

* Discredited state-A pregnant woman or mother whose HIV-positive diagnosis is
known to providers and/or her community.

* Stigma-health provider and/or minority community beliefs and attitudes that HIV-
positive females should not reproduce, are to blame for their infection, and are less
valuable than their HIV negative counterparts. Since stigma is defined as a feeling or
an attitude, it is not measured in this review of medical documentation.

* Discrimination-Absence of services available to others that facilitate participation in
HIV-related care or presence of policies that delay or prevent participation in HIV-
related care (specifically listed under variable definitions).

* Class protective effect-Presence of services that facilitate participation in HIV-
related care or absence of policies that delay or prevent participation in HIV-related
care (specifically listed under variable definitions, structural barriers and structural
facilitators).

* Diminished health outcome-Missed appointments, high viral loads and/or poor CD4
cell counts by laboratory measure, or missed doses of antiretroviral medications.

* Optimized health outcome-Adherence to recommended appointments, therapeutic
viral loads and CD4 cell counts by laboratory measure, and adherence to
antiretroviral medications.

Though HIV-related literature is peppered with Stigma Theory, other reasons for

treatment failure prevail. Most studies focus on health behavior theories that emphasize

individual or intrinsic empowerment. Many studies detailed in the next chapter describe

intrinsic qualities that make individuals likely to fail treatment. No study was identified

that specifically measured the role of structural or extrinsic barriers to care. It may not be

just the labeling or stigma that keeps HIV-positive mothers out of care, but rather the

preferential treatment others may receive in their presence if these same mothers are

denied equal treatment.















CHAPTER 2
REVIEW OF THE LITERATURE

Purpose

The purpose of this study was to determine if previously identified barriers to

HIV treatment adherence affect the perinatal dyad, the postpartum mother, or the

exposed infant equally; and to determine if the proposed constructs of structural

facilitators or structural barriers are associated with HIV treatment adherence in the

perinatal dyad, the postpartum mother, or the exposed infant.

Perinatal Transmission

Positive effects on fetal outcomes (HIV status) were reported by the first pediatric

clinical drug trials if pregnant HIV-infected women took zidovudine (AZT) during the

course of their pregnancy and delivery. The odds of an infected baby being born to an

infected mother decreased from about 30% to about 2% with the use of AZT (CDC,

2004). This was a landmark finding for prevention of pediatric AIDS, and was viewed as

the end of AIDS in children. Since children would no longer contract HIV from their

mothers, the reasoning followed that there would no longer be pediatric HIV in the

United States. Many states began to require HIV testing of pregnant females for the

protection of the fetuses, and thus the large cohort of infected, minority, pregnant

adolescents and young adults continue to be uncovered through diagnosis during

pregnancy (CDC, 2004). Existing literature did not reflect the real risk that without

changing social and economic circumstances, these same HIV-negative babies and their







16

friends would grow into adolescents and acquire their own HIV infections because the

same social and economic challenges that plagued their mothers now place them at risk.

Perinatal HIV literature has an overwhelming focus on the prevention of HIV-infection in

the fetus/infant.

Highly Active Anti-Retroviral Therapy

It is well established that aggressive anti-retroviral therapy, commonly referred to

as highly active anti-retroviral therapy (HAART) increases both the duration and quality

of life (Rogers, 2001). Implementation of HAART at the clinically indicated time is the

standard of care for infected individuals. An entire subset of social science research

focuses exclusively on ways to increase adherence to HAART. Models have been

developed to assess readiness and stage change for maximum success with therapy

(Glanz, Rimer, & Lewis, 2002). In fact, current treatment guidelines strongly

recommend that HAART not be initiated on patients until they demonstrate a readiness to

comply with therapy for the long term (Steinhart, Orrick, & Simpson, 2002, p. 58). This

recommendation is based on what is viewed as best for the patient and society. Partial

compliance or short duration therapy often results in mutated viruses that are resistant to

available medication (Steinhart, Orrick, & Simpson, 2002, p. 103), and multiple drug

resistant strains of HIV present insurmountable treatment obstacles, increasing the

burdens of both cost and mortality to society. The sentinel inconsistency in the

guidelines is that the only demographic for whom readiness is not assessed, and patient-

centered counseling above and beyond the required post-test (results) is not the norm, is

pregnant females (Steinhart, Orrick, & Simpson, 2002, p. 58). According to the theories

discussed below, these are the most personally at-risk of the infected population related

to both their behavior and biology.







17

Risk and Prevention

The only scientifically validated risk factor for contracting HIV is engaging in

any activity that exchanges infected bodily fluid with someone who has HIV (termed

high risk behaviors) term (Steinhart, Orrick, & Simpson, 2002, p. 13). Since secondary

prevention (limiting the spread of HIV by identifying those already infected) is the

prioritized funding goal for HIV, the question to answer is why certain groups engage in

high-risk behaviors more often than others. Basu (2004) found that the top

epidemiological predictor for HIV worldwide is a low-income level. It is this low-

income level that decides the context and conditions of sexual behaviors. The poor have

an increased risk of exposure to pathogenic situations, whether sanitation, crowding, or in

this case, the presence of more HIV-infected contacts is the cause. In the United States,

being black is considered a risk factor for HIV, because being black is often used

interchangeably with the socioeconomic variable of being poor. Nazroo (2003) argued

against being black as a risk factor for disease by showing that after income adjustment,

health disparities within the United States leveled considerably. The balance of the

difference may be attributable to experiences and awareness of racism (Nazroo, 2003).

Krieger (1994) asserts that population patterns of health stem from economic and

social activities and inequalities. Behavior models that are the backbone of health

sciences assume individual agency that may not exist without public health efforts to

identify and remove barriers. Often, the community of scientists that ration resources to

the HIV-infected community disregards or negatively regards individuals who are in the

wrong group at the wrong time. Basu (2004) recounts two studies, one in which a miner

explains the context of his high-risk recreational behavior (his profession has an injury

rate of 42% and he is never certain he will live through his work day), and one in which







18

prostitutes complain of their lack of opportunity. The miner is labeled by the

psychologists as "in denial" and tagged with a "low self-esteem." The researchers label

the prostitutes as "liars" who are denying their own agency when they claim that

prostitution is the only source of available income. Basu (2004) calls AIDS the symptom

in these situations, not the disease.

Adolescent Adherence

A review of the literature was conducted using the CINAHL and Medline

databases dated 1997 to 2004 with a limitation to English language. Keywords

"adolescent," "adherence," and "HIV" were utilized. Findings were limited, and of

those, few reported successful methods of adherence. Most literature reported reasons

for failure instead of ways to promote adherence. Though multiple centers specialize in

adolescent HIV identification and treatment throughout the country, only one group of

providers have joined to formalize, test, and publish findings. Members of the REACH

Project (Reaching for Excellence in Adolescent Care and Health) comprise this group

(Murphy et al., 2001).

No research conducted by individual providers was identified in the literature.

This is probably related to the federal mandate that HIV funding to go to sub-specialty

treatment centers, and the resultant tendency of uninsured and underinsured youth to seek

care at these centers. Even with sub-specialty care, adolescence and motherhood remain

turbulent times in the lives of HIV-positive women. Pinch (1994) discusses the

vulnerability of adolescents in particular, stating that the invincibility perceived by

adolescent girls makes them a special concern with regard to their own HIV infections.

An additional concern is the inability for many adolescent females to problem-

solve due to psychological immaturity and incomplete educations (Pinch, 1994).







19

Looking at the theoretical perspectives to adherence to treatment in HIV-infected

adolescents, models in the literature generally discuss adolescent development and

behavior theories appropriate to reach adolescents and affect desired outcomes. As

mentioned earlier, one program of researchers in the United States has committed

themselves to researching the specific needs of adolescents with HIV infection.

The group of providers most often seen in the literature is physicians who provide

care at 15 clinics in 13 major cities throughout the United States to form the REACH

Project). Their findings report a strong association between medication adherence and

reduced HIV viral loads (Murphy et al., 2001). Findings also showed less than 50% of

subjects reported acceptable adherence to their treatment regimen, and that adherence

seems to be a serious problem among HIV-positive adolescents (Murphy et al., 2001).

The researchers cited both ease of medication use, also known as decreased pill burden,

and treatment of depression as two important issues to be addressed when treating

adolescents (Murphy et al., 2001). Though the REACH Project providers come closest to

addressing the population of pregnant adolescents, their findings do not provide separate

statistics on pregnant adolescents, nor do they suggest interventions specific to the needs

of pregnant adolescents.

In the analysis by Murphy et al. (2001), strong social support was reported as

necessary to adherence, and a linear decrease in adherence was noted when regimens

became more complex. This is not surprising, though it is important to note that

regimens for pregnant women are geared toward fetal benefit, and do not usually take pill

burden issues into consideration. This is another possible explanation for treatment

failure. This lack of consideration of maternal needs indicates a failure to consider

barriers to maternal long-term survival.







20

A subset of REACH providers in Los Angeles performed a simple survey of 31

HIV-positive youth from their clinic to identify factors associated with adherence

(Belzer, Fuchs, Luftman, & Tucker, 1999). By self-report, 22% of the youth reported "I

forgot" as the reason for missing ART, another 15% reported depression, and 43%

reported "too many pills," though the actual pill burden was not reported by the authors

for analysis (Belzer et al., 1999).

At one site the REACH providers developed and tested a program to address

adherence problems in adolescents called Therapeutic Regimens Enhancing Adherence

in Teens (TREAT) Program (Rogers et al., 2001). The program includes education on

benefits of compliance, psychological assessment and support to develop readiness for

therapy and to address depression, and tools and reminders, such as signaling watches

and pagers (Rogers et al., 2001). Rogers et al. (2001) released a description and

preliminary results of their adolescent adherence program based on the Stages of Change

Model discussed below. While theoretically, this program and approach should work to

increase adherence, it has not been shown to be effective in their cohort in the short term.

Their findings indicated less than 40% of participants were adherent to HAART, despite

intense interventions. Their recommendations include avoidance of the harmful effects of

premature prescription of HAART, further suggesting that the degree to which youth

accept their diagnosis predicts success on HAART (Rogers et al., 2001). Though not

discussed by the authors, this recommendation was not extended to pregnant adolescents

because of secondary prevention policies.

In addition to depression and forgetfulness, Futterman (1999) cites distrust of

medications and the medical establishment among teens of color (94% of infected

adolescents) as a reason for adherence problems. Finally, in a study yielding contrasting







21

results, two pharmacists conducted a study of 25 patients, ages 9-21 years using a pager

as a reminder for medication doses over a three-month period (Todd & Miller, 2000).

Their findings showed that compliance improved anywhere from 45-50% from baseline,

using only self-report as a measure. These reported increases in adherence were

supported by serologic improvements including lower viral loads and higher CD4 counts

(Todd & Miller, 2000).

In summary, there exists a paucity of research addressing compliance

improvement in adolescents. No research focuses specifically on the issues faced by

adolescent mothers. Much of what was identified focused on the assessment of

"readiness," or psychological measurements to determine if a youth was ready to start

taking medications based on whether they were willing and able to comply with

medication therapy, thereby reducing self harm over time in the form of ART resistance.

Pregnant HIV-positive teens get no assessment of "readiness," as the focus is on the

health of the fetus. Nothing addresses continuation of therapy when the teens are started

on ART precipitously and without psychological preparation or acceptance of their

diagnoses.

When it comes to individual adherence, most researchers ground their studies in

health behavior models. Willey et al. (2000) used the Stages of Change Model, or

Transtheoretical Model to assess why patients fail to take medications as prescribed.

They studied 161 HIV-infected patients using self-report of compliance and electronic

monitoring of adherence (electronic reminders and dose counters), and compared their

existing stages of readiness to comply with their actual compliance behavior. The

assumption in this study is that prescribing HAART is tantamount to prescribing a

lifestyle change, and should not be done without assessing the readiness of the individual







22

to accept this proposed change. For pregnant girls in particular, no such preparation is

made. The assumption is that the girls must change for the sake of their fetuses,

regardless of readiness. This is one explanation of the failure of maternal treatment

continuity. Their findings showed that reminders and dose counters, mainstays in the

adherence community, do not work if participants have not advanced to the action and

maintenance stages of change.

Adherence in Other Marginalized Groups

Tross (2001) published the findings of her ethnography of inner-city Hispanic

women within one zip code in New York City. Because of the limited geography and

comprehensive techniques, she draws a vivid picture of the barriers to sexual power and

health negotiation for these women. She found that despite high-risk lifestyles and

knowledge of HIV, none of the women perceived themselves to be at risk. She also

documented the absence of discussion of sex or HIV between these women and their

partners. According to participants, such discussions at times ended with violence

(Tross, 2001). Methodology was comprehensive in this study, utilizing the Community

Identification Process (CID) and included a review of community records, individual

interviews, and focus groups. Methods included identification of gatekeepers, in this

instance, the owner of a bodega and the owner of a laundermat. In both cases, the

community identified self-made women of substance in the community as leaders. Four

different levels of participants were identified: key participants, system interactors,

project staff, and the gatekeepers. This body of work contributes greatly to the sparse

body of knowledge regarding women with HIV by applying the inductive approach to

HIV prevention within a small community.







23

Specific interesting findings include the concept of support groups as a "whitey

thing," with a preference shown to provide support where women already go-the

supermarket, daycare, schools, and the laundermat. The women also expressed the desire

for a female condom that could be applied before engaging in foreplay or sexual activity,

and removed after her male partner has left. This illustrates the awkward nature of trying

to negotiate safer sex with intimate others in the community. These findings support the

idea of social marketing to change the attitudes of a community, as individual change is

predicated on community boundaries. They also support listening to group feedback on

the kind of care in which they would participate.

The Henry J. Kaiser Family Foundation (1999) also sought to provide a

comprehensive description of a "community" by conducting interviews and focus groups

on "higher risk" teens and their providers. Though the study lists no specific

methodology, it is ethnography-like in its scope. The study included groups with

demographic characteristics representative of overall HIV prevalence, so there is not a

preponderance of African American female adolescents. Still, there are 30 females, and

about half are African American. Again, despite repeated high risk behavior and a

cognitive understanding of what constitutes high risk, these adolescents do not view

themselves as being personally at risk. It is the only study specifically addressing

qualitative content of adolescents and HIV. The point is also advanced that teens

themselves express a feeling of hostility and judgment from the healthcare system, and

state "teen-friendly" care is a requirement for them.

One interesting finding is the concept of using condoms to "protect" oneself from

a sexual partner. Teens do not view sex partners as unclean or dangerous, and therefore

do not use condoms for "protection" (Henry J. Kaiser Family Foundation, 1999).







24

Different messages are either effective or fall on deaf ears, depending on how the

population feels they are regarded, and the context into which the message is received.

Stigma Theory and HIV

Many researchers have described the conditions experienced by HIV positive

people as being consistent with Stigma Theory (Ingram & Hutchinson, 1999a). It has

been applied to explain how the health of mothers is overlooked in the interest of their

fetuses. Taylor (2001) reported that sick women are differentiated from well women

during pregnancy by the mere diagnosis of HIV infection. Bunting (1996) noted that the

ability to disregard the needs of mothers is supported by stigmas that render her

discounted and tainted. Fathers, the primary source of infection for these mothers, are

notably absent from vertical transmission literature as a cause of infection for infants

(Pinch, 1994). Similar to feminist theory, Stigma Theory defines the community of HIV-

positive women as deviant, or "other" (Taylor, 2001). Their societal status places them

at risk for early pregnancy, and their early pregnancy leads to their HIV diagnosis, this

process of diagnosis during pregnancy is explored in the literature of stigma theory.

Taylor (2001) notes that "sick" women are separated from "well" women during

pregnancy by the mere diagnosis of HIV infection, regardless of whether or not

symptoms are present. She comments that this earns them the label, "Seemingly healthy,

but doomed." She takes care to separate the patient's emic viewpoint of feeling well

until medications (with their unpleasant side effects) are started, from the medical

profession's etic view of having an uncontrolled infection until medications are started.

How the concepts of medication, illness, and wellness are conveyed to these young

minority females has a lot to do with the way females communicate in general. A need

for connection is inherent in females (Pinch, 1994). Females have a need to be linked to







25

a supportive personal network more so than males. Knowing, including knowing about

self and HIV-infection, is generally based on passively received knowledge or subjective

knowing for poor women (Pinch, 1994). "Facts" are defined to these women as what

political leaders and health care providers say they are (Pinch, 1994). This means that in

the absence of empowering these girls, advice rendered to these mothers should be with

the full understanding of their position and vulnerabilities.

Although Taylor (2001) discusses both the Health Belief Model and the "rational

choice model," she seems to suggest the rational choice model is the dominant view in

HIV treatment communities. This model, though not elaborated in her writing, implies

that the medical community assumes that patients will adhere to prescribed treatments

and prevention practices once educated about HIV infection, simply because of medical

advice. It does not take into account any of the concepts discussed as barriers to

adherence in other studies. According to Taylor (2001) one primary barrier is the need to

hide the diagnosis because of stigma. If the mothers seek treatment, the child may be

labeled as infected, whether or not it is actually is. In this sense, by continuing her own

treatment, the mother is putting her child at risk of stigma. According to Bunting (1996),

the danger the mother represents to society though infecting innocent and unborn babies

rationalizes the choice to ignore her particular health concerns. She becomes

stigmatized, with discrimination by the health care system as a consequence Bunting

(1996).

Qualitative researchers have done the most work in describing the stigma of HIV.

Poindexter and Linsk (1999) conducted one-time semi-structured interviews on 19 older

African American women who were caregivers to family members with HIV/AIDS.

Though the subjects are HIV negative and older, they are members of the African







26

American culture and can speak to the context of being (or perceived as being) HIV-

positive in this community. This is the only study to look at the attitudes of African

Americans toward each other in regards to HIV infection. The point of the study was

interesting; in that the investigators were interested in the burden these women bear as

elders and caregivers. Their overall findings suggest that stigma should be addressed as

part of social work intervention. Of particular interest in the findings was that many of

these women rely on church for comfort, but do not disclose to any members or their

pastors that they have an HIV-positive family member. This comfort seems to be

regarded as contingent on secrecy and vulnerable to the threat of withholding comfort if

the secret is known. Three participants talked of their own discrimination of HIV-

positive community members-until one appeared in their family. One participant

viewed this as punishment from God for past discrimination.

Poindexter and Linsk (1999) review Goffman's stigma theory, summarizing the

three related types of stigma: associative stigma (ascribed to those attached to

stigmatized persons-also called courtesy stigma), internalized stigma (acceptance of

society's appraisal of one's reduced worth), and stigma management (being aware of real

or potential reactions of others). Methodology consisted of semi-structured questions

such as, "Tell me a story about how you or your family members have had negative

responses or have experienced discrimination about AIDS." Data collection included

field notes, observations, and researcher comments. Content was then coded and

analyzed, with results revealing that the majority of African American women

acknowledge stigma, but avoid it personally by carefully guarding disclosure of the HIV

status of their family members.







27

Another study looked specifically at the emergent fit of Goffman's stigma theory

to a grounded theory approach to HIV positive mothers. Ingram and Hutchinson (1999a)

published their discovery of an emergent fit of extant theory to their findings. In their

grounded theory approach to the mothering experience for HIV-positive women, themes

related to stigma kept arising. Through a review of existing literature on this

phenomenon, Goffman's theory seemed appropriate. The details of the grounded theory

study are listed separately below, but the themes that kept arising related to stigma are

privacy, passing, and covering in clinic. Stigma is an unanticipated difference that

discounts other attributes of an individual and causes the rejection of others (Ingram &

Hutchinson, 1999a). In relation to parenting, mothers also reported that this stigma

spread to their children, whether or not they were infected. Quotes such as, "No one

would touch my children" were common. Findings included leaving "double lives"

behind a "facade of normal mothering."

The concept of passing pertains to something that is not apparent being

concealed. The experience of appearing normal, with all of its benefits, is mitigated by

the constant threat of exposure and losing the benefits. The women going to church for

the benefit of support without expressing why they needed the support for fear of losing

the support is a good example. The need to pass is evidenced by the unwillingness to

have AZT in their homes because of its name recognition (Ingram & Hutchinson, 1999a).

Covering is the act of managing social tension and distracting attention away from the

stigmatizing quality (Ingram & Hutchinson, 1999a). Examples from the study include

lying about reasons for medication (i.e., back pain, cancer) and referring to visits to the

doctor as "shopping trips." In another, a woman struggling with weight loss from her

medications told her co-workers she was on a strict weight loss program. Their response







28

was that, "I would sure love to lose weight the way you did." She states she thought,

"No, you most definitely would not." The women believe that lies are justified as a

means for survival (Ingram & Hutchinson, 1999a).

Psychological Distress and HIV

Because of the well-documented link between depression and a suppressed

immune system, Pinch (1994) theorizes systematically oppressed adolescent females to

be at a higher risk for rapid HIV disease progression. Looking at the effects of stress

becomes important to this issue. Smith et al. (2001) use the Stress Process Model to

explain psychosocial functioning in African American recent mothers. The domains of

this model include stressors, coping resources, close relationships, coping responses, and

psychological distress. They found that though adherence and treatment literature

assumes both comprehension of and focus on the diagnosis of HIV, many of their

subjects rated money as a greater concern than HIV infection. This appears important, as

it addresses the concept of immediacy in these women. HIV infection is an arbitrary

concept with no immediate manifestation (symptoms); whereas poverty has daily

implications for housing, food, clothing, and transportation. Smith et al. (2001)

concluded this special population needs interventions to improve their social and

economic conditions, with interventions to address their HIV infection tailored to their

strongest measured resources.

Murphy et al. (2001) agreed with this association between psychological distress

and health risk behaviors, and sought to apply this theoretical perspective to adolescents

with HIV. They utilized Jessor's Problem Behavior Theory as their theoretical

framework. This theory focuses on three major concepts: behavior, personality, and

environment (Murphy et al., 2001). Their study suggests that higher levels of depression







29

predict increased risk taking behaviors, despite what they describe as youth-appropriate

risk reduction counseling. Anxiety related to health status, i.e. HIV infection, actually

increased the maladaptive coping behaviors of substance abuse and unprotected sex.

This could likely be studied in direct relationship to medication adherence with the same

results. This finding suggests that medical interventions in the absence of addressing

depression and environment are likely to fail.

Mellins, Ehrhardt, Rapkin, and Havens (2000) appear to agree with the above

models in their study ofpsychosocial factors and adaptation in HIV-infected mothers.

Though no explicit framework is cited, the concepts of environment, stress, protective

factors, and depression are listed. They found that mothers with HIV have additional

stressors of racism, isolation, poverty, discrimination, and single motherhood. These

make them more disposed to diagnoses of depression and post-traumatic stress disorder.

They also found that relationship breakup, assault, abuse, and removal of children played

a major role in causing these disorders. Ironically, some providers offer the fear of

removal of exposed infants as the primary reason mothers adhere to treatment protocols.

This is proffered as an explanation by some of the staff in the clinic to be studied, though

no legal precedent for the removal of exposed babies is established in Florida.

Minorities and HIV

bell hooks (2003) describes an early movement within the African-American

culture toward increased self-esteem and self-love, regardless of external events. She

describes a revolution where success was defined as the ability to value and revere

oneself despite the presence of dehumanization. This movement, largely led by males,

also demanded that men regain their historical position of being "revered by their

women." By definition, this would make the subjugation of women necessary for the







30

advancement of men. Not only is it undoubtedly frustrating for men to be expected to

appear intact and invincible in the presence of ongoing subjugation, but in a people

struggling for power, domination of women is the only area in which this is easily

achievable. This leaves black women in the terrible place of being less regarded even

within the culture that should provide harbor from discrimination.

Wilkinson (1999) states that in places with more income inequality, more deaths

can be attributed to violence. Nondisclosure of status before unprotected sex, and the

refusal to wear condoms are both acts of violence. Though the attributable cause of death

may be pneumocystis pneumonia secondary to AIDS, the primary cause of death is an act

of violence. Wilkinson (1999) further elaborates with reference to the "shame rage

spiral," where in effect, hostility mounts in the presence of unacknowledged shame.

Many black men are infected during incarceration, and this is a reasonable explanation

for black men feeling entitled to not disclose their HIV status to their female partners

upon release from prison. This, combined with the expectation of machismo, explains

why impregnating a woman (sign of virility) is the bragging right of many men, without

regard for the current or future health of the woman or the fetus. Farmer (2003) makes a

similar argument by stating that people subordinated by their social superiors and

threatened with humiliation, attempt to regain their sense of control by asserting authority

and control over those below them. The power imbalance between men and women in

minority communities is discussed extensively in the literature.

Beatty, Wheeler, and Gaitner (2004) conducted a review of HIV prevention

literature to look for inclusion of African American subjects with the intent of making

recommendations for the development of more effective prevention strategies for this

group. They reported a lack of culturally based theory to guide the existing research and







31

at the same time suggested that black Americans are too large and diverse a group to be

placed within one cultural definition. They found that research aimed at behavioral

lifestyle change was over represented in the literature, with the Health Beliefs Model, the

Theory of Reasoned Action, and the Transtheoretical Model of Behavior Change being

most commonly used. It was reported that often, even if included as subjects in a study,

minorities were not later broken out and analyzed as a separate group. They suggested

that models sensitive to different groups of minorities be developed and that not all effort

should be extended to individual change, as individuals do not always have the control

necessary to initiate changes. Women rarely have control over the elements of survival

sex, and childhood sexual abuse and unwanted adult sexual activity are both reported as

risk factors for risky sexual behavior, and for substance abuse that leads to risky sexual

behavior (Cargill, Stone, & Robinson, 2004).

The idea that any of this behavior is fully volitional is a stretch. Lynch, Kaplan,

and Salonen (1997) question the popular theories now that attribute health behaviors to

the choices that individuals make, because they make health behavior both individual and

volitional; neither of which are usually the case for women at risk for HIV. The theories

predict that if groups at high risk for HIV infection are simply told what behaviors to

avoid and why, they will stop the behaviors. Their findings showed that while there is

correlation between current SES and health behaviors in adulthood, there also exists

correlation between health behaviors in adulthood and SES at specific temporal

milestones throughout childhood. In other words, behaviors are in some way linked to

parental SES-an influence on behavior that neither an individual nor a provider could

go back in the past to change (Lynch et al., 1997). This is a strong argument for

increased economic parity in society, though it would be unattractive to most public







32

health providers and planners specifically because it lacks agency, even on the part of

programs. Programs act in the present and occasionally for the future; never in the past.

The effects of SES parity now, even if attained, would not be measurable for a lifespan.

Since program evaluations must measure changes each funding cycle, this would never

happen.

In their recent study, Cargill, Stone, and Robinson (2004) found evidence that

minorities are less likely to receive antiretroviral medication therapy (ART) even if the

guidelines indicate it as the standard of care. Often this is attributed to the provider's

sense that the patient will be unable to adhere to care. Basu (2004) observed that if ART

were reserved only for those who seem most likely to adhere to regimens, then only those

least likely to be in need of ARTs would receive them.

This is not the only recent work to support the idea that the behaviorist theories

are shortsighted. Aynalem, Mendoza, Frederick, and Mascola (2004) found in their study

of pregnant women who refused HIV-testing reported that 4% of women had to seek

their husband's permission prior to consenting to the test. This clearly negates personal

agency. Many others feared stigma, discrimination, disclosure, violence, or had a lack of

trust in the provider to prevent any of these. The authors fell short by suggesting that

through additional education and time to trust the providers, that many of these women

would change their minds. This seems naive in that it addresses none of the structural or

domestic problems that were identified as driving the refusal in the first place.

Another commentary on the behavior theories is the finding of Whyte, Standing,

and Madigan (2004) that there was actually a positive correlation between HIV-related

knowledge and high-risk sexual behaviors in African American women in the southeast.

Income, age at first sexual encounter, number of partners, and education level were all







33

analyzed, with only education level having a significant impact on behavior. This shows

that the population-specific messages are reaching the community; they are just not

having an impact on behavior. Ebrahim, Anderson, Weidle, and Purcell (2004) analyzed

data from the 2001 Behavioral Risk Surveillance Survey. They found that though

African Americans had a higher rate of having ever been tested for HIV, they had a lower

level of knowledge about available treatment. Overall, even those who had been tested

were not aware that treatment for infection now exists. This shows the durability of the

targeted messages to avoid contracting this fatal disease, and about it being a death

sentence. The danger of targeting these types of messages at a community already

devoid of hope for the future is that the messages ablate any hope for a future once a

person is diagnosed with HIV.

Positive Adaptation

Despite the grim outlook, some women, devoid of many identified barriers, do

exceptionally well with their HIV diagnoses. Speigel and Schrimshaw (2001) conducted

a series of two semi-structured interviews that they then subjected to thematic analysis.

The subjects were HIV-positive women who claimed to have had positive changes as a

result of their diagnoses. Exclusion criteria included intravenous drug use, and the

average age was 36 years, with a sample size of 54. One-third was African-American.

The mix of subjects is not reflective of HIV-positive women in general, so the findings

are not generalizable. These women, unlike any other described in literature or seen in

personal experience, describe HIV infection as a catalyst to resolving their relationship

conflicts, a queue to enjoy life, and a step to becoming strong advocates for themselves.

While there may be some merit to the idea advanced that perceptions of positive

outcomes may imply positive outcomes-certainly supported in psychoneuro-







34

immunology literature, it is not consistent with the findings closer to the population of

interest.

Dunbar, Meuller, Medina, and Wolf (1998) conducted similar semistructured

interviews in the hopes of supporting their previously created model relating to positive

adaptation. They conducted interviews with 34 women in a study that also appears to be

nongeneralizable because of the geographic demographics of their convenience sample.

Unlike HIV-infected women, their participants were almost half Pacific-Islanders;

college educated, and had an average age of 36. They found five components that

supported their model of adaptation: reckoning with death, life affirmation, creation of

meaning, self-affirmation, and redefining relationships. All of these processes they

described require reflexivity and a level of acceptance of diagnosis that far exceed that

documented in the target population of this paper. However, the concept of

understanding personal growth, as a precedent to effective coping is consistent with

previously explored stress models.

Neither of these studies published clear descriptions of their data analysis. Both

of these studies suggest a population that has adjusted to their diagnosis, presents for

care, acts in their own best interest, and is therefore largely incongruent with the

population of interest. Identifying the barriers faced by minority HIV-positive

adolescent mothers from this differently adjusted population may be of some value.

Studies addressing barriers to adherence provide some explanation as to the contextual

differences experienced by these groups.

Mothering by HIV-Infected Women

Ingram and Hutchinson (1999b) described their grounded theory study in

anotherpublication in greater detail. The purpose was to describe the mothering







35

experience of HIV-positive women. The goal was to understand actions in the context of

their beliefs. A sample of 18 HIV-positive mothers aged 18-44, half of whom was

minority, was subjects of extensive interviews. Using a grounded theory technique, a

basic social psychological process (BSPP) of defensive mothering was discovered. The

obvious connection to stigma is described in one woman's comment, "Everything you do

in your life you have to worry about if someone is going to find out. It is hard. You have

to watch everything you say and do." They concluded that stigma provides the context of

HIV-mothering, and sets the stage for defensive mothering (Ingram & Hutchinson,

1999b). Mothering consists primarily of ways to protect the child from both herself and

society's prejudices. This relationship, that seems to explain a mother's preference for

her child's welfare over her own, is further explored in another less rigorous qualitative

study in the literature.

Andrews, Williams, and Neil (1993) conducted a series of two qualitative

interviews: the Norbeck Social Support screen and a substance abuse screen to a sample

of 80 HIV-infected mothers. A weakness in this study is that no description of data

evaluation was included. No significant themes were identified, and no methodology

other than "interviews" was given. Quotes were discussed generally, but not truly

analyzed. The two tested screens that could be reported on an ordinal basis were more

fully described. When evaluating social support, it was impressive to reveal how many

of these mothers relied on young children for emotional support (Andrews et al., 1993).

In regards to the medical community, most women expressed distrust. One woman, in

commenting on care offered to her HIV-positive son, said "They want to put him on

DDI. The AZT is making him too anemic. I'm going to try the DDI dose they want him

on first, before they give it to him. My thinking is that he and I are almost the same, you







36

know, with the same genes, so I'll know what side effects he'll get from the DDI"

(Ingram & Hutchinson, 1999a). This statement shows a total lack of understanding of the

relationship between body surface area and side effects, but more alarming is that her

views are built on the foundation of mistrust.

The research reviewed describes the women who are likely to fail treatment, or

describes the society at large that stigmatizes and alienates women from caring for

themselves. Ample descriptive data were identified that described women who fail

treatment, but this data does not differ from the descriptions of women most likely to be

infected with HIV. A small but important section of the literature tried to determine how

women experienced living with HIV, and allusions were made to perceived judgment and

hostility in their worlds at large, including in the HIV treatment community. The next

logical step is to gather quantitative data to describe whether or not this perceived

difference in treatment is associated with differences in outcomes, because if it is, then

modifying the treatment environment will be fundamental to improving maternal

outcomes.















CHAPTER 3
METHODS

Purpose

The purpose of this study was to determine if previously identified barriers to

HIV treatment adherence affect the perinatal dyad, the postpartum mother, or the

exposed infant equally; and to determine if the proposed constructs of structural

facilitators or structural barriers are associated with HIV treatment adherence in the

perinatal dyad, the postpartum mother, or the exposed infant.

Design

This was a descriptive and analytic epidemiologic case comparison study. The

goals were both to describe frequencies of exposures and outcomes in the three groups

and to analyze which independent variables best predicted outcomes. All data collected

were collected through retrospective chart reviews. Data were analyzed through both

descriptive statistics and logistic regression analysis. A convenience sample was obtained

through systematic review of the charts of HIV-positive pregnant women in the high risk

obstetrical clinic (see Figure 3-1). The study was constructed of 100 maternal cases and

200 comparisons. The 100 maternal cases consisted of postpartum HIV-positive women

with the group name "postpartum mothers." The 200 comparisons were comprised of

two groups. One hundred of the comparisons comprised the group "dyad" and consisted

of the same mothers as the case group during a different time interval. For this group,

data were collected during the prenatal period instead of the postnatal period. The last








38

100 comparison subjects were in the "infant" group. This group was comprised of the

babies born to the dyad group. As such, these infants shared maternal risk factors at the

same point in time as the maternal group. The subjects were all patients of the same HIV

subspecialty clinic and its affiliated high risk obstetrical clinic. Table 3-1 depicts the

distribution of the independent variables for the maternal sample.


Convenience Sampling of HROB Charts for
HIV(+) Pregnancies 2000-2004


Mother maintained primary custody of infant to
18 mos postpartum: Included in study


Did not meet inclusion criteria -
Excluded from study


Mother & fetus assigned to
DYAD Group


After delivery, infant
assigned to Infant Group


After delivery mother
assigned to Postpartum
Mother Group


I ________

Figure 3-1. Subject selection

Setting

This study was conducted at an urban outpatient HIV clinic and its affiliated high

risk obstetrical clinic in the southeastern United States. Data were collected through

chart extraction without contact with subjects. The clinics are located in an urban


i







39

medical complex and are staffed by nurses, physicians, students, and residents. Patients

in the high risk obstetrical clinic receive specialty prenatal care for a variety of conditions

including HIV seropositivity, diabetes, and epilepsy. The HIV clinic is located in the

same complex and is part of the same health system. It is a federally funded (Ryan

White) and state funded (Children's Medical Services/Medicaid) HIV sub-specialty care

clinic. Care is provided in tandem by nurse practitioners and infectious disease

physicians on an outpatient basis. HIV treatment, psychological support, dietary support,

social work support, and primary care are all offered on site, unless specifically excluded

by insurance or funding regulations.

Subjects

Sample Selection

Convenience sampling was used to select the sample from eligible dyads. Based

on a prediction that 30% of the subjects were in adherence to HIV care, approximately

229 subjects were required given a 0.05 level of significance, the 10 predictors, and 90%

power of the test.

Inclusion Criteria

* Subjects were HIV (+) females whose prenatal medical care was provided by the high
risk obstetrical clinic at the study site from 2000 to 2004.

* The mother and infant both survived through 18 months postpartum.

* The infant remained in the primary care of the biological mother.

Exclusion Criteria

* Neonatal, infant, or maternal subjects who spent greater than 2 weeks inpatient during
the 18-month observation period.

* Neonates and infants who spent greater than 2 weeks outside of the primary care of
their biological mothers during the observation period.

* The mother or the baby died before the end of the observation period.







40

Data Analysis

The Statistical Analysis System (SAS) (Version 9.1) was used for all statistical

analyses and for writing the scientific report of the quantitative data. Descriptive

statistics were used to obtain the summary measures for all data including a description

of the sample characteristics. Descriptive statistics included means, medians, modes,

ranges, interquartile ranges, and standard deviations for continuous variables. Categorical

variables were statistically represented in frequency distributions, percentage

distributions, and graphical illustrations. A p-value of less than 0.05 was considered

statistically significant.

To address the major hypothesis of the study, stepwise logistic regression with

forward elimination techniques were used to both control for confounders and estimate

independent relationships between the predictive variables and the outcome. Logistic

regression analysis was also used to explore potential differences in predictor variables

between who participated in recommended HIV care from those who did not. All

predictor variables were included that had either at least a marginal bivariate association

with the outcome variables or for which there was some rationale that the variable may

have been a confounder or effect modifier for other variables. This initial model also

included hypothesized interaction terms for which stratification analyses suggest

potential interactive effects.

The point and interval estimates of the odd ratios of the categorical predictor

variables were reported. The hypothesis of homogeneity was addressed by running

frequencies. In most cases, Chi-square statistics were presented. In cases where any cell

in the 2x2 table or 3x2 table was valued at less than 5, the Fisher's Exact Test for small

samples results was instead used for increased validity.







41

Procedure

Following graduate committee approval, paperwork was processed to gain

approval of the University of Florida Institutional Review Board (IRB-01). Included in

the packet was a letter of support from the administration of the study site. Expedited

processing was requested and approved, as subjects were fully de-identified for their

protection (in compliance with the Health Information Portability and Accountability Act

of 1996). This option was chosen because the identifying data omitted was not essential

to study outcomes. Subjects were deemed vulnerable because data on mental illness and

substance abuse were collected from pregnant, HIV-positive and potentially pediatric

subjects. After clearing the University of Florida IRB-01, approval of the study site IRB

was also requested and obtained under the same conditions.

Data Collection

Upon approval of the University of Florida IRB-01 and study site IRB,

de-identified data were collected through chart extraction onsite at the clinic. The data

collection form was constructed using Snap software. The principal investigator

personally conducted or oversaw data collection at all times. Each subject was assigned

a unique numeric identifier not traceable back to the subject at any time and not recorded

any place that contains protected health information (PHI). As the site is a federally

funded HIV clinic, superconfidential requirements were already in place onsite. Off-site,

de-identified data were transported via password protected laptop computer and

transferred to the secure server of the University of Florida College of Nursing. A back-

up copy of the de-identified data remains in the possession of the Principal Investigator

and is stored securely.







42

Potential Health Risks

There were no physical or psychological risks to subjects of this study. There

was no intervention and no contact at any point with subjects. All data collected were

de-identified from the outset, eliminating the risk associated with disclosure of HIV

status.

Potential Health Benefits

There were no direct health benefits from participation in this study as there was

no contact with subjects and no intervention. There were potential benefits to society as

knowledge was gained to explain reasons HIV-positive mothers are lost to follow up care

at a disproportionate rate. Considering the potential impact premature parental death has

on children, families, and society, the benefit of decreasing parental morbidity and

mortality outweighs the risk to subjects.

Potential Financial Risk

There was no cost to subjects, no information that would trace back to subjects,

and no contact with subjects. Therefore, there was no potential financial risk.

Potential Financial Benefits

There was no reimbursement or compensation of any kind to subjects. Therefore,

there was no potential financial benefit to subjects.

Conflicts of Interest

The principal investigator and subinvestigators did not have any conflict of

interest regarding this protocol. No benefit, beyond professional and academic

development was derived from participation in this project.















CHAPTER 4
RESULTS

Purpose

The purpose of this study was to determine if previously identified barriers to

HIV treatment adherence affect the perinatal dyad, the postpartum mother, or the

exposed infant equally; and to determine if the proposed constructs of structural

facilitators or structural barriers are associated with HIV treatment adherence in the

perinatal dyad, the postpartum mother, or the exposed infant.

Description of Sample

The convenience sample in this case comparison study was constructed of 100

maternal cases and 200 comparisons. The 100 maternal cases consisted of postpartum

HIV-positive women with the group name "postpartum mothers." The 200 comparisons

were comprised of two groups. One hundred of the comparisons comprised the group

"dyad" and consisted of the same mothers as the case group during a different time

interval. For this group, data were collected during the prenatal period instead of the

postnatal period. The last 100 comparison subjects were in the "infant" group. This

group was comprised of the babies born to the dyad group. As such, these infants shared

maternal risk factors at the same point in time as the maternal group. The subjects were

all patients of the same HIV subspecialty clinic and its affiliated high risk obstetrical

clinic. Tables 4-1 and 4-2 depict the distribution of the independent variables for the

maternal sample.







44

Table 4-1. Number and percent distribution of maternal sample by age and race
Age 14-18 19-21 22-24 >24 Total (%)
4 13 21 62 100

Carribean
Race AA Island Hispanic Caucasian Other Total

40 21 16 12 11 100


Table 4-2. Number and percent distribution of maternal sample by dichotomous
variables
Variable Yes No Total (%)


Mental illness

New diagnosis

Poverty

HS/GED

Drugs or alcohol

Pill burden


23

62

67

61

4

0


77

38

33

39

96

100


100

100

100

100

100

100


Variables

The independent variables are maternal depression or other mental illness,

proximity (newness) of HIV diagnosis, race, income, adolescence (age), high school

education, substance abuse, pill burden, structural barriers to access, and structural

facilitators to access. The dependent variables are consistent, but defined differently for

each group because the standard of care is different during each period. These outcome

measures include laboratory values, percentage of missed medication doses, and delayed

or missed appointments for care.

Hypothesis

The following two-tailed hypothesis was tested: Maternal depression or other

mental illness, proximity of HIV diagnosis, race, income, adolescence, high school







45

education, substance abuse, pill burden, structural barriers to access, and structural

facilitators to care cause a significant difference in adherence as measured by the

laboratory results, missed medication doses, and delayed or missed clinic appointments

of HIV-positive pregnant women, postpartum mothers, or their infants. In order to test

the hypothesis, the following four research questions were asked and answered.

Research Question One

Are the three groups (postpartum mothers, infants, and dyads) homogeneous with

respect to the three outcome measures (laboratory results, medication adherence, and

appointment adherence)?

An analysis of frequency was used to evaluate this question. Chi-Square statistics

and p-values were computed by cross-classifying the variables and groups with each

outcome variable.

The results presented in Table 4-3 indicate that the three groups were not

homogeneous with respect to laboratory results (Chi-Square=98.82, p=0.0001).

Laboratory results were significantly less likely to be in the acceptable range for

postpartum mothers (47%) than for infants (96%) or dyads (96%). The result indicated

that the three groups were not homogeneous with respect to ART medication adherence

(Chi-Square=74.12, p=0.0001). Medication adherence was significantly less likely to be

in the acceptable range for postpartum mothers (36%) than for infants (90%) or dyads

(92%). The three groups were also not homogeneous with respect to appointment

adherence (Chi-Square=57.16, p=0.0001). Appointment adherence was significantly less

likely to be in the acceptable range for postpartum mothers (52%) than for babies (80%)

or dyads (97%). Table 4-3 depicts the distribution of outcomes in each group.







46

Table 4-3. Distribution of cases and comparison groups by independent variables of
exposure to barriers and exposure to facilitators, and by outcome measures
of favorable laboratory results, medication adherence, and appointment
adherence
Cases Comparison groups
Postpartum
mothers Dyads Infants
Total Total Total
Yes No % Yes No % Yes No % x2 p
Exposure to
barriers 31 69 100 5 95 100 31 69 100 87.70 .0001
Exposure to
facilitators 6 94 100 100 0 100 97 3 100 177.36 .0001
Favorable
laboratory
results 47 53 100 96 4 100 96 4 100 98.80 .0001
Medication
adherence
>90% 46 54 100 90 10 100 92 8 100 74.10 .0001
Appointment
adherence 52 48 100 80 20 100 97 3 100 57.20 .0001

Research Question Two

Is there a significant relationship between the outcome variables (laboratory

results, appointment adherence, and medication adherence) and demographic variables

(age, race, mental illness, poverty, new diagnosis, pill burden, high school education,

substance abuse) among the three groups?

Analysis of frequency and Cochran-Mantel-Haenszel Statistic were used to

address this question. The results shown in Table 4-4 indicated that the relationship

between laboratory results and age, maternal race, mental illness, poverty, pill burden,

and new diagnosis were statistically nonsignificant for each group. The relationship

between laboratory results and education were statistically nonsignificant for infants and

for postpartum mothers (Chi-Square=4.79, p=0.03). The Cochran-Mantel-Haenszel

Statistic indicated an overall significant relationship between laboratory results and

education (Chi-Square=4.92, p=0.03). Mothers with a high school education were 2.24

times more likely to have favorable laboratory results (C.I: 1.1-4.6). There was also a







47

significant association between laboratory results and substance abuse for postpartum

mothers (Chi-Square=4.97, p=0.03) and for dyads (Chi-Square=4.78, p=0.03), though

there was no significant relationship to infant outcomes. The overall association was

statistically nonsignificant.

Table 4-4. Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure
of laboratory results
Cases Comparison groups
Postpartum
mothers Dyads Infants
Odds
x2 p x2 p x2 p ratio 95% CI
Age 0.79 0.85 0.85 0.84 2.55 0.47
Race 3.02 0.55 6.25 0.18 1.72 0.79
Mental illness 1.09 0.30 1.72 0.19 0.01 0.92
Poverty 0.40 0.53 2.05 0.15 0.54 0.46
New diagnosis 0.22 0.64 0.25 0.61 0.25 0.61
HS/GED 4.79 0.03 0.21 0.65 0.21 0.65 2.24 1.1-4.6
Substance abuse 4.70 0.05 4.79 0.15 0.13 0.72

The results shown in Table 4-5 indicated that the relationship between medication

adherence and age, maternal race, mental illness, poverty, pill burden, and new diagnosis

were statistically nonsignificant for each group. There was a significant relationship

between maternal education and medication adherence for infants (Chi-Square=4.48,

p=0.03) and postpartum mothers (Chi-Square=5.97, p=0.01), though there was no

significant association with medication adherence and education for dyads. The overall

association of medication adherence and education for groups was statistically significant

(Chi-Square=8.07, p=0.005). Medication adherence was 2.49 times more likely to be in

the acceptable range if mothers had a high school education (C.I.: 1.32-4.7). There was

a significant relationship between medication adherence and substance abuse for mothers

(Chi-Square=4.89, p=0.03), but not for dyads or infants. The general association test

between medication adherence and substance abuse for the groups was statistically









significant (Chi-Square=5.11, p=0.02). There was no significant relationship between

the independent variables and appointment adherence, as shown in Table 4-6.

Table 4-5. Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure
of medication adherence


Cases
Postpartum
mothers


Age
Race
Mental illness
Poverty
New diagnosis
HS/GED
Substance abuse


Comparison groups


Dyads Infants


x2 p
0.42 0.94
4.33 0.36
7.66 0.01
1.65 0.20
0.01 0.98
0.01 0.93
0.36 0.55


X2 p
0.96 0.81
1.11 0.89
1.06 0.30
0.05 0.83
1.53 0.22
4.49 0.04
0.34 0.56


Odds
ratio 95% CI





2.49 1.3-4.7


Table 4-6. Relationship between maternal age, race, mental illness, poverty, education
level, substance abuse, and objective pill burden and the outcome measure
of appointment adherence
Cases Comparison groups
Postpartum mothers Dyads Infants
x2 p x2 p x2 p
Age 4.91 0.18 8.70 0.07 2.68 0.44
Race 4.09 0.39 3.20 0.52 1.98 0.74
Mental illness 2.09 0.15 0.92 0.34 0.13 0.72
Poverty 1.46 0.23 0.00 0.99 0.55 0.46
New diagnosis 0.26 0.61 1.08 0.20 0.09 0.76
HS/GED 1.81 0.18 0.99 0.32 0.38 0.54
Substance abuse 3.85 0.12 0.13 0.72 0.77 0.38

Research Question Three

Is there a significant association between the outcome variables (medication

adherence, laboratory results, and appointment adherence) and structural facilitators and

structural barriers?

Analysis of frequency and Cochran-Mantel-Haenszel Statistic were used to

address this question. The results shown in Table 4-7 indicated that the relationship

between laboratory results and structural barriers (insurance companies, including


x2
0.52
1.74
1.33
0.60
0.39
5.97
4.89


P
0.91
0.78
0.25
0.44
0.53
0.01
0.04







49

Medicaid HMOs, that limit access to comprehensive care in the subspecialty HIV clinic

or requirement of an individual to link self with follow-up care, including the

requirement of self-initiated case management before clinic appointment is given) were

statistically nonsignificant for babies and dyads but were significant for postpartum

mothers (Chi-Square=5.53, p=0.02). The overall association according to Cochran-

Mantel-Haenszel Statistic was not statistically significant. In addition, the result

indicated that the relationship between laboratory results and structural facilitators

(transportation to clinic, Targeted Outreach to Pregnant Women Act involvement [to

provide transportation, emotional support, and/or expedited paperwork], HIV nurse and

social worker involvement, medication dispensed [instead of prescriptions given],

primary care available onsite, intake paperwork not necessary, and/or previously

established and ongoing relationship with case manager) was not statistically significant

for each group.

Table 4-7. Relationship between structural barriers and structural facilities and the
outcome measures of laboratory results, medication adherence, and
appointment adherence
Cases Comparison groups
Postpartum
mothers Dyads Infants
Odds
x2 p x2 p x2 p ratio 95% CI
Laboratory results
Structural barriers 5.53 0.02 1.87 0.17 0.64 0.22
Structural facilitators 3.38 0.07 0.13 0.72
Medication adherence
Structural barriers 4.23 0.04 0.15 0.70 0.58 0.44
Structural facilitators 3.58 0.06 0.27 0.60
Appointment adherence
Structural barriers 4.46 0.03 1.84 0.18 1.32 0.25 0.36 0.17-0.78
Structural facilitators 5.90 0.03 0.09 0.76 32 1-1034









Research Question Four

Are maternal depression or other mental illness, proximity of HIV diagnosis,

race, income, adolescence, high school education, poverty, substance abuse, pill burden,

structural barriers to access, and structural facilitators to care significantly different as

measured by the laboratory results, medication adherence, and appointment adherence of

HIV-positive pregnant women, postpartum mothers, or their infants?

Logistic regression was used to address this hypothesis. The forward selection

procedure was utilized to obtain the optimal model for each outcome variable. As shown

in Table 4-8, the final analysis of Maximum Likelihood Estimates shows the following

variables were significant to laboratory results: age (p = 0.04), maternal education

(p = 0.008), membership in the baby group (p = 0.0001), or the dyad group (p = 0.0001).

The association between positive laboratory results and young maternal age was

inversely related, with an odds ratio of 0.6 (95% C. I. 0.40-0.98). Maternal education

provided an odds ratio of 2.8 (95% C. I. 1.31-6.06). Comparing membership in the baby

group or the dyad group to that of the postpartum mother group had an odds ratio of 31.7

(95% C. I.: 10.49-95.94). This means that young maternal age, maternal lack of a high

school education or its equivalent, and simply being in the postpartum maternal group

were most closely associated with poor laboratory findings. Babies were 31 times more

likely to have therapeutic laboratory results than for those postpartum mothers or dyads.

Table 4-8. Summary of logistic regression analysis predicting favorable laboratory
results
95% Wald
Wald Odds confidence
Variable B SE statistic p ratio limits
Age -0.46 0.23 4.21 0.04 0.63 0.4-0.9
Education 1.04 0.39 7.03 0.008 2.80 1.3-6.1
Group: Baby 3.46 0.56 37.48 0.0001 31.72 10.49-95.94
Group: Dyad 3.46 0.56 37.48 0.0001 31.72 10.49-95.94







51

The final analysis of Maximum Likelihood Estimates shows the following

variables (Table 4-9) were significant to medication adherence: maternal education

(p = 0.005), being in the baby group (p = 0.0001), and being in the dyad group

(p = 0.0001). The odds ratio estimates showed that mothers with a high school education

or its equivalent had a 2.5 times greater chance (95% C. I.: 1.32-4.72) of adhering to

their antiretroviral medications. Members of the baby group were almost 12 times more

likely to receive their medications (95% C. I.: 5.29-25.39), and members of the dyad

group were almost 15 times more likely to adhere to their medications (95% C. I. 6.4-

34.58).

Table 4-9. Summary of logistic regression analysis predicting favorable medication
adherence
95% Wald
Wald Odds confidence
Variable B SE statistic p ratio limits
Education 0.91 0.33 7.85 0.0050 2.49 1.3-4.7
Group: Baby 2.45 0.4 37.5 0.0001 11.59 5.29-25.39
Group: Dyad 2.7 0.43 39.34 0.0001 14.87 6.4-34.6

Table 4-10 shows that the final analysis of Maximum Likelihood Estimates

indicates the following variables were significant to adherence to the recommended clinic

appointments: maternal age (p = 0.04), structural barriers (p = 0.0037) and membership

in the dyad group (p = 0.0001). The odds ratio estimates (0.66) showed that younger age

had a negative relationship to appointment adherence (95% C. I. 0.45-0.97), as did the

presence of structural barriers odds ratio 0.32 (95% C. I. 0.15-0.69). The dyad group was

23 times more likely (95% C. I. 6.71-79.02) to have favorable appointment adherence.

Support for Hypothesis

Based on the results presented above, the hypothesis: Maternal depression or

other mental illness, proximity of HIV diagnosis, race, income, adolescence, high school







52

education, substance abuse, pill burden, structural barriers to access, and structural

facilitators to care are associated with a significant difference in adherence as measured

by the laboratory results, missed medication doses, and delayed or missed clinic

appointments of HIV-positive pregnant women, postpartum mothers, or their infants is

supported and statistically significant.

Table 4-10. Summary of logistic regression analysis predicting favorable appointment
adherence
95% Wald
Wald Odds confidence
Variable B SE statistic p ratio limits
Age -0.41 0.20 4.37 0.04 0.66 0.4-0.9
Education -1.15 0.39 8.45 0.004 0.32 0.15-0.67
Group: Baby 0.62 0.41 2.28 0.13 1.86 0.83-4.15
Group: Dyad 3.14 0.63 24.85 0.0001 23.03 6.7-79















CHAPTER 5
DISCUSSION AND IMPLICATIONS

Purpose

The purpose of this study was to determine if previously identified barriers to

HIV treatment adherence affect the perinatal dyad, the postpartum mother, or the

exposed infant equally; and to determine if the proposed constructs of structural

facilitators or structural barriers are associated with HIV treatment adherence in the

perinatal dyad, the postpartum mother, or the exposed infant. The dependent variables

were operationalized as follows: adherence to antiretroviral medication = self report of

missing <10% of doses; adherence to appointment schedule = all primary and

subspecialty care received within the recommended time frame; and therapeutic

laboratory results = a stable or falling HIV viral load, and a stable or rising CD4 count.

The results showed that all outcome measures were significantly different

between the infant and the postpartum mother groups. Further, exposure to the tested

constructs of structural facilitators and structural barriers accounted for the majority of

the differences. The use of the women and babies as their own comparisons (at different

points in the care system) assured that the exposures to maternal factors were the same

for all three groups. This allows more confidence that the differences expressed were

likely associated with the exposures to structural barriers and structural facilitators.




























































Improved Medication
&Appointment
Adherence Leading
toTherapeutic Lab
Results


Poor Adherence to
Appointments Medication
Leading to Increased Viral
Loads & Decreased CD4
Counts


Figure 5-1. Stigma theory in this study, as adapted from Goffman (1963)







55

Descriptive Statistics

Structural barriers were present for 69% of postpartum mothers, 31% of dyads,

and 5% of babies. This shows heterogeneity between groups (p<.0001), with postpartum

mothers significantly more likely to experience structural barriers to care than babies or

dyads.

Structural facilitators also appeared heterogeneous between groups (p<.0001),

being present for 100% of babies, 97% of dyads, and 6% of postpartum mothers. CD4

counts and HIV viral loads were significantly less likely (p<.0001) to be in the acceptable

range for postpartum mothers (47%) than for babies (96%) or dyads (96%). Greater than

10% of medication doses were missed in 54% of postpartum mothers, while only 10% of

babies and 8% of dyads reported missing >10% of doses. This shows a difference at the

(p<.001) level of significance.

The last outcome that differed between groups was missed or delayed

appointments. Again, postpartum mothers were less likely (p<.0001) to attend clinic

appointments (48% missed) than were their babies (20% missed) or the same mothers

during their pregnancy (3% missed).

Laboratory Results

Therapeutic laboratory results consisted of stable or decreasing viral loads and

stable or increasing CD4 counts. For postpartum mothers, therapeutic laboratory results

were positively impacted by maternal achievement of a high school diploma or

equivalent (p=0.0286). In fact, even after adjusting for group membership, subjects in all

groups were 2.24 times more likely to have positive laboratory findings if the mothers

had a high school education or its equivalent.







56

Therapeutic laboratory results for postpartum mothers were negatively affected

by prenatal maternal substance abuse at diagnosis according to the Fisher's Exact Test

(p=0.046). This effect is limited to the postpartum mothers, there was no general

association (p=0.21). There were few subjects who were positive for illicit substances at

the onset of their pregnancy (n=3), and those who were entered residential treatment

programs to maintain custody of their babies. These residential programs required

routine drug-testing, so ongoing use was not in question. Since there was not ongoing

substance abuse, the clinical significance of this finding is not readily apparent.

Maternal CD4 counts and viral loads were also negatively affected by the

presence of structural barriers to care (p=0.019). Again, no significant general

association was found (p=0.11) to suggest these barriers affected the laboratory results of

infants or dyads. This means that young maternal age, maternal lack of a high school

education or its equivalent, and simply being in the postpartum maternal group were

most closely associated with increased viral loads and decreased CD4 counts.

Medication Adherence

Medication adherence during pregnancy (group=dyad) was negatively affected by

maternal mental illness according to the Fisher's Exact Test (p=0.015), though maternal

mental illness was not shown to have a general association once adjustment was made for

group membership (p=0.84). Again, this was a relatively small group of women (n=21),

and the clinical significance is not readily apparent.

Infant medication adherence was correlated with maternal education using the

Fisher's Exact Test (p=0.045). Postpartum maternal medication adherence was also

positively correlated with maternal education (p=0.015). After adjusting for group







57

membership, a general association (p=0.005) existed. Subjects in all groups were 2.49

times more likely to report medication adherence if mothers had received a high school

education or its equivalent.

Postpartum maternal medication adherence was negatively associated with prior

maternal substance abuse according to analysis using the Fisher's Exact Test (p=0.04).

In fact, there was a general association (p=0.024) after adjustment for group membership.

This means subjects in all groups were twice as likely to miss medications if the mother

had a history of prior substance abuse.

Finally, there was a correlation between medication adherence and the presence

of structural barriers noted (p=0.04) only for the postpartum mothers group. This means

that only subjects in the postpartum mother group were likely to experience structural

barriers that affected their medication adherence.

The logistic regression model showed that the best predictors of medication

adherence were maternal education (p=.005), membership in the infant group (p<.0001),

and membership in the dyad group (p<.0001). The odds ratio estimates show that

postpartum mothers with a high school education or its equivalent have a 2.5 times

greater chance of successful adherence to their antiretroviral medications. Simply being

in the infant group meant subjects were almost 12 times more likely to receive their

medications. Being in the dyad group meant subjects were almost 15 times more likely to

adhere to their medications.

This shows that something about the infant and prenatal dyad groups significantly

enhanced medication adherence. For postpartum mothers, the best predictor of

therapeutic medication adherence was education.









Appointment Adherence

Postpartum maternal appointment adherence was negatively correlated with the

presence of structural barriers (p=0.03). Statistical testing also showed a general

association (p=0.008). In this case, the influence of structural barriers was so strong, that

after adjusting for group membership subjects were about 3 times less likely to attend

clinic appointments if there were structural barriers to doing so.

In contrast, postpartum maternal appointment adherence was positively affected

by the presence of structural facilitators to care (p=0.03). There was also a general

association after adjustment for group membership (p=0.02). Regardless of the group,

subjects were 32 times more likely to adhere to the recommended clinic appointment

schedule if structural facilitators were present.

The logistic regression model showed that older maternal age (p=.0365), presence

of structural barriers (p=.0037) and membership in the dyad group (p<0.0001) best

explained appointment adherence. The odds ratio estimates showed that young age had a

negative relationship to appointment adherence (OR=0.66), as did the presence of

structural barriers (OR=0.32). Membership in the dyad group increased the likelihood of

adherence to appointments. In fact, women were 23 times more likely to attend clinic

visits during the prenatal period than they were to continue their care after delivery.

Summary of Findings

Structural barriers were present for 69% of postpartum mothers, 31% of dyads,

and 5% of infants (p<.0001). Structural facilitators were present for 100% of infants,

97% of dyads, and only 6% of mothers (p<.0001). This means that extrinsic structural

barriers, including insurance limitations and complicated clinic admission procedures,

were present for a majority of mothers and almost no infants. Conversely, extrinsic







59

structural facilitators, including comprehensive funding and streamlined clinic admission

procedures were present for all of the infants and almost none of the mothers. This

shows that policies favor infant outcomes by almost never presenting barriers to infant

care, and almost always giving mothers the tools needed to succeed with the care of their

infants. On the other hand, mothers were rarely the recipients of structural help for

themselves, and experienced structural barriers over two-thirds of the time.

As for outcome measures, 96% of infants and dyads had acceptable laboratory

results. This is in contrast to only 47% of postpartum mothers. This showed a

significant difference (p<.0001). Medication adherence had a similar distribution, with

only 10% of infants and 8% of dyads reporting poor adherence. Over half (54%) of

postpartum mothers reported poor adherence, a significant difference again (p<.0001).

Given these figures, the last outcome, adherence to recommended clinic appointments, is

not surprising. Dyads were rarely nonadherent (3%), infants (20%), and postpartum

mothers almost half of the time (48%). Postpartum mothers who had already

demonstrated the ability to adhere to treatment during their pregnancies experienced the

majority of nontherapeutic laboratory results, missed medications, and missed

appointments.

Missing and Excluded Data

Complete data sets were obtained from 300 subjects. No data points were

missing at the time of analysis due to availability of redundant charting via paper charts,

computer programs, and across disciplines within the clinic.

The independent variable "pill burden" was operationalized to measure number

and times of pills per day against the treatment norm or standard at the time of therapy.

Because most subjects began medication therapy during their pregnancy (treatment







60

naive), progression to burdensome regimens due to drug resistance was not found in any

instance. Had this not been a chart review, self-reported pill burden as perceived by

subjects would likely have been a valid dichotomous measure to add to the model. For

purposes of this analysis, pill burden is shown as frequency only. It was not

operationalized in this study in a manner that showed heterogeneity.

Maternal substance abuse is underrepresented in this sample because of the study

purpose and its exclusion criteria. The purpose was to examine outcomes of mothers and

babies when challenged with the same maternal environment. This prohibited analysis of

babies that were not in maternal custody because the environment then differed from that

of the mother. Mothers that utilized illicit drugs during pregnancy were likely to lose

custody of their infants at least temporarily, and thus exclusion criteria eliminated these

infants, their mothers, and the dyad from the sample.

Ethical Considerations

As previously stated, postpartum mothers who had already demonstrated the

ability to adhere during their pregnancies experienced the majority of missed

medications, missed appointments, and a resulting increase in their viral loads and

decrease in their CD4 counts (immune function). With the exception of education, these

results were not found to be associated with ways of describing these mothers, (i.e.,

black, poor, drug-addicted, or depressed). What did appear to influence outcomes was

the structure of the health care system in which they were expected to function. The

complexity of the system in fact, may be why an increase in maternal education was

associated with successful treatment. For example, the mother of a neonate is given an

actual physical supply of Retrovir for her baby and required to return-demonstrate proper

administration by a HIV nurse specialist. The same nurse gives the mother an







61

appointment time and date for the infant to follow-up with the subspecialty nurse

practitioner in the HIV clinic after discharge home.

The postpartum mother, on the other hand, has no mechanism in place to help her

receive or continue her own medications, and is in fact, not allowed to make an

appointment for subspecialty care until after her 6 week visit in the obstetrical clinic

postpartum; a clinic that no longer addresses her HIV-related needs once she has

delivered her baby. An additional requirement is that mothers "connect" with case

management prior to entering care in the subspecialty clinic. Further, she likely loses

access to medications in the interim, because the Medicaid that covered the pregnancy

usually expires before (or if) she links with case management to begin the process to

access the AIDS Drug Assistance Program (ADAP). It does appear that decisions made

about who will receive what type of help through the maze of care is important to the

overall health of HIV-positive women.

The people who decide what constitutes a scientific problem shape the world,

because the world is shaped by research (Kane & Thomas, 2000). The development of

knowledge is a political enterprise that reflects society's current dominant values,

including the determination of which groups and subjects should be researched and by

whom (Browne, 2001). Since research is the foundation for best practice, it is important

to include maternal outcome measures in treatment protocol. The results of this study

suggest that a need for a more inclusive approach to maternal health in the perinatal and

postnatal periods is justified by basic ethical concepts, including personhood and value of

mothers and fetuses.

According to Beauchamp and Childress (1994), the first principle in bioethics is

the respect for autonomy. This means persons who are able to make decisions about their







62

own care must be allowed to, and researchers and providers must respect these decisions.

This does not happen when HIV-positive mothers are judged incapable of understanding

risk and treatment options, or when treatment counseling consists of discussing fetal

health in the absence of discussions of the impact of that treatment on maternal health.

Anti-retroviral prophylaxis given to the fetus via the mother is not free from impact on

the mother's health. Viral mutation and resultant drug resistance are documented

consequences of even short-term prophylaxis during the perinatal period (Toni et al.,

2005).

Moreover, it is important to include maternal outcomes as a measure in any study

that uses the bodies of mothers to impact the health of others. Successful adherence

overall is predicated on patients understanding and accepting their diagnosis, and

expression of a willingness to take medications (Rogers et al., 2001). Guidelines are

evolving rapidly to allow delay in the introduction of antiretroviral medications until

patients are ready. The aim is to minimize self and community harm by minimizing

multiple drug resistant viral mutations and the waste of resources. Pregnant women are

the exception to this rule, as the protection of the fetus is an immediate need, and waiting

months or years for maternal readiness is not considered a viable option. Guidelines state

that antiretroviral therapy must be started by week 14 of pregnancy, or as soon thereafter

as the woman is identified as HIV-positive (Steinhart, Orrick, & Simpson, 2002). But

ethicists caution not to overrule the decisions of mothers regarding their babies' health,

as the general view that being HIV-positive is the worst possible outcome for a baby may

not be accurate when viewed in the full context of a mother's world (Fu-Chang Tsai,

2001).







63

Reinforcement of the position that being HIV-positive is the worst possible thing

that can happen devalues the lives of the mothers, whether or not that is the intention.

Without a focus on the health outcomes of a baby and its mother, the message is

reinforced at each encounter that the baby is the true patient.

One approach is to consider Clarke's (1999) assertion that the mother and the

fetus should be considered jointly as "the person" when making decisions about "the

person's" future. This appears to be the most sound answer to the question of which is of

more value; to simply state that the question is void because they are the same individual.

This is beneficial to the fetus, because the only relationship the fetus can form is one of

biological dependence on the mother. The mother's psychological and emotional

dependence on the fetus is also honored. This position values the dyadic and

interdependent nature of the mother and fetus, and as such, served as the ethical

perspective of this research study.

The ethical position remains that to compromise either person would harm both.

Since the mother and the fetus are both entitled to moral standing, there are duties owed

to the dyad by the medical and research community. The absence of treatment for the

millions of women infected with HIV while the prevention of mother to child

transmission (MTCT) remains the focus of researchers and clinicians worldwide, uses the

justification of the utilitarian argument that protecting exposed babies does the most good

for the most people, with limited allocated resources. After all, the mothers are already

infected. Alleviation of suffering in mothers is ignored in the name of cost-effectiveness.

This is a contrived poverty because the resources exist to do both in the United States.

It is essential to remember that whether in the United States or elsewhere in the

world, imbued in the value of the mother is the fact that she is a person with HIV. Harris







64

(1999) addressed the cost of acknowledging someone's personhood-it brings her into

the same moral category as those who are uninfected. Judging someone as a nonperson is

a way of distancing her reality from everyone else's. To acknowledge her sameness

would be to admit that everyone is vulnerable to HIV-infection. This can be incredibly

unsettling for the uninfected world. It is the duty of policy makers and providers to keep

actual medical and research practice within the confines of moral behavior. This means

that reinforcing the stigma placed on HIV-positive mothers through practices that

excessively weight fetal outcome is both immoral and unethical.

Significance

Providers and policy makers focus on the cause most proximal to the symptom

they are treating (Krieger, 1994). In the case of perinatal HIV transmission, the cause is

exposure to the mother. But the more distal exposures that rendered the mother

susceptible to infection go unchecked. Failure to address the larger disparities makes the

prevention effort shortsighted. The same infants who benefited by the PACTG 076

protocol in the mid-1980s have now grown into impoverished, at-risk adolescents who

engage in activities (other than exposure to their mothers) that could result in HIV

infection. Children of minority groups whose contemporaries were spared by this

protocol are now coming of age to bear their own exposed infants.

The value-laden term "mother-to-child transmission" (MTCT) is the acronym

used by the scientific community worldwide to describe perinatal transmission. It

assumes that the fetus is already born (child, not fetus, though the majority of

transmissions do not occur after birth), and it assigns blame to the mother for giving the

baby HIV. Though mothers often report primary risk for HIV infection to be unprotected







65

sex with the fathers of these babies, any assignment of responsibility to the father is

absent.

Mother-to-child-transmission assumes the mother is exclusively to blame if her

child is born with HIV. It makes her the executor of an immoral act upon a defenseless

child, and justifies hostility toward her. Vertical transmission also denotes a direct route

from mother to child. No value-neutral term exists (i.e., parent to child transmission).

Since the trajectory of infection is more accurately father to child via mother, vertical

transmission could more accurately be described as angular. This is supported by

McNair and Prather's (2004) recent findings indicating that a partner's risk behaviors

exert more influence on a woman's HIV status than her own behavior. It follows that if a

mother's health is predicted by the behavior of the father, so then is the baby's.

Interestingly, both the medical community and society place the sole

responsibility and blame for the "high-risk" pregnancy on mothers. An example is the

message conveyed in the High Risk Obstetrical Clinic (HROB) where the predominant

message to HIV-positive mothers is one of saving the babies from the high-risk situations

in which their mothers have placed them. In one qualitative study (Ingram &

Hutchinson, 1999a), a black woman spoke of the hostility she felt every time she walked

into the clinic to see a beautiful black baby on a state-sponsored prevention sign saying,

"She has her daddy's eyes and her mother's AIDS" (Figure 5-2). She states, "When you

are positive, you learn undertones." These undertones, even in the healthcare

community, may be a fundamental barrier to continued treatment for mothers after they

have safely delivered their babies.




















































Figure 5-2. Florida Department of Health announcement, circa 1995

A mother whose social worth is reinforced by her ability to procreate must

confront the knowledge that she will never have a nonpathological pregnancy. This

distortion potentially impacts decisions about engagement in self-care, and encourages







67

further development of the already low expectations of the medical community. In

effect, their fates in the biomedical equation are already cast.

Thus is created the new risk category for under-treated HIV- that of pregnant

women. But it also creates the risk category of fetus of a minority mother, when again;

the real risk category is sexual partner or offspring of a male who has sex with males.

Providers speak candidly of the need to curtail reproductive activity in these mothers

because of their presumptive treatment failure and the burden their orphaned or infected

babies will place on society. The simple fact is, despite the increased number of

pregnancies in which babies are exposed to HIV, less than 10 were reported as infected in

Florida in 2004 (CDC, 2005).

Implications for Future Research

Existing literature does not reflect the real risk that without changing social and

economic circumstances, these same HIV-negative babies and their friends would grow

into adolescents and acquire their own HIV infections because the same social and

economic challenges that plagued their mothers now place them at risk. Perinatal HIV

literature has an overwhelming focus on the prevention of HIV-infection in the

fetus/infant.

It can be argued that the real disease to be battled is hopelessness and a lack of

agency-not just HIV infection. If this is so, then the public health establishment is not

addressing the root problems, and will likely not make adequate gains in reducing HIV

infections in minority populations. HIV-positive mothers seem in full recognition of the

circumstances of their lives. Providers, however well intentioned, are unable to address

what really needs to be changed to lower risk; namely the introduction of the hope for a







68

future and agency in individual lives. These barriers must be lifted through policy shifts

at least at a community level. This is well out of the reach of individual providers.

In the literature, disparity within communities is frequently noted to affect health.

It is not simply the absence of a wealth, but the also the presence of jealousy or envy of

others' wealth that affects health. Studies indicate that poor societies without income

stratification have better health than wealthy societies with stratification (Farmer, 2003).

In the most generous individual, existing beneath another's level must create self-

loathing. Farmer (2003) refers to the equalization of this disparity as creating "freedom

from want." It can be described as more than a freedom from want, but also the freedom

from feeling less than others. In poor neighborhoods deprived of hope, and in the High

Risk Obstetrical Clinic (HROB), deprived of joy that other mothers-to-be experience, this

feeling of inadequacy is reinforced. The economy of want and despair thrive in the

minority populations in the United States. Because social segregation is still the norm,

minority people engage in most behaviors, including high-risk behaviors, with someone

who is of the same minority, because that is who is closest and readily available.

Prevention efforts in Florida consist primarily of messages aimed at minority

communities, a practice known as social marketing (Figure 5-3). People are encouraged

to know their individual HIV status because knowledge is power. But without the power

to decide when and with whom to engage in high-risk behaviors, and the social and

structural support to seek treatment if infected, the expensive prevention message is

pointless, because knowledge alone is nothing. Without the power to demand the honest

disclosure of, and protection from infected partners, no prevention is likely to occur.

What is needed is the power of minority females to demand to know the status of their

partners, and once infected, to expect adequate medical treatment to extend their lives to






69
see their children grow. This begins with economic power and extends to social power,
and is not adequately addressed by the prevalent message of "know your status." A
requirement for survival with HIV is the presence of enough social and economic capital
to withstand the structural violence in place, including that in place in the health care
system.


Did you know
that 8 out of
10 babies
born with HIV
are black?

If you are pregnant, take
care of yourself.Get
prenatal care and ask
your doctor for an HI-IIV
test.
If you have HIV or AIDS,
medical treatment can
help you hove a healthy

Call 1.800.FLA.AIDS
for more information.

C4teutdt Wasfara






Figure 5-3. Florida Department of Health announcement, circa 2005
Research must begin with "why" infected individuals were disempowered even
before they became infected, and what keeps them so now. These factors are what







70

Krieger (1994) refers to as the social and political determinants of health. Recent

attention has been focused on the effects that groups, or memberships in groups, convey

on the agency of individuals. When there is the lack of distinction between individuals

and populations, and a tendency to view populations as the sum of the individuals that

comprise them; the practice is termed biomedical individualism (Krieger 1994).

Any approach to understanding HIV-positive mothers and the decisions they

make as individuals must be rooted in an understanding of the contextual elements of the

communities that both support and constrain them. Health and behavior decisions made

by these women regarding both themselves and their children must also be evaluated

with this knowledge prior to labeling them unreachable or noncompliant. Effective

interventions will not be derived and implemented without an understanding of the

environment and constructs under which individuals operate. Often survival-not

health-is the goal of these women; health decisions will always be subjugated by

survival decisions. Understanding the context of decisions will help change

environments that constrain individual agency.

Suggestions for Further Research

This study sought to determine if mothers and infants have different health

outcomes during the perinatal and postnatal periods. It measured whether they

experienced structural barriers and facilitators to care, and tried to determine if these or

other exposures explained some of the differences in the outcomes they experienced.

Certainly, this work has demonstrated that the health outcomes measured differ

significantly between postpartum mothers and both groups containing their children

(dyad and infant). It also indicates that postpartum mothers are much more likely to

encounter barriers, and much less likely to encounter facilitators. The inverse is true for







71

dyads and infants-they are much more likely to experience facilitators and highly

unlikely to encounter any barriers.

This research suggests that these findings are a natural result of assigning blame

and stigma. An ethical orientation argues for inclusion of maternal health as an outcome

measure in future studies. Left unanswered is the question of whether or not equal

facilitation of maternal health care would in fact lead to improved maternal outcomes.

However, this research suggests that HIV-positive mothers would go to any length to

protect the health of their babies. Studies that introduce interventions to increase

maternal role efficacy may show mothers the importance of surviving to see their

children thrive.

Expansion of the current perinatal prevention protocol to include maternal health

outcomes as measures of success may prompt more interventions. The creation of

federal requirements to link infant and maternal outcomes at federally funded clinics

would also improve maternal health at the clinic level. Educational efforts and surveys

directed at provider attitudes may provide additional insight into discrimination at the

clinic level. And importantly, qualitative studies are needed to describe how women

experience the HIV treatment system, and to interpret how the messages given are heard.

Further qualitative study of the role women feel they play in the lives of their babies

would add tremendously to a mixed methods design that included an intervention to

increase their role efficacy and to decrease their levels of depression.




















APPENDIX A
DATA COLLECTION INSTRUMENT




DATA COLLECTION TOOL


ID#


Collector Initials


Q1 Maternal Age
14-18 years................................ ............ .....
19-21 years................................. .......... ......
22-24 years................................ ............ .....
>24 years................................... .................


Q2 Is there a history of maternal depression,
cognitive impairment, or other mental illness?
Yes................................ .... ...................
No .............................. .................. ................


Q3 Maternal Race
African American .............................................
Caribbean Islander .............................................
Hispanic......................................................
Caucasian................................. ..................
Other....................................... ....................


Q4 Was the initial diagnosis of HIV made during
this pregnancy?
Yes...........................................................
No.........................................................


05 Was the household income below the poverty
level at the time of delivery (see table)?
Yes........................................... ..................
No................................... ............................


Q6 Did mother have a high school diplomalGED at
the time of delivery?
Yes............................................................... .........
No ......................................................... ..........


Date Collected


Q7 Is there documented Illicit drug or ETOH use
during pregnancy or delivery?
Yes..................................... .. ...................
No .......................... ............................. ......


Q8 Did the daily pill burden exceed the normative
value for the time It was prescribed (see table)?
Yes................................ ... ...........................
N o ............................................................ ..........


Q9 Were potential structural barriers to care
present (see table)?
Yes.................................... ... ..................
N o .........................................................................


Q10 Were potential structural facilitators to care
present (see table)?
Yes................................... ...........................
N o ......................................................................


Q11 Were lab values within expected guidelines
(see table)?
Yes...................................................................
N o ........................................................... ............


Q12 Did subject miss less than 10% of prescribed
doses of ART?
es............................................. ..................
N o .........................................................................


Q13 Were primary and HIV care obtained within one
month of recommended schedule (see table)?
Yes.............................................................
No ............................................................. ...















APPENDIX B
EXPECTED VALUES FOR DATA COLLECTION

Maternal Age at Time of Delivery
* 14 years to 18 years of age
* 19 years to 21 years of age
* 22 years to 24 years of age
* Over 24 years of age

Maternal Race
* African American
* Caribbean Islander
* Hispanic
* Caucasian
* Other

Pill Burden
* Infant-dose is four times daily for six weeks, then two times daily three days a week
until six months of age or until instructed to stop
* Adult (1998-1999)-dose is 3 tabs three times a day and 1 tab two times a day
* Adult (2000-2004)-dose is five tabs twice a day

Structural Barriers
* Requirement to obtain own case manager prior to transfer of care from HROB to
HUG Me clinic
* Requirement of additional intake process prior to transfer of care from HROB to
HUG Me clinic
* Insurance coverage that prohibits mother or infant from receiving primary and
specialty care at the same site

Structural Facilitators
* Automatic transfer of paperwork without intake appointment
* TOPWA or social work assistance to arrange transportation to appointments and to
assist with financial paperwork
* ART medication dispensed instead of prescription given
* Nurse specialist and social worker assigned to attend clinic, follow progress and
provide teaching

Lab Values
* Maternal HIV RNA PCR value (viral load) is stable or falling
* Maternal CD4 count (immune system) is stable or increasing
* Infant HIV DNA PCR remains negative
* Infant remains negative for PCP









Recommended Primary and Subspecialty Care Schedule
* Adult postpartum HIV clinic visits scheduled and maintained at least every 90 days
* Prenatal visits kept according to periodicity recommended in chart on prior visits
* Infant HIV testing and treatment visits performed within one month of recommended
time
o DNA PCR #1 performed by 4 weeks of age
o DNA PCR #2 performed by 8 weeks of age
o Bactrim obtained by 10 weeks of age
o DNA PCR #3 performed by 5 months of age
o EIA performed by 19 months of age

* Infant primary care performed within one month of recommended time, whether at
HUG Me or other pediatric site
o 2 month EPSDT performed by 3 months of age
o 4 month EPSDT performed by 5 months of age
o 6 month EPSDT performed by 7 months of age
o 12 month EPSDT performed by 13 months of age
o 18 month EPSDT performed by 19 months of age

Federal Poverty Guidelines
Year (one person, two person, three person, four person, each addt'l person)

1 2 3 4 +ea
1998 (8050, 10850, 13650, 16450, + 2800)
1999 (8240, 11060, 13880, 16700, + 2820)
2000 (8350, 11250, 14150, 17050, + 2900)
2001 (8590, 11610, 14630, 17650, + 3020)
2002 (8860, 11940, 15020, 18100, + 3080)
2003 (8980, 12120, 15260, 18400, + 3140)
2004 (9310, 12490, 15670, 18850, + 3180)















REFERENCES


Andersen, R. M. (1995). Revisiting the behavioral model and access to medical care:
Does it matter? Journal of Health and Social Behavior, 36, 1-10.

Andrews, S., Williams, A. B., & Neil, K. (1993). The mother-child relationship in the
HIV-1 positive family. Image: Journal of Nursing Scholarship, 25(3), 193-198.

Aynalem, G., Mendoza, P., Frederick, T., & Mascola, L. (2004). Who and why? HIV-
testing refusal during pregnancy: Implication for pediatric HIV epidemic disparity.
AIDS and Behavior, 8(1), 25-31.

Basu, S. (2004). AIDS, empire, and public health behaviorism. International Journal of
Health Services, 34(1), 155-167.

Beatty, L.A., Wheeler, D., & Gaiter, J. (2004). HIV prevention for African Americans:
Current and future directions. Journal of Black Psychology, 30(1), 40-58.

Beauchamp, T. L., & Childress, J. F. (1994). Principles of biomedical ethics (4th ed).
New York: Oxford University Press.

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(5), 316-319.

Browne, A. J. (2001). The influence of liberal political ideology on nursing science.
Nursing Inquiry, 8(2), 118-129.

Bunting, S. M. (1996). Sources of stigma associated with women with HIV. Diversity in
Health Care, 19(2), 64-74.

Cargill, V. A., Stone, V. E., & Robinson, M. R. (2004). HIV treatment in African
Americans: Challenges and opportunities. Journal of Black Psychology, 30(1),
24-39.

Centers for Disease Control and Prevention (CDC) (2004). HIV/AIDS surveillance
report (Vol. 16). Atlanta: Author.

Centers for Disease Control and Prevention (CDC) (2005). WISQARS leading causes of
death reports, 1999 2003. Retrieved February 25, 2006, from
http://webappa.cdc.gov/Sasweb/ncipc/leadcaus 10.html









Clarke, L. (1999). The person in abortion. Nursing Ethics, 6(1), 37-47.

Dunbar, H. T., Mueller, C. W., Medina, C., & Wolf, T. (1998). Psychological and
spiritual growth in women living with HIV. Social Work, 43(2), 144-155.

Ebrahim, S. H., Anderson, J. E., Weidle, P., & Purcell, D. W. (2004). Race/Ethnic
disparities in HIV testing and knowledge about treatment for HIV/AIDS: United
States, 2001. AIDS Patient Care and STDs, 18(1), 27-33.

Farmer, P. (2003). Pal uthgic ofpower: Health, human rights, and the new war on the
poor. Berkeley: University of California Press.

Fu-Chang Tsai, D. (2001). How should doctors approach patients? A Confucian
reflection on personhood. Journal of Medical Ethics, 27(1), 44-51.

Glanz, K., Rimer, B. K., & Lewis, F. M. (Eds.) (2002). Health behavior and health
education (3rd ed.). San Francisco: Jossey-Bass.

Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. New York:
Simon & Schuster.

Hall, W. A., & Callery, P. (2001). Enhancing the rigor of grounded theory: Incorporating
reflexivity and relationality. Qualitative Health Research, 11(2), 257-273.

Harris, J. (1999). The concept of personhood and the value of life. Kennedy Institute of
Erthi < Journal, 9(4), 293-308.

Henry J. Kaiser Family Foundation. (1999). Hearing their voices: A qualitative research
study on HIV testing and higher-risk teens. Menlo Park, CA: Author. Retrieved April
10, 2006, from http://www.kff.org/youthhivstds/1492-index.cfm

hooks, b.(2003). Rock my soul: Blackpeople and self-esteem. New York: Atria.

Ingram, D., & Hutchinson, S. A. (1999a). HIV-positive mothers and stigma. Health Care
for Women International, 20(1), 93-104.

Ingram, D., & Hutchinson, S. A. (1999b). Defensive mothering in HIV-positive mothers.
Qualitative Health Research, 9(2), 243-258.

Kane, D., & Thomas, B. (2000). Nursing and the "F" word. Nursing Forum, 35(2),
17-24.

Krieger, N. (1994). Epidemiology and the web of causation: has anyone seen the spider?
Social Science & Medicine, 39(7), 887-903.









Lynch, J. W., Kaplan, G. A., & Salonen, J. T. (1997). Why do poor people behave
poorly? Variation in adult health behaviours and psychosocial characteristics by
stages of the socioeconomic lifecourse. Social Science & Medicine, 44(6), 809- 819.

McNair, L. D., & Prather, C. M. (2004). African American women and AIDS: Factors
influencing risk and reaction to HIV disease. Journal of Black Psychology, 30(1),
106-123.

Mellins, C. A., Ehrhardt, A. A., Rapkin, B., & Havens, J. F. (2000). Psychosocial factors
associated with adaptation in HIV-infected mothers. AIDS and Behavior, 4(4),
317-328.

Murphy, D. A., Wilson, C. M., Durako, S. J., Muenz, L. R., & Belzer, M. (2001).
Antiretroviral medication adherence among the REACH HIV-infected adolescent
cohort in the U.S.A. AIDS Care, 13(1), 27-41.

Nazroo, J. Y. (2003). The structuring of ethnic inequalities in health: Economic position,
racial discrimination, and racism. American Journal of Public Health, 93(2), 277-
285.

National Institutes of Health/National Institute of Allergy and Infections Disease (1999).
Reassuring findings about infants exposed to zidovudine. NIHNews Release, January
12, 1999. Retrieved March 15, 2005, from
http://www3.niaid.nih.gov/news/newsreleases/1999/pactg219.htm

Pinch, W. J. (1994). Vertical transmission in HIV infection/AIDS: A feminist
perspective. Journal ofAdvanced Nursing, 19, 36-44.

Poindexter, C. C., & Linsk, N. L. (1999). HIV-related stigma in a sample of HIV-
affected older female African American caregivers. Social Work, 44(1), 46-54.

Rogers, A. S. (2001). HIV research in American youth. Journal ofAdolescent Health,
29(3S), 1-4.

Rogers, A. S., Miller, S., Murphy, D. A., Tanney, M., & Fortune, T. (2001). The TREAT
(therapeutic regimens enhancing adherence in teens) program: Theory and
preliminary results. Journal ofAdolescent Health, 29(3S), 30-38.

Smith, L., Feaster, D. J., Prado, G., Kamin, M., Blaney, N., & Szapocznik, J. (2001). The
psychosocial functioning of HIV+ and HIV- African American recent mothers. AIDS
and Behavior, 5(3), 219-231.

Speigel, K., & Schrimshaw, E. W. (2001). Women with HIV/AIDS described several
forms of positive changes resulting from their illness. Evidence-Based Nursing, 4(4),
126-129.









Steinhart, C., Orrick, J. J., & Simpson, K., Eds. (2002). HIV/AIDSprimary care guide.
Gainesville: University of Florida.

Taylor, B. (2001). HIV, stigma, and health: Integration of theoretical concepts. Journal
ofAdvanced Nursing, 35(5), 792-798.

Todd, T. J., & Miller, S. (2000). Increasing adherence to antiretroviral therapy in
adolescent patients with HIV via novel compliance tools. Journal ofPediatric
Pharmacy Practice, 5(5), 229-234.

Toni, T. D., Masquelier, B., Lazaro, E., Dore-Mbami, M., Ba-Gomeis, F. O., &
Tea-Diop, Y., (2005). Characterization of Nevirapine resistance mutations and HIV
type 1 subtype in women from Abidjan after Nevirapine single dose prophylaxis of
HIV type 1 mother to child transmission. AIDS Research and Human Retroviruses,
21(12), 1031-1034.

Tross, S. (2001). Women at heterosexual risk for HIV in inner-city New York: Reaching
the hard to reach. AIDS and Behavior, 5(2), 131-140.

Whyte, J., Standing, T., & Madigan, E. (2004). The relationship between HIV-related
knowledge and safe sexual behavior in African American women dwelling in the
rural southeast. JANAC, 15(2), 51-58.

Wilkinson, R. G. (1999). Putting the picture together: Prosperity, redistribution, health,
and welfare. In M. Marmot & R. G. Wilkinson (Eds.), Social determinants of health
(pp. 265-266). Oxford, UK: Oxford University Press.

Willey, C., Redding, C., Stafford, J., Garfield, F., Geletko, S., Flanigan, T., Melbourne,
K., Mitty, J., & Caro, J. J. (2000). Stages of change for adherence with medication
regimens for chronic disease: Development and validation of a measure. Clinical
Therapeutics, 22(7), 858- 872.

Yoder, S. D. (2002). Individual responsibility for health: Decision, not discovery.
Hastings Center Report 32(2), 22-31.















BIOGRAPHICAL SKETCH

Patricia Stearnes Robinson graduated from the University of Central Florida with

her Bachelor of Science in Nursing in 1994. She worked as a registered nurse with

pediatric brain-injured patients until she moved to Gainesville to pursue her Master of

Science in Nursing. While in Gainesville, she worked at Shands Hospital as a registered

nurse with both pediatric oncology patients and, later, in the pediatric intensive care unit.

She earned her Master of Science from the University of Florida in 2000. Since that time,

she has worked as a pediatric nurse practitioner in an HIV clinic in Central Florida. She

also taught both undergraduate and graduate nursing students at the University of Florida

while earning her doctorate. Her minor course of study for her doctorate was

epidemiology.




Full Text

PAGE 1

BA RRI ERS TO HUMAN I MMUNODEF I CI ENCY VI RUS (HI V) TREATMEN T PAR TI CI PAT I ON DU RI NG TH E PE RI NA TA L AN D PO STPA RTU M PE RI OD S: A COMPARI SON OF MA TERNAL AND I NFA NT HEAL TH OUTCOMES By PATRI CI A STEARNES ROBI NSON A DI SSER TATI ON PRESENTED TO THE G RADUATE SCHOOL OF T HE UNI VERSI TY OF FL ORI DA I N PARTI AL FUL FI L L MENT OF T HE REQUI REMENTS FOR THE DE GREE OF DOCTOR OF PHI L OSOPHY UNI VERSI TY OF FL ORI DA 2006

PAGE 2

Copy rig ht 2006 by Patricia Stear nes Robinson

PAGE 3

Th is d oc ume nt i s d e dic a te d to str on g h e a lth y mot he rs c hil dr e n, a nd fa mil ie s.

PAGE 4

iv ACKNOWL EDGMENTS I thank my committee, Dr. Shar leen Simpson, Dr. Donna T reloa r, Dr. Sandr a Sey mour, and Dr Nabih Asa l, for all of the ir patient work w ith me. I am also thankful for the suppor t and enc ourag ement of the fac ulty and the students of the FAM De pa rt me nt i n th e Col le g e of Nu rs ing T he y de mon str a te d a n e nd le ss a bil ity to l ist e n to complaints about the rig ors of doc toral work. I am thankful for the examples set by the doctora l fac ulty and espe cially the profe ssionalism of Dr. Ann Horg as and D r. Shawn Kn e ipp I a m a lso tha nk fu l to Dr H os se in Y a ra nd i f or his sig nif ic a nt s ta tis tic a l he lp ( p < 0 0 0 0 1 ) I t h a n k D r S t a c e y L a n g w i c k f o r t h e l e s s o n s i n l a n g u a g e v i e w s h u m i l i t y, and mostly for g etting me . This re sear ch could not have been c ompleted with out the support of Cather ine L ampre cht, MD and A lelia Munroe, MPH. Most of all, I thank Har riet Miller, Dr. E die Deve rs, and Sr. L ouise McEa cher n fo r b e ing bo th m y fa mil y a nd my str on g e st s up po rt e rs I mus t a lso a c kn ow le dg e my car eer mentors and pe rsonal inspirations, Dr. Jo S nider a nd Dr. L inda Hennig All of these pe ople saw more in me than I saw in my self. F or this rea son, I thank them a b s o l u t e l y.

PAGE 5

v TAB L E OF CONTENTS P age A C K N O W L E D G M E N T S ................................................ iv L I S T O F T A B L E S ..................................................... vii i L I S T O F F I G U R E S ..................................................... ix A B S T R A C T ........................................................... x CHAPTER 1 I N T R O D U C T I O N .................................................... 1 P r o b l e m S t a t e m e n t .................................................... 2 P u r p o s e ............................................................. 3 H y p o t h e s i s .......................................................... 3 V a r i a b l e s ........................................................... 4 T e r m i n o l o g y ......................................................... 4 A s s u m p t i o n s ......................................................... 6 L i m i t a t i o n s .......................................................... 6 B a c k g r o u n d a n d S i g n i f i c a n c e ........................................... 6 T h e o r e t i c a l F r a m e w o r k ............................................... 10 C o n c e p t D e f i n i t i o n s .............................................. 13 Oper ationalized Empirical I ndicators . . . . . . . . . . . . . . . . 13 2 R E V I E W O F T H E L I T E R A T U R E ...................................... 15 P u r p o s e ............................................................ 15 P e r i n a t a l T r a n s m i s s i o n ................................................ 15 H i g h l y A c t i v e A n t i R e t r o v i r a l T h e r a p y ................................... 16 R i s k a n d P r e v e n t i o n .................................................. 17 A d o l e s c e n t A d h e r e n c e ................................................ 18 Adher ence in Other Mar g inalized Groups . . . . . . . . . . . . . . . . 22 S t i g m a T h e o r y a n d H I V .............................................. 24 P s y c h o l o g i c a l D i s t r e s s a n d H I V ......................................... 28 M i n o r i t i e s a n d H I V .................................................. 29 P o s i t i v e A d a p t a t i o n .................................................. 33 M o t h e r i n g b y H I V I n f e c t e d W o m e n ..................................... 34

PAGE 6

vi 3 M E T H O D S ........................................................ 37 P u r p o s e ............................................................ 37 D e s i g n ............................................................ 37 S e t t i n g ............................................................ 38 S u b j e c t s ........................................................... 39 S a m p l e S e l e c t i o n ................................................. 39 I n c l u s i o n C r i t e r i a ................................................. 39 E x c l u s i o n C r i t e r i a ................................................ 39 D a t a A n a l y s i s ....................................................... 40 P r o c e d u r e .......................................................... 41 D a t a C o l l e c t i o n ..................................................... 41 P o t e n t i a l H e a l t h R i s k s ................................................ 42 P o t e n t i a l H e a l t h B e n e f i t s .............................................. 42 P o t e n t i a l F i n a n c i a l R i s k ............................................... 42 P o t e n t i a l F i n a n c i a l B e n e f i t s ............................................ 42 C o n f l i c t s o f I n t e r e s t .................................................. 42 4 R E S U L T S ......................................................... 43 P u r p o s e ............................................................ 43 D e s c r i p t i o n o f S a m p l e ................................................ 43 V a r i a b l e s .......................................................... 44 H y p o t h e s i s ......................................................... 44 R e s e a r c h Q u e s t i o n O n e ............................................ 45 R e s e a r c h Q u e s t i o n T w o ............................................ 46 R e s e a r c h Q u e s t i o n T h r e e ........................................... 48 R e s e a r c h Q u e s t i o n F o u r ............................................ 50 S u p p o r t f o r H y p o t h e s i s ............................................... 51 5 D I S C U S S I O N A N D I M P L I C A T I O N S ................................... 53 P u r p o s e ............................................................ 53 D e s c r i p t i v e S t a t i s t i c s ................................................. 55 L a b o r a t o r y R e s u l t s ................................................ 55 M e d i c a t i o n A d h e r e n c e ............................................. 56 A p p o i n t m e n t A d h e r e n c e ........................................... 58 S u m m a r y o f F i n d i n g s ................................................. 58 M i s s i n g a n d E x c l u d e d D a t a ............................................ 59 E t h i c a l C o n s i d e r a t i o n s ................................................ 60 S i g n i f i c a n c e ........................................................ 64 I m p l i c a t i o n s f o r F u t u r e R e s e a r c h ........................................ 67 S u g g e s t i o n s f o r F u r t h e r R e s e a r c h ....................................... 70

PAGE 7

vii A P P E N D IX A D A T A C O L L E C T I O N I N S T R U M E N T .................................. 72 B EXPECTED VAL UES FOR DAT A COL L ECTI ON . . . . . . . . . . . . 73 R E F E R E N C E S ........................................................ 75 B I O G R A P H I C A L S K E T C H .............................................. 79

PAGE 8

vii i L I ST OF TAB L ES T ab l e p age 4-1 Number a nd perc ent distribution of maternal sa mple by ag e and r ace . . . . 44 4-2 Number a nd perc ent distribution of maternal sa mple by dichotomous v a r i a b l e s ........................................................ 44 4-3 Distribution of cases a nd compar ison gr oups by independe nt variable s of exposure to barrie rs and e x posure to fa cilitators, and by outcome mea sures of favor able labor atory results, medica tion adhere nce, a nd appointment a d h e r e n c e ....................................................... 46 4-4 Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of l a b o r a t o r y r e s u l t s ................................................. 47 4-5 Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of m e d i c a t i o n a d h e r e n c e .............................................. 48 4-6 Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of a p p o i n t m e n t a d h e r e n c e ............................................ 48 4-7 Relationship between struc tural bar riers a nd structura l fac iliti es and the outcome mea sures of la boratory results, medica tion adhere nce, a nd a p p o i n t m e n t a d h e r e n c e ............................................ 49 4-8 Summary of log istic reg ression ana ly sis predicting favor able labor atory r e s u l t s .......................................................... 50 4-10 Summary of log istic reg ression ana ly sis predicting favor able a ppointment a d h e r e n c e ....................................................... 52

PAGE 9

ix L I ST OF F I GURES Fi gu re p age 1-1 Routine prenata l scree ning a nd car e of H I V-positive women . . . . . . . . 7 1 2 S t i g m a t h e o r y .................................................... 11 1 3 P r o p o s e d r e v i s i o n t o s t i g m a t h e o r y ................................... 12 3 1 S u b j e c t s e l e c t i o n ................................................. 38 5 1 S t i g m a t h e o r y i n t h i s s t u d y.......................................... 54 5-2 Florida Depa rtment of He alth announce ment, circa 1995 ..................66 5-3 Florida Depa rtment of He alth announce ment, circa 2005 ..................69

PAGE 10

x AB STR AC T Abstrac t of Dissertation Prese nted to the Gra duate School of the Unive rsity of F lorida in Partial Fulf illment of the Requirements for the Deg ree of Doc tor of Philosophy BA RRI ERS TO HUMAN I MMUNODEF I CI ENCY VI RUS (HI V) TREATMEN T PAR TI CI PAT I ON DU RI NG TH E PE RI NA TA L AN D PO STPA RTU M PE RI OD S: A CO MPA RI SON OF MA TE RN AL AN D I NF AN T H EA L TH OU TCO ME S By Patricia Stear nes Robinson May 2006 Chair: Sharlee n Simps on Major De partment: Nursing Pub lic he a lth e ff or ts h a ve y ie lde d a dr a ma tic de c re a se in t he nu mbe r o f i nf a nts born Human I mmunodeficienc y Virus (HI V)-positive throug h ear ly identification of materna l HI V status during preg nancy Despite this overw helming suc cess with fe tal and inf a nt o utc ome s, ma te rn a l ou tc ome s r e ma in d ism a l. T he pu rp os e of thi s st ud y wa s to deter mine if previously published barr iers to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equa lly ; and to deter mine if the proposed c onstructs of structur al fa cilitators or structura l barrie rs af fec t adher ence in the pe ri na ta l dy a d, the po stp a rt um m oth e r, or the e xpos e d in fa nt. A d e sc ri pti ve a nd a na ly tic c a se c omp a ri so n s tud y wa s c on du c te d o n 1 00 pr e g na nt w ome n, a nd the y a nd the ir inf a nts wer e followe d from entr y into the high r isk obstetrical clinic with a dia g nosis of HI Vi n fe ct i o n t o 1 8 m o n t h s p o s t p ar t u m Al l d at a c o l l ec t ed we re co l l ec t ed t h ro u gh retrospe ctive cha rt revie ws. Data wer e ana ly zed throug h both descriptive statistics and

PAGE 11

xi logistic re g ression ana ly sis. St ructura l fac ilitators (transportation to clinic, Ta rg eted Outrea ch to Preg nant Women Act involvement [to provide transportation, emotional su pp or t, a nd /or e xpe dit e d p a pe rw or k], H I V n ur se a nd so c ia l w or ke r i nv olv e me nt, medication dispensed [instead of pr escr iptions given], primary car e ava ilable onsite, int a ke pa pe rw or k n ot n e c e ssa ry a nd /or pr e vio us ly e sta bli sh e d a nd on g oin g re la tio ns hip with case manag er) wer e assoc iated with infant a dhere nce to HI V-re lated ca re, a nd str uc tur a l ba rr ie rs (i ns ur a nc e c omp a nie s, inc lud ing Me dic a id H MO s, tha t li mit a c c e ss t o compre hensive c are in the subspecialty HI V clinic or r equire ment of an individual to link herse lf with follow-up ca re, inc luding the r equire ment of selfinitiated case manag ement be fo re c lin ic a pp oin tme nt i s g ive n) we re a sso c ia te d w ith de c re a se d ma te rn a l a dh e re nc e in the postpartum per iod. Sugg estions were made f or fur ther re sear ch.

PAGE 12

1 CH APT ER 1 I NTRODUCTI ON Hu ma n im mun od e fi c ie nc y vir us (H I V) inf e c tio ns c on tin ue to i nc re a se in incidence and pre valenc e ac ross minority populations in the United States (Centers f or Disease Control [ CDC] 2005). I n minorities, 64% of transmissions occur in wome n, 90% of whom we re infe cted throug h heter osexual contact, with 74% of those inf ections occur ring in non-Hispanic blac ks (CDC, 2005). Once infec ted, the dea th rate f rom a c qu ir e d im mun e de fi c ie nc y sy nd ro me (A I DS) fo r b la c k w ome n in the Un ite d St a te s is nine times highe r than for Cauca sian women (CDC, 2004). Minority adolesc ent fe males demonstrate the most rapidly incre asing incidence of new infec tions (Roge rs, 2001). The HI V diag nosis usually comes a s the result of r outine prena tal testing. I t is not l inked to sy mptoms ex perie nced or any perc eived hig h-risk beha viors; rather the tests are mandated by public health policy to reduc e ra tes of tra nsmissi on to fetuses. When a ne w H IV d i a g n o s i s i s m a d e p r e g n a n c y i s t u r n e d i n t o p a t h o l o g y f o r m a n y m i n o r i t y w o m e n It is particular ly hard to e ng ag e and r etain y oung postpartum mothers in car e, bec ause the y do not link well to the curre nt network of a dult HI V trea tment center s desig ned to trea t men. HI V-positive women a re disenf ranc hised from both society and tre atment fac iliti es due to multiple inequalities. Stigma re lated to ra ce, inc ome, ag e, g ender educa tion level, ear ly -onset par enting sexualit y substance a buse, and H I V infec tion are prese nt, and a pe rce ived indiffer ence by the medica l community is often re ported (I ng ra m & Hu tc hin so n, 19 99 a ). Con ve nti on a l w isd om w ith in t he tr e a tme nt c omm un ity

PAGE 13

2 re ite ra te s th a t a do le sc e nts a re lik e ly to f a il t o a dh e re to t re a tme nt b e c a us e of the se barr iers (Rog ers, Miller, Murphy Tanne y & F ortune, 2001). I n addition, denial of the HI V diag nosis and ref usal to act in their own be st interest are considere d important re a so ns fo r c on tin ue d r isk y be ha vio rs (R og e rs e t a l., 20 01 ). Ye t, a t th e sa me po int s in tim e y ou ng mot he rs of te n c a re fo r t he ir e xpos e d in fa nts wi th v ig or a nd su c c e ss t ha t is not evident in their own c are This raises a number of que stions. S pecif ically what ar e the conditions and attributes that a llow y oung mothers to provide this specia lty car e succe ssfully for their infants, and wha t are the bar riers tha t preve nt them from participa ting in selfcar e? Why are trea tment plans that are eff ective f or other p o p u l a t i o n s n o t e f f e c t i v e f o r m i n o r i t y a d o l e s c e n t m o t h e r s ? P r obl e m St at e m e nt The he alth car e sy stem has strateg ies in place to identify HI V-positive women during preg nancy for the pur pose of pr eventing viral transmission to ex posed fe tuses. Structural fa cilitators are in place to incr ease the ability of HI V-positive pre g nant women to c omp ly wi th p re na ta l a nd inf a nt c a re T he se fa c ili ta tor s in c lud e a uto ma tic re fe rr a l to the HI V t re a tme nt t e a m in the Hi g h Ri sk Ob ste tr ic a l Cli nic (H RO B ) u po n d ia g no sis imm e dia te a c c e ss t o a nti re tr ov ir a l me dic a tio n ( AR T) f a st t ra c k p a pe rw or k to g a in emer g ency Medica id cover ag e, fie ld worker s from the Ta rg eted Outr eac h to Preg nant Women Act (TOPWA) to provide support and tr ansportation to clinic and f inancial appointments, and a r eg istered nur se and a social worke r who both attend H ROB. Once born, neona tes rec eive their initial HI V blood test, a two we ek follow-up appointment at the HI V spec ialty clinic for r esults, and ART is dispensed for use at h o m e U n l i k e t h e i r n e o n a t e s p o s t p a r t u m m o t h e r s h a v e n o a u t o m a t i c r e f e r r a l t o t h e H IV sp e c ia lty c lin ic a nd ha ve no me c ha nis m in pla c e to e ns ur e the y ob ta in A RT p ri or to

PAGE 14

3 discharg e af ter de livery Postpartum mothers are often lost to follow-up af ter de livery until they are preg nant ag ain or until they prese nt for ca re a fter the onset of HI V-re lated sy mptoms. The interruption of ca re be tween de livery and sy mptom onset has the po te nti a l to c re a te the bu rd e ns of inc re a se d u til iza tio n o f t e rt ia ry he a lth c a re re so ur c e s, infant/child aba ndonment or orpha nag e, and pe rsonal suff ering for the mother s. The lac k of mater nal selfcar e is attributed in the litera ture to pover ty educa tion, y outh, and clinical depr ession (Murphy Wil son, Durako, Mue nz, & B elzer, 2001). The se mater nal char acte ristics are share d by both the mother and the baby postpartum, but may not alway s serve as bar riers to he alth car e for them both. P ur pos e Th e pu rp os e of thi s st ud y wa s to de te rm ine if pr e vio us ly ide nti fi e d b a rr ie rs to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equally ; and to deter mine if the propose d constructs of struc tural fac ilitators or structural ba rrier s are associate d with HI V trea tment adher ence in the pe ri na ta l dy a d, the po stp a rt um m oth e r, or the e xpos e d in fa nt. Hyp ot he sis The following two-tailed hy pothesis was tested: Mate rnal de pression or other mental illness, prox imity of HI V diag nosis, race income, adole scenc e, hig h school educa tion, substance abuse pill burden, structura l barrie rs to acc ess, and struc tural fac ilitators to care are associate d with a sig nificant diff ere nce in a dhere nce a s measure d by the la bo ra tor y re su lts mi sse d me dic a tio n d os e s, a nd de la y e d o r m iss e d c lin ic a pp oin tme nts of HI Vpo sit ive pr e g na nt w ome n, po stp a rt um m oth e rs o r t he ir inf a nts I n order to determine the answe r, the following rese arc h questions were posed.

PAGE 15

4 1. Ar e the thr e e g ro up s ( po stp a rt um m oth e rs in fa nts a nd dy a ds ) h omo g e ne ou s w ith respe ct to the three outcome mea sures (la boratory results, medica tion adhere nce, a nd appointment adher ence )? 2. I s th e re a sig nif ic a nt r e la tio ns hip be tw e e n th e ou tc ome va ri a ble s ( la bo ra tor y re su lts appointment adher ence and medica tion adhere nce) and demog raphic varia bles (ag e, rac e, mental illness, pover ty new diag nosis, pill burden, hig h school educ ation, su bs ta nc e a bu se ) a mon g the thr e e g ro up s? 3. I s there a signific ant associa tion between the outcome var iables (medic ation adher ence labora tory results, and a ppointment adhere nce) and structur al fa cilitators a nd str uc tur a l ba rr ie rs ? 4. I s materna l depre ssion or other mental illness, proxim ity of HI V diag nosis, race income, adole scenc e, hig h school educ ation, poverty substance a buse, pill burden, structura l barrie rs to acc ess, and struc tural fa cilitators to car e sig nificantly differ ent as mea sured by the labora tory results, medica tion adhere nce, a nd appointment a dh e re nc e in H I Vpo sit ive pr e g na nt w ome n, po stp a rt um m oth e rs o r t he ir inf a nts ? Variables Th e ind e pe nd e nt v a ri a ble s a re ma te rn a l de pr e ssi on or oth e r m e nta l il lne ss, proxi mity (new ness) of H I V diag nosis, race income, adole scenc e (a g e), hig h school educa tion, substance abuse pill burden, structura l barrie rs to acc ess, and struc tural fac ilitators to acce ss. The depe ndent var iables ar e consistent, but def ined diffe rently for e a c h g ro up be c a us e the sta nd a rd of c a re is d if fe re nt d ur ing e a c h p e ri od T he se ou tc ome measure s include labora tory values, pe rce ntag e of missed doses, a nd delay ed or missed appointments for c are Term inology F or the pu rp os e of thi s st ud y th e stu dy va ri a ble s a re op e ra tio na lize d a s f oll ow s. Adolesce ncema terna l ag e less than 24 y ear s at time of conc eption Delay ed ca reH I V-re lated ca re not provide d during the re commended w indow of time, or postpartum materna l car e or infa nt immuni zations not provided during the re c omm e nd e d w ind ow of tim e Dy adunit that includes the pre g nant woman a nd her f etus during the pre natal and antena tal periods

PAGE 16

5 Educa tionattainment of a hig h school diploma, GED, or c urre nt enrollment in an equivalent pr og ram I ncomehouse hold income less than pover ty threshold for the y ear under investiga tion L abora tory results (dy ad)CD4 a nd HI V viral loads within expected r ang e base d on CDC information for the y ear under obse rvation L abora tory results (infa nt)HI V DNA PCRs remain neg ative at the thr ee rec ommended interva ls, and HI V EL I SA /W estern B lot is nega tive at 18 months L abora tory results (mater nal)CD4 and H I V viral loads r emain within expected rang e base d on CDC recommenda tions for the per iod under obser vation Materna l depre ssion or other mental illnessAny psy cholog ical diag nosis in the prena tal medical c hart that is found in DSM-I V, whethe r or not tre ated by medication or thera py Mis se d a pp oin tme nts a pp oin tme nts mis se d a nd no t r e sc he du le d > 10 % o f t he tim e Mis se d d os e s > 10 % o f A RT d os e s mi sse d b y se lf re po rt a t c lin ic vis it Pill burdennumbe r of pills/doses per da y g rea ter than e stablished norm during the y ear presc ribed (i.e ., 5 Virac ept and 1 Combivir tabs twice da ily = pill burden of 12 t a b s / d a y = a v e r a g e p r e n a t a l p i l l b u r d e n f r o m 2 0 0 0 t o 2 0 0 4 ; a n d A Z T s yr u p e v e r y 6 hours = 4 dose s/day = ave rag e neona tal pill burden during entire study period) Prox imity of HI V diag nosisdetermination of whe ther HI V seropositivity was e sta bli sh e d d ur ing the c ur re nt p re g na nc y ; a me a su re of the ne wn e ss o f t he dia g no sis Ra c e s e lf -r e po rt e d r a c e (i n a c c or da nc e wi th f e de ra l g uid e lin e s) Structural bar riers to a cce ssI nsuranc e compa nies, including Medica id HMOs that limit acc ess to compre hensive c are in the subspecialty HI V clinic or r equire ment of an individual to link herself with follow-up c are including the require ment of selfinitiated case manag ement bef ore c linic appointment is given Str uc tur a l f a c ili ta tor s tr a ns po rt a tio n a va ila ble to c lin ic T a rg e te d O utr e a c h to Preg nant Women Act (TOPWA) involvement to provide tra nsportation, emotional su pp or t, a nd /or e xpe dit e d p a pe rw or k, HI V n ur se a nd so c ia l w or ke r i nv olv e me nt, medication dispensed ( instead of pr escr iptions given), pr imary car e ava ilable onsite, int a ke pa pe rw or k n ot n e c e ssa ry o r p re vio us ly e sta bli sh e d o ng oin g re la tio ns hip wi th case manag er S u b s t a n c e a b u s e i l l i c i t d r u g o r a l c o h o l u s e d u r i n g p r e g n a n c y

PAGE 17

6 As sum pt ion s As with any retrospe ctive cha rt ana ly sis, it was assumed that information contained in the c harts wa s acc urate and complete in both reporting by clients and rec ording by staff. Lim it at ion s Th e g e ne ra liza bil ity of thi s st ud y is l imi te d to wo me n in so c ia l a nd e c on omi c circ umstances similar to those of the subje cts. I ncar cer ated a nd/or tox icolog y scre enpo sit ive mot he rs a t de liv e ry we re e xclu de d f ro m th is s tud y a s w e re ne on a te s b or n w ith pr e ma tur ity sig nif ic a nt e no ug h to pr e ve nt m oth e rs fr om p ro vid ing c a re fo r t he ir inf a nts Thoug h foster pa rents or e x tended f amily and fr iends ca re f or many HI V-e x posed infants, this study soug ht to ex amine mothers who w ere primary car e provide rs so that inf a nts a nd mot he rs sh a re d th e sa me ma te rn a l r isk fa c tor s a t th e sa me tim e s. Th is l imi ts this study s ability to describe the circ umstances of broken f amilies. Also excluded were the sparse subjects that elec ted to seek the ir specia lty car e fr om community providers. Background and S ignif icance The public he alth and medica l communities have g iven extensive eff ort to the pr ote c tio n o f f e tus e s e xpos e d to HI V. Th e g old sta nd a rd of c a re in t he Un ite d St a te s is the pu bli c he a lth ini tia tiv e to e ng a g e the mot he rba by dy a d in the pr e ve nti on e ff or ts detailed in the Pediatr ic AI DS Cli nical Tria l Group Study Protocol 076 (PACTG 076). Th is e ff or t f oc us e s o n th e ou tc ome me a su re of pr e ve nti on of HI V t ra ns mis sio n to exposed babies. The pr otocol is overwhe lmingly succe ssful in its i ntent to reduc e transmission within t he United States, re ducing the ra te from a natura l history of 25-28% (NI H/NI AI D, 1999) to about 2% ( NI H/NI AI D, 1999). The follow-up study PACTG 219C, continues to look at the health of the exposed infant as it g rows into adulthood

PAGE 18

7 (NIH/NIAID, 1999). Neither study considers the outcome of maternal health. Since this research produces such convincing evidence of its efficacy, funding and policy decisions are made based on adherence of clinics to this protocol. In the era of evidence-based practice, the separation of maternal and child health at the time of delivery may unintentionally inform providers decisions to cease to address maternal health after delivery. The treatment guidelines continue in a linear fashion along the prevention of transmission, stating guidelines for medicating the neonate with antiretroviral medication, with no recommendations for maternal care after delivery (Figure 1). Mothers cease to be treated by the protocol for their own infections, because they cease to be regarded once they are not carrying babies, and thus not a threat of transmission to children. Figure 1-1.Routine prenatal screening and care of HIV-positive women

PAGE 19

8 Yoder (2002) c ontends that participa tion in the social instit utions of medicine and public health ca n shape va lues, assumptions, ex planatory models, and problemsolving a p p r o a c h e s o f p r o v i d e r s A l t r u i s t i c n o v i c e p r o v i d e r s w h o e n t e r t h e f i e l d o f H IV trea tment and pre vention are quickly influence d by the exist ing tr eatment c ulture and explanatory models. Mothers that do not follow-up with their own H I V-re lated ca re a fter delivery are often view ed by providers a s being intrinsically unmotivated or selfdestructive. Provide rs often f eel that the a mount of attention and ef fort that has be en g iven to mothers during the pre natal per iod should be enoug h to retain them in ca re a fter de liv e ry So th e ir fa ilu re is e xpla ine d a wa y in t he lit e ra tur e by de sc ri bin g the m a s li ke ly to fail based on the ir intrinsic char acte ristics (Ander sen, 1995) without looking a t why that is so. Often, intrinsic fac tors such as r ace income, and a g e ar e used to de scribe awa y failure without ex amination of the c ontext in which the fa ilure occ urs. The healthca re c ommunity cre ates this context. Thus, it is important to challeng e the e x isting beliefs a nd actions of the public he alth community and hea lth providers who dismiss the mot he rs a s u nr e a c ha ble by loo kin g a t th e sp e c if ic c ir c ums ta nc e s th a t c on tr ibu te to t he ir health outcomes. This re quires that re sear ch take into account struc tural compone nts that ma y inf or m ma te rn a l c ho ic e s. The public he alth community deter mines when tre atment occ urs, when it does not, and on what pr iorities resea rch f unds are spent. The pr evale nt themes in the literature that underlie be liefs about HI V pre vention, infec tion, and adher ence should be compar ed to the life c ircumstance s of minority y oung fema les infec ted with HI V. B e fo re a vu lne ra ble po pu la tio n is dis mis se d a s r e fr a c tor y to h e lp, a n e a rn e st s c ie nti fi c

PAGE 20

9 e f f o r t s h o u l d b e e m p l o y e d t o u n d e r s t a n d t h e f i t b e t w e e n t h e s y s t e m a n d t h e s e m o t h e r s It is imperative to stop dismis sing a s unrea chable those who may simply be misunderstood. B e c a us e of the re pe tit ive c la ims tha t mi no ri tie s c a nn ot o r d o n ot a dh e re to thera py studies show that minorities are less likely to rec eive a nti-retrovira l medications (Carg ill, S tone, & Robinson, 2004), though these are the people most likely in need of the ra pe uti c e ff or ts ( B a su 2 00 4) O uts ide of the Un ite d St a te s, po pu la tio ns pr e vio us ly dismiss ed as unr eac hable a re showing signific ant adhe renc e throug h the re moval of str uc tur a l ba rr ie rs to c a re I n th e c e ntr a l pl a in o f H a iti a dh e re nc e to A RT h e ig hte ne d in the pre sence of the c oncomitant provision of food, opportunity to ge nera te income, a nd a continuity and par ity of hea lth care distribution (Farme r, 2003). To unde rstand the dismiss al of a dhere nce pote ntial in HI V-infe cted y oung mothers, a look at the history of HI V i n th e Un ite d St a te s is he lpf ul. HI V first appe are d in the United States in the ea rly 1980s as a disea se that inf e c te d a nd kil le d ma le s w ho we re ide nti fi e d a s h a vin g se x with oth e r m a le s ( MSM ) (CDC, 2004). Soon after HI V and the r esultant AI DS started infe cting and killing int ra ve no us dr ug us e rs (I VD Us ), a s w e ll a s th e fe ma le se xua l or ne e dle -s ha ri ng c on ta c ts of bo th g ro up s ( CD C, 2 00 4) H I V i n th e Un ite d St a te s is we ll e sta bli sh e d th ro ug h th is his tor y a s a dis e a se of sti g ma tize d a nd vu lne ra ble g ro up s. I n 1 98 1, tw o p hy sic ia ns in New Y ork re quested funding from the Na tional I nstitut es of He alth and the Ame rica n Founda tion for AI DS Resear ch to investig ate the pa rticular e ffe cts of HI V on women. B o t h g r a n t o r s r e j e c t e d t h e p r o p o s a l s s t a t i n g o n t h e r e t u r n c r i t i q u e s t h a t t h e e f f e c t o f H IV infec tion on women was not an ur g ent topic for study (B unting, 1996) Resear ch money and ef fort in the United States a nd the world ha ve stay ed foc used on the tre atment of infec ted men and the preve ntion of transmission to fetuses, while the epide mic itself has

PAGE 21

10 moved disproportionately to killi ng y oung minority women (CDC, 2005). Rese arc hers are now fa ced w ith the majority of new infec tions occurr ing in minority populations within the United Statespopulations that likely have dif fer ent priorities and c ir c ums ta nc e s th a n a re a dd re sse d b y the c ur re nt t re a tme nt g uid e lin e s. Th e fi rs t st e p to examining the fit betwe en tre atment ce nters, policy and y oung mothers is to ex plore a nd de sc ri be a ny dif fe re nc e s th a t ma y e xist in the wa y the HI V t re a tme nt n e tw or k tr e a ts y ou ng mot he rs a nd a ny po ssi ble su bs e qu e nt d if fe re nc e s in ou tc ome s. Theoretic al F ram ework S t i gm a t h eo ry i s a m i d d l e r an ge t h eo ry t h at wa s fi rs t p u b l i s h ed i n s o ci o l o gy literature by Goffma n (1963). Stigma was de fined by the Gre eks as a bodily sign of a m o ra l fl aw Go ff m an (1 9 6 3 ), an d wa s a c o m m o n co n ce p t i n t h e f i el d o f s o ci al p s y ch o l o gy in the 1960s. I t was deve loped in to a sociolog ical theor y to expl ain the ca uses of discrimination in society (Goff man, 1963). Conce pts in the theory are the stigma tiz ed attribute, discre dited state, discre ditable state, stig ma, and discr imination (Fig ure 2) Th e se c on c e pts a re inc or po ra te d in to t he the or e tic a l f ra me wo rk of thi s st ud y a lon g wi th four a dditional concepts that be g in to allow for an individuals r esponse to being stigmatized. Goff mans (1963) stigma the ory is comprised of the f ollowing r elational sta te me nts : A s tig ma tize d a ttr ibu te ma y le a d to dis c re dit Discre dit causes stig ma. Stigma justifies discrimination. Di sc ri min a tio n th e n o c c ur s. Discrimination become s part of a n individuals or g roups e nvironment, the ef fec t of which may incre ase the stigma e x perie nced.

PAGE 22

11 Figure 1-2.Stigma theory (Goffman,1963) What happens if a decision is made not to discredit someone with the potential for a stigmatized attribute? It is possible to confer on them protection from the already stigmatized source of this potential stigma. One of the purposes of stigma is to separate normals from the undesirable outcome, or the risk of becoming undesirable One way that normals distance themselves from perceived risk is to blame stigmatized individuals and groups for having the very trait that renders them stigmatized (Goffman, 1963). Because groups are already blamed for being HIV-positive, other groups that are at risk of becoming infected by these flawed individuals would logically be viewed (intentionally or unintentionally) as victims.

PAGE 23

12 It is plausible that there is an actual protective effect of being perceived as a victim, as much as there is a harmful effect of being perceived as stigmatized or flawed. In this study, a model is proposed (Figure 3) that is reflective of both sides of a potential stigmatizing attributeone that reflects possible positive and negative outcomes, simply dependent upon how the health care community and health care policy makers regard the individual or group at entry into care. Figure 1-3.Proposed revision to stigma theory, as adapted from Goffman (1963)

PAGE 24

13 Co nc e pt De finit ion s Stig ma tize d a ttr ibu te Ch a ra c te ri sti c tha t r e du c e s a pe rs on fr om wh ole a nd us ua l to discounted and ta inted; and that may be a pe rce ived phy sical flaw char acte r flaw or tribal membership (G offman, 1963) I n his origina l work, examples of phy sical flaw s wer e disfig ureme nt or blindness; ex amples of c hara cter flaws would be homosex uality or being a convic ted cr iminal; and examples of tribal flaws would be a trait passed throug h family or tradition, such a s skin color or re ligion. Discre ditable stateState in which a n individual with a stigmatized attribute that ca n be hidden e x ists. Hiding an a ttribute such as HI V infec tion prevents being discredited a nd possible stigma. The state involves incre ased stre ss and worr y of being discovere d. Di sc re dit e d s ta te St a te in w hic h a n in div idu a l e xists wh e n a sti g ma tize d a ttr ibu te is kn ow n to no rm a ls. StigmaA be lief or a n attitude that a per son is less than human that is constructed and held by normals as justification for discrimination, or a situation in which an individual is denied full social ac cepta nce ( Goffma n, 1963). DiscriminationThe a ct of pre judicial trea tment. Pr ote c te d c la ss s ta te M e mbe rs hip in a g ro up tha t ha s a sti g ma tize d a ttr ibu te y e t is vie we d a s a vic tim of sti g ma tize d in div idu a ls i ns te a d o f a s a sti g ma tize d in div idu a l. Diminished health outcomeF inal health status re sulting fr om failure to r ece ive health ca re a nd support to the extent they are available for individuals with simil ar me dic a l c on dit ion s. I nc re a se d p ro te c tiv e e ff or tS tr uc tur a l f a c ili ta tor s o utn umb e ri ng tho se a va ila ble to other individuals and g roups with the same stig matized attribute; structural bar riers a re le ss t ha n th os e e nc ou nte re d b y oth e r i nd ivi du a ls a nd g ro up s w ith the sa me stigmatized attribute. Op tim ize d h e a lth ou tc ome F ina l he a lth sta tus re su lti ng fr om b e st p os sib le a c c e ss t o health re source s and support as c ompare d to individuals wit h simil ar me dical c on dit ion s. Operationalized Em pirical Ind icator s Stigmatized attributesRacial minority ag ed 15-23 y ear s, preg nant, HI V-positive, income below pove rty level, educ ation below hig h school level, doc umented history of dr ug or a lc oh ol u se du ri ng pr e g na nc y o r a do c ume nte d h ist or y of me nta l il lne ss.

PAGE 25

14 Di sc re dit a ble sta te A n H I Vpo sit ive mot he r w ho do e s n ot d isc los e he r d ia g no sis to pr ov ide rs a nd /or to h e r c omm un ity o r a n H I Ve xpos e d f e tus or inf a nt w ho se sta tus is e ith e r u nk no wn or is n e wl y kn ow n u po n d isc los ur e of mot he r s H I V s ta tus Di sc re dit e d s ta te A pr e g na nt w oma n o r m oth e r w ho se HI Vpo sit ive dia g no sis is k n o w n t o p r o v i d e r s a n d / o r h e r c o m m u n i t y. Stigmahea lth provider and/or minority community beliefs a nd attitudes that HI Vpo sit ive fe ma le s sh ou ld n ot r e pr od uc e a re to b la me fo r t he ir inf e c tio n, a nd a re le ss valuable tha n their HI V neg ative counte rpar ts. Si nce stig ma is defined a s a fe eling or an attitude, it is not measured in this revie w of medic al documenta tion. Di sc ri min a tio n Ab se nc e of se rv ic e s a va ila ble to o the rs tha t f a c ili ta te pa rt ic ipa tio n in HI V-re lated ca re or prese nce of policies that delay or pre vent par ticipation in HI Vrela ted ca re ( specific ally listed under var iable def initions ). Class protective e ffe ctPresenc e of se rvice s that fac ilitate participation in HI Vrela ted ca re or absenc e of policie s that delay or pre vent par ticipation in HI V-re lated car e (spe cifica lly listed under var iable def initions structura l barrie rs and struc tural fac ilitators). Diminished health outcomeMissed appointments, hig h viral loads and/or poor CD4 c e ll c ou nts by la bo ra tor y me a su re o r m iss e d d os e s o f a nti re tr ov ir a l me dic a tio ns Op tim ize d h e a lth ou tc ome A dh e re nc e to r e c omm e nd e d a pp oin tme nts th e ra pe uti c vir a l lo a ds a nd CD 4 c e ll c ou nts by la bo ra tor y me a su re a nd a dh e re nc e to a nti re tr ov ir a l me dic a tio ns Thoug h HI V-re lated litera ture is peppe red w ith St igma Theory other re asons for tr e a tm e n t f a il u r e p r e v a il M o s t s tu d ie s f o c u s o n h e a lt h b e h a v io r th e o r ie s th a t e mp h a s iz e individual or intrinsic empowerme nt. Many studies detailed in the next chapter descr ibe intrinsic qualities that make individuals likely to fail trea tment. No study was identified that specific ally measure d the role of structura l or extrinsi c bar riers to c are I t may not be just the labeling or stigma that keeps H I V-positive mothers out of c are but rather the pref ere ntial treatment other s may rec eive in their pr esenc e if these same mothers a re de nie d e qu a l tr e a tme nt.

PAGE 26

15 CH APT ER 2 REV I EW O F TH E L I TE RA TU RE P ur pos e Th e pu rp os e of thi s st ud y wa s to de te rm ine if pr e vio us ly ide nti fi e d b a rr ie rs to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equally ; and to deter mine if the propose d constructs of struc tural fac ilitators or structural ba rrier s are associate d with HI V trea tment adher ence in the pe ri na ta l dy a d, the po stp a rt um m oth e r, or the e xpos e d in fa nt. P erinatal Transm ission Pos iti ve e ff e c ts o n f e ta l ou tc ome s ( HI V s ta tus ) w e re re po rt e d b y the fi rs t pe dia tr ic clinical drug trials if preg nant HI V-infe cted wome n took zi dovudine (AZ T) during the course of their pr eg nancy and deliver y The odds of a n infec ted baby being born to an inf e c te d mo the r d e c re a se d f ro m a bo ut 3 0% to a bo ut 2 % w ith the us e of AZ T ( CD C, 2004). This was a landmark f inding f or pre vention of pedia tric AI DS, and was vie wed a s the e nd of AI DS i n c hil dr e n. Sin c e c hil dr e n w ou ld n o lo ng e r c on tr a c t H I V f ro m th e ir mothers, the re asoning followed that ther e would no long er be pediatric HI V in the United States. Many states beg an to re quire HI V testing of pre g nant fe males for the protec tion of the fetuse s, and thus the larg e cohor t of infec ted, minority preg nant adolesc ents and y oung adults continue to be unc overe d throug h diag nosis during preg nancy (CDC, 2004). Exi sting litera ture did not ref lect the r eal r isk that without c ha ng ing so c ia l a nd e c on omi c c ir c ums ta nc e s, the se sa me HI Vne g a tiv e ba bie s a nd the ir

PAGE 27

16 friends w ould gr ow into adolesce nts and ac quire their ow n HI V infec tions because the same socia l and ec onomic challeng es that plag ued their mother s now place them at risk. Pe ri na ta l H I V l ite ra tur e ha s a n o ve rw he lmi ng fo c us on the pr e ve nti on of HI Vinf e c tio n in the fe tus/infant. Hig hly Ac ti ve Ant iRe tr ov ir al T he r apy I t is we ll e sta bli sh e d th a t a g g re ssi ve a nti -r e tr ov ir a l th e ra py c omm on ly re fe rr e d to a s h ig hly a c tiv e a nti -r e tr ov ir a l th e ra py (H AA RT) inc re a se s b oth the du ra tio n a nd qu a lit y of life ( Roge rs, 2001). I mplementation of HAA RT at the clinically indicated time is the standard of car e for infec ted individuals. An entire subset of social scienc e re sear ch focuse s exclusively on way s to increa se adhe renc e to HAART. Mode ls have be en develope d to assess re adiness and sta g e cha ng e for maxi mum success with thera py (G la nz, Rime r, & L e wi s, 20 02 ). I n f a c t, c ur re nt t re a tme nt g uid e lin e s st ro ng ly re c omm e nd tha t H AA RT n ot b e ini tia te d o n p a tie nts un til the y de mon str a te a re a din e ss t o c omp ly wi th t he ra py fo r t he lon g te rm (S te inh a rt O rr ic k, & Simp so n, 20 02 p 5 8) T his rec ommendation is based on wha t is viewed as be st for the pa tient and society Partial c omp lia nc e or sh or t du ra tio n th e ra py of te n r e su lts in m uta te d v ir us e s th a t a re re sis ta nt t o available medication (Steinhar t, Orrick, & Simps on, 2002, p. 103), and multiple drug resistant strains of H I V pre sent insurmountable trea tment obstacles, incr easing the burdens of both cost and mortality to society The sentinel inconsistenc y in the g uid e lin e s is tha t th e on ly de mog ra ph ic fo r w ho m r e a din e ss i s n ot a sse sse d, a nd pa tie ntc e nte re d c ou ns e lin g a bo ve a nd be y on d th e re qu ir e d p os tte st ( re su lts ) i s n ot t he no rm is preg nant fe males (Steinhar t, Orrick, & Simps on, 2002, p. 58). Ac cording to the theories discussed below, the se ar e the most persona lly at-risk of the infec ted population rela ted t o b o t h t h e i r b e h a v i o r a n d b i o l o g y.

PAGE 28

17 Risk an d Pr evention Th e on ly sc ie nti fi c a lly va lid a te d r isk fa c tor fo r c on tr a c tin g HI V i s e ng a g ing in any activity that exchang es infec ted bodily fluid with someone who has H I V (ter med high r isk behaviors) te rm (Steinhart, Or rick, & Simps on, 2002, p. 13). Since sec ondary preve ntion (limi ting the spr ead of HI V by identify ing those alre ady infec ted) is the pr ior iti zed fu nd ing g oa l f or HI V, the qu e sti on to a ns we r i s w hy c e rt a in g ro up s e ng a g e in highrisk behavior s more ofte n than others. B asu (2004) found that the top epidemiolog ical pre dictor for HI V worldwide is a low-income le vel. I t is thi s lowincome leve l that decides the context and conditions of sex ual beha viors. The poor ha ve a n in c re a se d r isk of e xpos ur e to p a tho g e nic sit ua tio ns w he the r s a nit a tio n, c ro wd ing o r i n thi s c a se th e pr e se nc e of mor e HI Vinf e c te d c on ta c ts i s th e c a us e I n th e Un ite d St a te s, being black is consider ed a r isk factor for HI V, bec ause be ing bla ck is often use d intercha ng eably with the socioec onomic varia ble of be ing poor Nazroo (2003) arg ued a g a ins t be ing bla c k a s a ri sk fa c tor fo r d ise a se by sh ow ing tha t a ft e r i nc ome a dju stm e nt, health disparities within the United States leve led consider ably The ba lance of the differ ence may be attributable to experience s and aw are ness of ra cism (Nazroo, 2003). Krieg er ( 1994) asse rts that population patterns of he alth stem from ec onomic and so c ia l a c tiv iti e s a nd ine qu a lit ie s. B e ha vio r m od e ls t ha t a re the ba c kb on e of he a lth sc ie nc e s a ssu me ind ivi du a l a g e nc y tha t ma y no t e xist w ith ou t pu bli c he a lth e ff or ts t o ide nti fy a nd re mov e ba rr ie rs O ft e n, the c omm un ity of sc ie nti sts tha t r a tio n r e so ur c e s to the HI V-infe cted c ommunity disreg ards or neg atively reg ards individuals who ar e in the wrong g roup at the w rong time. Ba su (2004) re counts two studies, one in which a miner explains t he conte x t of his highrisk rec rea tional behavior ( his profession has a n injury rate of 42% a nd he is neve r ce rtain he w ill li ve throug h his work day ), and one in which

PAGE 29

18 prostitutes complain of their lac k of opportunity The miner is labe led by the psy cholog ists as in denial a nd tag g ed with a low self-e steem. The rese arc hers labe l the prostitutes as liar s who a re de ny ing the ir own ag ency when they claim that prostitution i s the only source of ava ilable income. B asu (2004) calls AI DS the sy mptom in these situations, not the disease. Adolescent Adhere nce A re view of the literature was c onducted using the CI NAHL and Medline databa ses dated 1997 to 2004 with a limitation to English lang uag e. Ke y words adole scent, adhe renc e, a nd HI V we re utilized. Finding s were limit ed, and of those, few repor ted succ essful methods of a dhere nce. Most litera ture re ported re asons fo r f a ilu re ins te a d o f w a y s to pr omo te a dh e re nc e T ho ug h mu lti ple c e nte rs sp e c ia lize in adolesc ent HI V identifica tion and trea tment throug hout the country only one g roup of pr ov ide rs ha ve joi ne d to fo rm a lize te st, a nd pu bli sh fi nd ing s. Me mbe rs of the REA CH Project (Rea ching for Excellenc e in Adolesc ent Care and He alth) comprise this group (Murphy et al., 2001). No re sear ch conduc ted by individual providers wa s identified in the literature Th is i s p ro ba bly re la te d to the fe de ra l ma nd a te tha t H I V f un din g to g o to su bsp e c ia lty trea tment center s, and the re sultant tendency of uninsured a nd underinsure d y outh to seek c a re a t th e se c e nte rs E ve n w ith su bsp e c ia lty c a re a do le sc e nc e a nd mot he rh oo d r e ma in turbulent times in the lives of HI V-positive women. Pinch (1994) discusses the vulnera bility of adole scents in par ticular, stating that the invincibility perc eived by adolesc ent g irls makes them a spe cial conc ern w ith reg ard to their ow n HI V infec tions. An a dd iti on a l c on c e rn is t he ina bil ity fo r m a ny a do le sc e nt f e ma le s to pr ob le msolve due to psy cholog ical immaturity and incomplete e ducations (Pinch, 1994).

PAGE 30

19 L ooking a t the theore tical per spectives to adhe renc e to trea tment in HI V-infe cted adolesc ents, models in the literature g ener ally discuss adolesce nt development and behavior theories a ppropriate to rea ch adole scents and a ffe ct desire d outcomes. As mentioned ea rlier, one prog ram of r esea rche rs in the United States has c ommitt ed themselves to re sear ching the spec ific nee ds of adolesc ents with HI V infec tion. The g roup of pr oviders most often see n in the literature is phy sicians who provide c a re a t 15 c lin ic s in 13 ma jor c iti e s th ro ug ho ut t he Un ite d St a te s to fo rm the REA CH Project). The ir finding s repor t a strong association betwe en medica tion adhere nce a nd reduc ed HI V viral loads ( Murphy et al., 2001). F indings a lso showed less than 50% of subjects re ported a cce ptable adhe renc e to their tre atment re g imen, and that adhe renc e seems to be a serious problem a mong H I V-positive adolesc ents (Murphy et al., 2001). The re sear cher s cited both ea se of medic ation use, also known a s decr ease d pill burden, and tre atment of de pression as two important issues to be a ddresse d when tre ating a do le sc e nts (M ur ph y e t a l., 20 01 ). Th ou g h th e REA CH Pr oje c t pr ov ide rs c ome c los e st t o a dd re ssi ng the po pu la tio n o f p re g na nt a do le sc e nts th e ir fi nd ing s d o n ot p ro vid e se pa ra te statistics on preg nant adolesc ents, nor do they sug g est interventions spec ific to the nee ds of pre g nant adolesc ents. I n the ana ly sis by Murphy et al. (2001) strong social support wa s repor ted as nece ssary to adher ence and a linea r dec rea se in adhe renc e wa s noted when r eg imens beca me more c omplex This is not surprising, thoug h it is i mportant to note that re g ime ns fo r p re g na nt w ome n a re g e a re d to wa rd fe ta l be ne fi t, a nd do no t us ua lly ta ke pil l burden issues into consider ation. This is another possible explanation for tre atment failure This lack of c onsideration of ma terna l needs indica tes a fa ilure to consider ba rr ie rs to m a te rn a l lo ng -t e rm su rv iva l.

PAGE 31

20 A subset of REACH provide rs in L os Ang eles pe rfor med a simple surve y of 31 HI V-positive y outh from their c linic to identify fac tors associate d with adher ence (B elzer, F uchs, L uftman, & Tucke r, 1999). B y self-r eport, 22% of the y outh reporte d I forg ot as the r eason f or missing ART, a nother 15% r eporte d depre ssion, and 43% repor ted too many pills, though the a ctual pill burden wa s not reporte d by the authors for a naly sis (Be lzer et al., 1999). At on e sit e the REA CH pr ov ide rs de ve lop e d a nd te ste d a pr og ra m to a dd re ss adher ence problems in adolesc ents ca lled Thera peutic Reg imens Enhancing Adher ence in Teens ( TREAT) Prog ram (Rog ers e t al., 2001). The pr og ram include s educa tion on benef its of compliance, psy cholog ical asse ssment and support to develop r eadine ss for thera py and to addr ess depr ession, and tools and re minders, such a s signa ling wa tches and pag ers ( Roge rs et al., 2001) Roge rs et al. ( 2001) re lease d a desc ription and p re l i m i n ar y re s u l t s o f t h ei r a d o l es ce n t ad h er en ce p ro gra m b as ed o n t h e S t age s o f C h an ge Mo de l di sc us se d b e low Wh ile the or e tic a lly th is p ro g ra m a nd a pp ro a c h s ho uld wo rk to incre ase a dhere nce, it has not be en shown to be e ffe ctive in their c ohort in the short term. Th e ir fi nd ing s in dic a te d le ss t ha n 4 0% of pa rt ic ipa nts we re a dh e re nt t o H AA RT, de sp ite intense interve ntions. Their rec ommendations include avoida nce of the har mful eff ects of pr e ma tur e pr e sc ri pti on of HA AR T, fu rt he r s ug g e sti ng tha t th e de g re e to w hic h y ou th acc ept their diag nosis predicts succ ess on HAART ( Roge rs et al., 2001) Thoug h not dis c us se d b y the a uth or s, thi s r e c omm e nd a tio n w a s n ot e xten de d to pr e g na nt a do le sc e nts be c a us e of se c on da ry pr e ve nti on po lic ie s. I n addition to depression and f org etfulness, F utterman (1999) cites distrust of medications and the medical e stablishment among te ens of c olor (94% of infec ted adolesc ents) as a rea son for a dhere nce pr oblems. Finally in a study y ielding c ontrasting

PAGE 32

21 results, two pharma cists conducted a study of 25 patients, ag es 9-21 y ear s using a pag er as a r eminder f or medica tion doses over a three -month period (Todd & Miller, 2000). Their f indings showe d that compliance improved any wher e fr om 45-50% fr om baseline, using only self-r eport a s a mea sure. The se re ported incr ease s in adhere nce w ere su pp or te d b y se ro log ic imp ro ve me nts inc lud ing low e r v ir a l lo a ds a nd hig he r C D4 c ou nts (Todd & Miller, 2000). I n summary there exis ts a pauc ity of re sear ch addr essing compliance improvement in adolesc ents. No re sear ch foc uses spec ifically on the issues fac ed by adolesc ent mothers. Much of what wa s identified foc used on the a ssessment of re adiness, or psy cholog ical mea surements to dete rmine if a y outh was re ady to start ta kin g me dic a tio ns ba se d o n w he the r t he y we re wi lli ng a nd a ble to c omp ly wi th medication ther apy there by reduc ing se lf harm ove r time in the form of A RT resistance Preg nant HI V-positive teens g et no assessment of re adiness, a s the focus is on the health of the fetus. Nothing addre sses continuation of ther apy when the te ens ar e starte d on AR T p re c ipi tou sly a nd wi tho ut p sy c ho log ic a l pr e pa ra tio n o r a c c e pta nc e of the ir dia g no se s. Whe n it c ome s to ind ivi du a l a dh e re nc e mo st r e se a rc he rs g ro un d th e ir stu die s in health beha vior models. Wil ley et al. (2000) used the Stag es of Chang e Model, or Transthe oretica l Model to assess why patients fail to take me dications as pre scribed. Th e y stu die d 1 61 HI Vinf e c te d p a tie nts us ing se lf -r e po rt of c omp lia nc e a nd e le c tr on ic mon ito ri ng of a dh e re nc e (e le c tr on ic re min de rs a nd do se c ou nte rs ), a nd c omp a re d th e ir exis ting stag es of r eadine ss to comply with their ac tual compliance behavior The as s u m p t i o n i n t h i s s t u d y i s t h at p re s cr i b i n g HA AR T i s t an t am o u n t t o p re s cr i b i n g a lifesty le cha ng e, and should not be done without assessing the rea diness of the individual

PAGE 33

22 to a c c e pt t his pr op os e d c ha ng e F or pr e g na nt g ir ls i n p a rt ic ula r, no su c h p re pa ra tio n is ma de T he a ssu mpt ion is t ha t th e g ir ls m us t c ha ng e fo r t he sa ke of the ir fe tus e s, reg ardle ss of rea diness. This is one expl anation of the failure of mater nal trea tment continuity Their f indings showe d that reminder s and dose c ounters, mainstay s in the adher ence community do not work if par ticipants have not a dvance d to the action and maintenanc e stag es of c hang e. Adh e r e nc e in Ot he r M ar gin ali z e d G r oups Tr os s ( 20 01 ) p ub lis he d th e fi nd ing s o f h e r e thn og ra ph y of inn e rc ity Hi sp a nic women within one zip code in New York City Be cause of the limited g eog raphy and compre hensive tec hniques, she dra ws a vivid picture of the bar riers to sexual power and health neg otiation for these w omen. She found that de spite highrisk lifesty les and kn ow le dg e of HI V, no ne of the wo me n p e rc e ive d th e mse lve s to be a t r isk Sh e a lso do c ume nte d th e a bs e nc e of dis c us sio n o f s e x or H I V b e tw e e n th e se wo me n a nd the ir partne rs. Acc ording to participants, such discussions at times ende d with violence (T ro ss, 20 01 ). Me tho do log y wa s c omp re he ns ive in t his stu dy u til izin g the Com mun ity I dentifica tion Process (CI D) a nd included a r eview of community rec ords, individual int e rv ie ws a nd fo c us g ro up s. Me tho ds inc lud e d id e nti fi c a tio n o f g a te ke e pe rs in thi s instance, the owner of a bode g a and the owner of a la underma t. I n both case s, the community identified selfmade wome n of substance in the community as lea ders. F our dif fe re nt l e ve ls o f p a rt ic ipa nts we re ide nti fi e d: k e y pa rt ic ipa nts s y ste m in te ra c tor s, pr oje c t st a ff a nd the g a te ke e pe rs T his bo dy of wo rk c on tr ibu te s g re a tly to t he sp a rs e bo dy of kn ow le dg e re g a rd ing wo me n w ith HI V b y a pp ly ing the ind uc tiv e a pp ro a c h to H IV p r e v e n t i o n w i t h i n a s m a l l c o m m u n i t y.

PAGE 34

23 Specific inter esting finding s include the c oncept of support g roups as a whitey thing, with a pre fer ence shown to provide support whe re w omen alre ady g othe supermar ket, day car e, schools, and the launder mat. The wome n also expressed the de sire f o r a f e m a l e c o n d o m t h a t c o u l d b e a p p l i e d b e f o r e e n g a g i n g i n f o r e p l a y o r s e x u a l a c t i v i t y, and re moved af ter he r male pa rtner ha s left. This illust rate s the awkw ard na ture of tr y ing to neg otiate safe r sex with int imate others in the c ommunity These f indings support the ide a of so c ia l ma rk e tin g to c ha ng e the a tti tud e s o f a c omm un ity a s in div idu a l c ha ng e is predic ated on c ommunity boundarie s. They also support listening to g roup fe edbac k on the kind of ca re in whic h they would participa te. The He nry J Kaiser Fa mily Founda tion (1999) also soug ht to provide a compre hensive de scription of a community by conducting interviews a nd focus g roups on hig he r r isk te e ns a nd the ir pr ov ide rs T ho ug h th e stu dy lis ts n o s pe c if ic me tho do log y it is e thn og ra ph y -l ike in i ts s c op e T he stu dy inc lud e d g ro up s w ith de mog ra ph ic c ha ra c te ri sti c s r e pr e se nta tiv e of ov e ra ll H I V p re va le nc e s o th e re is n ot a preponde ranc e of A frica n Americ an fe male adole scents. Still, t here are 30 fema les, and about half a re A frica n Americ an. Ag ain, despite re peate d high r isk behavior a nd a cog nitive understanding of wha t constitutes high risk, these adolesc ents do not view themselves as be ing pe rsonally at risk. I t is the only study specific ally addre ssing qualitative content of adolesc ents and HI V. The point is also adva nced tha t teens themselves express a f eeling of hostility and judg ment from the he althcar e sy stem, and state te en-f riendly ca re is a r equire ment for them. One inter esting finding is the conce pt of using c ondoms to protect oneself f rom a sexual partner Tee ns do not view sex partners a s unclea n or dang erous, a nd there fore do not use condoms for protec tion (He nry J Kaiser Fa mily Founda tion, 1999).

PAGE 35

24 Diffe rent messa g es ar e either eff ective or fall on dea f ea rs, depe nding on how the population fee ls they are reg arde d, and the c ontext into which the messag e is rec eived. Stigm a Theory and H IV Many rese arc hers ha ve desc ribed the c onditions ex perie nced by HI V positive people a s being consistent with Sti g ma Theor y (I ng ram & Hutchinson, 1999a). I t has be e n a pp lie d to e xpla in h ow the he a lth of mot he rs is o ve rl oo ke d in the int e re st o f t he ir fetuses. Ta y lor (2001) r eporte d that sick women a re diff ere ntiated from we ll women during preg nancy by the mere diag nosis of HI V infec tion. Bunting (1996) noted tha t the ability to disreg ard the needs of mothers is supported by stigmas that r ender her discounted and ta inted. Fa thers, the pr imary source of infec tion for these mother s, are no ta bly a bs e nt f ro m ve rt ic a l tr a ns mis sio n li te ra tur e a s a c a us e of inf e c tio n f or inf a nts (Pinch, 1994). Similar to feminist theory Stigma The ory define s the community of HI Vpositive women as devia nt, or other (Ta y lor, 2001). The ir societal status plac es them a t r isk fo r e a rl y pr e g na nc y a nd the ir e a rl y pr e g na nc y le a ds to t he ir HI V d ia g no sis th is proce ss of diag nosis during pr eg nancy is ex plored in the litera ture of stig ma theory Tay lor (2001) note s that sick w omen ar e sepa rate d from w ell women dur ing preg nancy by the mere diag nosis of HI V infec tion, reg ardle ss of whethe r or not s ym p t o m s a r e p r e s e n t S h e c o m m e n t s t h a t t h i s e a r n s t h e m t h e l a b e l S e e m i n g l y h e a l t h y, bu t do ome d. She ta ke s c a re to s e pa ra te the pa tie nt s e mic vie wp oin t of fe e lin g we ll until medications (with their unpleasa nt side eff ects) a re star ted, from the me dical profe ssions etic view of having an uncontr olled infec tion until medications ar e starte d. How the c oncepts of me dication, illness, and wellness a re c onvey ed to these y oung minority fema les has a lot to do with the way fema les communicate in ge nera l. A need fo r c on ne c tio n is inh e re nt i n f e ma le s ( Pin c h, 19 94 ). F e ma le s h a ve a ne e d to be lin ke d to

PAGE 36

25 a supportive pe rsonal networ k more so than male s. Knowing including know ing a bout self and H I V-infe ction, is ge nera lly based on pa ssively rec eived knowle dg e or subjec tive knowing for poor women (Pinch, 1994). Fa cts a re de fined to these w omen as wha t po lit ic a l le a de rs a nd he a lth c a re pr ov ide rs say the y a re (P inc h, 19 94 ). Th is m e a ns tha t in the a bs e nc e of e mpo we ri ng the se g ir ls, a dv ic e re nd e re d to the se mot he rs sh ou ld b e wi th the full understa nding of their position and vulnera biliti es. Although T ay lor (2001) discusse s both the Hea lth Belief Model and the rational c ho ic e mod e l, sh e se e ms t o s ug g e st t he ra tio na l c ho ic e mod e l is the do min a nt v ie w i n HI V trea tment communities. This m odel, thoug h not elabora ted in her w riting, implies tha t th e me dic a l c omm un ity a ssu me s th a t pa tie nts wi ll a dh e re to p re sc ri be d tr e a tme nts and pre vention prac tices once educa ted about HI V infec tion, sim ply beca use of medic al a dv ic e I t do e s n ot t a ke int o a c c ou nt a ny of the c on c e pts dis c us se d a s b a rr ie rs to a dh e re nc e in o the r s tud ie s. Ac c or din g to T a y lor (2 00 1) on e pr ima ry ba rr ie r i s th e ne e d to hide the diag nosis because of stigma I f the mothers se ek tre atment, the child may be labeled a s infecte d, whether or not it is actually is. I n this sense, by continuing he r own trea tment, the mother is putting her child at risk of stig ma. Acc ording to Bunting (1996), the dang er the mother re prese nts to society though inf ecting innocent and unbor n babies rationalizes the choice to ignor e her particula r hea lth concer ns. She become s stigmatized, with discrimination by the hea lth care sy stem as a c onsequenc e B unting (1996). Qualitative re sear cher s have done the most work in descr ibing the stig ma of HI V. Poindex ter a nd L insk (1999) conduc ted onetime semi-structure d interviews on 19 older Afric an Amer ican wome n who wer e ca reg ivers to fa mily members with HI V/AI DS. Thoug h the subjects ar e HI V neg ative and olde r, they are members of the Afr ican

PAGE 37

26 Americ an culture and ca n speak to the c ontext of being (or pe rce ived as be ing) HI Vpositive in thi s community This is the only study to look at the attitudes of Afr ican Americ ans towar d eac h other in re g ards to HI V infec tion. The point of the study was interesting ; in that the investiga tors wer e intere sted in the burden the se women be ar a s elder s and ca reg ivers. The ir overa ll findings sug g est that stigma should be addre ssed as part of soc ial work interve ntion. Of particula r intere st in the findings wa s that many of the se wo me n r e ly on c hu rc h f or c omf or t, b ut d o n ot d isc los e to a ny me mbe rs or the ir pastors that they have a n HI V-positive fa mily member. This c omfort see ms to be re g a rd e d a s c on tin g e nt o n s e c re c y a nd vu lne ra ble to t he thr e a t of wi thh old ing c omf or t if the sec ret is known. Thre e par ticipants talked of their own discrimination of HI Vpositive community membersuntil one appe are d in their family One pa rticipant viewed this as punishment fr om God for pa st discrimination. Poindex ter a nd L insk (1999) re view Gof fmans stig ma theory summarizing the thr e e re la te d ty pe s o f s tig ma : a sso c ia tiv e sti g ma (a sc ri be d to tho se a tta c he d to stigmatized personsa lso called c ourtesy stigma) internalized stigma ( acc eptanc e of society s appr aisal of one s re duced w orth), and stig ma manag ement (be ing a war e of r eal or potential re actions of other s). Methodolog y consisted of semi-struc tured que stions such as, Tell me a story about how y ou or y our fa mily members ha ve had ne g ative response s or have experience d discrimination about AI DS. Data collection included field notes, obser vations, and re sear cher comments. Content was then c oded and analy zed, with results revea ling that the ma jority of Afr ican Ame rica n women a c k n o w l e d g e s t i g m a b u t a v o i d i t p e r s o n a l l y b y c a r e f u l l y g u a r d i n g d i s c l o s u r e o f t h e H IV sta tus of the ir fa mil y me mbe rs .

PAGE 38

27 Another study looked spec ifically at the eme rg ent fit of Gof fmans stig ma theory to a g rounded the ory approa ch to HI V positive mothers. I ng ram a nd Hutchinson (1999a) pu bli sh e d th e ir dis c ov e ry of a n e me rg e nt f it o f e xtan t th e or y to t he ir fi nd ing s. I n th e ir g rounded the ory approa ch to the mothering experience for HI V-positive women, theme s re la te d to sti g ma ke pt a ri sin g T hr ou g h a re vie w o f e xistin g lit e ra tur e on thi s phenomenon, Gof fmans the ory seeme d appropr iate. The details of the g rounded the ory study are listed separa tely below, but the themes tha t kept arising rela ted to stigma a re privac y passing and cove ring in clinic. Stigma is an una nticipated diffe renc e that dis c ou nts oth e r a ttr ibu te s o f a n in div idu a l a nd c a us e s th e re je c tio n o f o the rs (I ng ra m & Hu tc hin so n, 19 99 a ). I n r e la tio n to pa re nti ng mo the rs a lso re po rt e d th a t th is s tig ma sprea d to their children, w hether or not they wer e infe cted. Quote s such as, No one wo uld tou c h my c hil dr e n we re c omm on F ind ing s in c lud e d le a vin g do ub le liv e s behind a fac ade of normal mothering . The c oncept of passing perta ins to something that is not appare nt being conce aled. The experience of appe aring normal, with all of its benef its, is m itigate d by the constant thre at of e x posure a nd losing the be nefits. The w omen g oing to c hurch f or the bene fit of support without expressing why they neede d the support for f ear of losing the su pp or t is a g oo d e xamp le T he ne e d to pa ss i s e vid e nc e d b y the un wi lli ng ne ss t o have A Z T in their homes bec ause of its name rec og nition (I ng ram & Hutchinson, 1999a). Covering is the act of ma nag ing soc ial tension and distrac ting a ttention away from the stigmatizing qua lity (I ng ram & Hutchinson, 1999a). E x amples fr om the study include ly ing a bout rea sons for medica tion (i.e., bac k pain, ca ncer ) and r efe rring to visit s to the doctor a s shopping trips. I n another a woman strug g ling with we ight loss from he r me dic a tio ns tol d h e r c owo rk e rs sh e wa s o n a str ic t w e ig ht l os s p ro g ra m. T he ir re sp on se

PAGE 39

28 wa s th a t, I wo uld su re lov e to l os e we ig ht t he wa y y ou did . She sta te s sh e tho ug ht, No y ou mos t de fi nit e ly wo uld no t. Th e wo me n b e lie ve tha t li e s a re jus tif ie d a s a means for survival (I ng ram & Hutchinson, 1999a). P sy c hol og ic al D ist r e ss a nd HI V Be cause of the we ll-documented link betwe en depr ession and a suppr essed imm un e sy ste m, P inc h ( 19 94 ) t he or ize s sy ste ma tic a lly op pr e sse d a do le sc e nt f e ma le s to be a t a hig he r r isk fo r r a pid HI V d ise a se pr og re ssi on L oo kin g a t th e e ff e c ts o f s tr e ss be c ome s im po rt a nt t o th is i ssu e Sm ith e t a l. ( 20 01 ) u se the Str e ss P ro c e ss M od e l to explain psy chosocia l functioning in Africa n Americ an re cent mother s. The domains of this model include stressors, coping resour ces, c lose rela tionships, coping re sponses, and psy cholog ical distress. The y found that thoug h adher ence and tre atment literature a ssu me s b oth c omp re he ns ion of a nd fo c us on the dia g no sis of HI V, ma ny of the ir subjects ra ted money as a g rea ter c oncer n than HI V infec tion. This appear s important, as it addresse s the conc ept of immediac y in these women. H I V infec tion is an arbitrar y c on c e pt w ith no imm e dia te ma nif e sta tio n ( sy mpt oms ); wh e re a s p ov e rt y ha s d a ily implications for housing, f ood, clothing, a nd transporta tion. Sm ith et al. (2001) conclude d this special population needs inter ventions to improve their social a nd e c on omi c c on dit ion s, wi th i nte rv e nti on s to a dd re ss t he ir HI V i nf e c tio n ta ilo re d to the ir str on g e st m e a su re d r e so ur c e s. Mu rp hy e t a l. ( 20 01 ) a g re e d w ith thi s a sso c ia tio n b e tw e e n p sy c ho log ic a l di str e ss a nd he a lth ri sk be ha vio rs a nd so ug ht t o a pp ly thi s th e or e tic a l pe rs pe c tiv e to a do le sc e nts with HI V. They utiliz ed Jessors Problem Be havior The ory as their theor etical fra mework. This theor y focuse s on three ma jor conc epts: behavior, pe rsonality and environment (Mur phy et al., 2001). The ir study sug g ests that highe r leve ls of depre ssion

PAGE 40

29 pr e dic t in c re a se d r isk ta kin g be ha vio rs d e sp ite wh a t th e y de sc ri be a s y ou tha pp ro pr ia te ri sk re du c tio n c ou ns e lin g A nxie ty re la te d to he a lth sta tus i. e H I V i nf e c tio n, a c tua lly incre ased the maladaptive c oping be haviors of substanc e abuse and unprote cted se x Th is c ou ld l ike ly be stu die d in dir e c t r e la tio ns hip to m e dic a tio n a dh e re nc e wi th t he sa me results. This finding sug g ests that medical inter ventions in the absenc e of a ddressing de pr e ssi on a nd e nv ir on me nt a re lik e ly to f a il. Mellins, Ehrhardt, Rapkin, and H avens ( 2000) appe ar to a g ree with the above models in their study of psy chosocia l fac tors and ada ptation in HI V-infe cted mother s. Thoug h no expli cit fra mework is cited, the conce pts of environment, stre ss, protective fac tors, and depr ession are listed. They found that mothers with HI V have additional str e sso rs of ra c ism is ola tio n, po ve rt y d isc ri min a tio n, a nd sin g le mot he rh oo d. Th e se make them more disposed to diag noses of de pression and post-tra umatic stress disorder They also found that re lationship breakup, a ssault, abuse, a nd remova l of children pla y ed a major r ole in causing these disorde rs. I ronica lly some provider s offe r the fe ar of remova l of exposed infants as the primar y rea son mothers adhe re to tre atment protocols. T h i s i s p ro ff er ed as an ex p l an at i o n b y s o m e o f t h e s t af f i n t h e c l i n i c t o b e s t u d i ed t h o u gh no leg al pre cede nt for the r emoval of e x posed babie s is established in Florida. M ino r it ie s a nd HI V bell hooks (2003) de scribes a n ear ly movement within the Afric an-A merica n culture towa rd incre ased se lf-este em and se lf-love, r eg ardle ss of external eve nts. She descr ibes a re volution where suc cess wa s defined a s the ability to value and r ever e on e se lf de sp ite the pr e se nc e of de hu ma niza tio n. Th is m ov e me nt, la rg e ly le d b y ma le s, a lso de ma nd e d th a t me n r e g a in t he ir his tor ic a l po sit ion of be ing re ve re d b y the ir women. B y definition, this would make the subjug ation of women ne cessa ry for the

PAGE 41

30 a dv a nc e me nt o f m e n. No t on ly is i t un do ub te dly fr us tr a tin g fo r m e n to be e xpe c te d to a pp e a r i nta c t a nd inv inc ibl e in t he pr e se nc e of on g oin g su bju g a tio n, bu t in a pe op le str ug g lin g fo r p ow e r, do min a tio n o f w ome n is the on ly a re a in w hic h th is i s e a sil y achie vable. This lea ves blac k women in the ter rible plac e of be ing le ss reg arde d even within the culture that should provide ha rbor f rom discrimination. Wil kinson (1999) states that in plac es with more income inequality more de aths can be attributed to violence Nondisclosure of status before unprotec ted sex, and the re fu sa l to we a r c on do ms a re bo th a c ts o f v iol e nc e T ho ug h th e a ttr ibu ta ble c a us e of de a th may be pneumoc y stis pneumonia seconda ry to AI DS, the primary cause of dea th is an act o f v i o l en ce W i l k i n s o n (1 9 9 9 ) f u rt h er el ab o ra t es wi t h re fe re n ce t o t h e s h am e r age spiral, whe re in e ffe ct, hostilit y mounts in t he pre sence of unac knowledg ed shame Many black men a re infe cted dur ing inc arc era tion, and this is a reasona ble explanation for blac k men fe eling entitled to not disclose their HI V status to their fema le par tners upon rele ase f rom prison. This, combined with the expectation of mac hismo, ex plains why impreg nating a woman ( sign of virility ) is the bra g g ing r ight of many men, without re g a rd fo r t he c ur re nt o r f utu re he a lth of the wo ma n o r t he fe tus F a rm e r ( 20 03 ) m a ke s a simil ar a rg ument by stating tha t people subordinate d by their social supe riors and thr e a te ne d w ith hu mil ia tio n, a tte mpt to r e g a in t he ir se ns e of c on tr ol b y a sse rt ing a uth or ity a nd c on tr ol o ve r t ho se be low the m. T he po we r i mba la nc e be tw e e n me n a nd wo me n in minority communities is dis cussed e x tensively in the literature Be atty Wheeler, a nd Gaitner (2004) c onducted a revie w of HI V pre vention literature to look for inclusion of Afr ican Ame rica n subjects with the intent of making re c omm e nd a tio ns fo r t he de ve lop me nt o f m or e e ff e c tiv e pr e ve nti on str a te g ie s f or thi s g roup. They repor ted a lac k of cultura lly based the ory to guide the exist ing r esea rch a nd

PAGE 42

31 at the same time sugg ested that blac k Americ ans ar e too larg e and dive rse a g roup to be place d within one cultural de finition. They found that re sear ch aimed a t behaviora l lifesty le cha ng e wa s over r epre sented in the litera ture, with the He alth Be liefs Model, the Theory of Reasone d Action, and the T ranstheor etical Mode l of Be havior Chang e being m o s t c o m m o n l y u s e d It w a s r e p o r t e d t h a t o f t e n e v e n i f i n c l u d e d a s s u b j e c t s i n a s t u d y, minorities were not later broke n out and ana ly zed as a sepa rate g roup. They sug g ested that models sensitive to differe nt gr oups of minorities be deve loped and that not all ef fort should be extended to individual chang e, as individuals do not alway s have the control nece ssary to initi ate c hang es. Women rar ely have c ontrol over the e lements of survival sex, and childhood sex ual abuse and unwa nted adult sexual activity are both reporte d as risk fac tors for r isky sexual behavior, and f or substance abuse tha t leads to risky sexual behavior (Carg ill, S tone, & Robinson, 2004). The idea that any of this behavior is fully volitional is a stretc h. L y nch, Ka plan, a nd Sa lon e n ( 19 97 ) q ue sti on the po pu la r t he or ie s n ow tha t a ttr ibu te he a lth be ha vio rs to the choice s that individuals make, beca use they make he alth behavior both individual and volitional; neither of which a re usua lly the ca se for women at r isk for HI V. The the ories pr e dic t th a t if g ro up s a t hi g h r isk fo r H I V i nf e c tio n a re sim ply tol d w ha t be ha vio rs to a vo id a nd wh y th e y wi ll s top the be ha vio rs T he ir fi nd ing s sh ow e d th a t w hil e the re is c or re la tio n b e tw e e n c ur re nt S ES a nd he a lth be ha vio rs in a du lth oo d, the re a lso e xists corr elation betwe en hea lth behaviors in adulthood and SES at specif ic tempora l mil e sto ne s th ro ug ho ut c hil dh oo d. I n o the r w or ds b e ha vio rs a re in s ome wa y lin ke d to pa re nta l SE Sa n in fl ue nc e on be ha vio r t ha t ne ith e r a n in div idu a l no r a pr ov ide r c ou ld g o back in the pa st to chang e (L y nch et a l., 1997). This is a strong arg ument for inc re a se d e c on omi c pa ri ty in s oc ie ty th ou g h it wo uld be un a ttr a c tiv e to m os t pu bli c

PAGE 43

32 health provide rs and planne rs spec ifically beca use it lacks ag ency even on the part of prog rams. Prog rams ac t in the present a nd occa sionally for the f uture; neve r in the past. The e ffe cts of SES parity now, eve n if attained, w ould not be measura ble for a lifespan. Since prog ram e valuations must measure c hang es ea ch funding cy cle, this would neve r happen. I n their re cent study Carg ill, S tone, and Robinson (2004) f ound evidenc e that minorities are le ss likely to rec eive a ntiretrovira l medication thera py (ART) e ven if the g uidelines indicate it as the sta ndard of car e. Of ten this is attributed to the providers se ns e tha t th e pa tie nt w ill be un a ble to a dh e re to c a re B a su (2 00 4) ob se rv e d th a t if AR T we re re se rv e d o nly fo r t ho se wh o s e e m mo st l ike ly to a dh e re to r e g ime ns th e n o nly tho se least likely to be in need of ARTs would rec eive them. This is not t he only rec ent work to support the ide a that the be haviorist theories are shortsighte d. Ay nalem, Mendoza, F rede rick, a nd Mascola ( 2004) found in their study of pre g nant women who r efuse d HI V-testing repor ted that 4% of w omen had to see k their husband s permission prior to consenting to the test. This clear ly neg ates pe rsonal ag ency Many others fe are d stigma, discr imination, dis closure, violenc e, or ha d a lac k of trust in the provider to pre vent any of these. T he author s fell short by sug g esting that throug h additional educa tion and time to trust the providers, that many of these w omen would chang e their minds. This seems nave in that it addresses none of the struc tural or domestic problems that we re ide ntified as driving the re fusal in the first plac e. An o t h er co m m en t ar y o n t h e b eh av i o r t h eo ri es i s t h e f i n d i n g o f W h y t e, S t an d i n g, and Madig an (2004) that there was a ctually a positive corr elation betwe en HI V-re lated knowledg e and hig h-risk sexual behaviors in Af rica n Americ an women in the southe ast. I nc ome a g e a t f ir st s e xua l e nc ou nte r, nu mbe r o f p a rt ne rs a nd e du c a tio n le ve l w e re a ll

PAGE 44

33 analy zed, with only educa tion level having a sig nificant impac t on behavior. This shows that the population-spec ific messag es ar e re aching the community ; they are just not having an impact on be havior. Ebr ahim, Ander son, Weidle, and Purcell (2004) analy zed d at a f ro m t h e 2 0 0 1 Be h av i o ra l R i s k S u rv ei l l an ce S u rv ey T h ey fo u n d t h at t h o u gh Afric an Amer icans ha d a hig her r ate of having ever been te sted for H I V, they had a lowe r level of know ledg e about a vailable tre atment. Over all, even those who had be en tested wer e not awa re tha t treatment for infec tion now exi sts. This shows the durability of the ta rg e te d me ssa g e s to a vo id c on tr a c tin g thi s f a ta l di se a se a nd a bo ut i t be ing a de a th sentenc e. The dang er of targ eting these ty pes of messa g es at a community alre ady devoid of hope for the f uture is that the messag es abla te any hope for a future once a person is diag nosed with HI V. P ositive Adaptation Despite the g rim outlook, some women, devoid of many identified bar riers, do exceptionally well with their HI V diag noses. Speig el and Schr imshaw (2001) c onducted a ser ies of two semi-struc tured intervie ws that they then subjecte d to thematic ana ly sis. Th e su bje c ts w e re HI Vpo sit ive wo me n w ho c la ime d to ha ve ha d p os iti ve c ha ng e s a s a result of their dia g noses. Exclusion criteria included intra venous drug use, and the aver ag e ag e wa s 36 y ear s, with a sample size of 54. Onethird was Af rica n-Amer ican. T h e m i x o f s u b j ec t s i s n o t re fl ec t i v e o f H I Vp o s i t i v e w o m en i n gen er al s o t h e f i n d i n gs a re no t g e ne ra liza ble T he se wo me n, un lik e a ny oth e r d e sc ri be d in lit e ra tur e or se e n in pe rs on a l e xpe ri e nc e d e sc ri be HI V i nf e c tio n a s a c a ta ly st t o r e so lvi ng the ir re la tio ns hip conflicts, a que ue to enjoy life, and a step to becoming strong advoca tes for the mselves. Whil e ther e may be some mer it to the idea adva nced tha t perc eptions of positive outcomes may imply positive outcomescer tainly supported in psy choneur o-

PAGE 45

34 immunology literature it is not consistent with the finding s closer to the popula tion of int e re st. Dunbar Meuller, Me dina, and Wolf (1998) c onducted similar semistructure d interviews in the hope s of supporting their pre viously cre ated model r elating to positi ve adapta tion. They conducte d interviews with 34 women in a study that also appe ars to be nong ener alizable beca use of the g eog raphic demog raphic s of their c onvenienc e sample. Un lik e HI Vinf e c te d w ome n, the ir pa rt ic ipa nts we re a lmo st h a lf Pa c if ic -I sla nd e rs ; colleg e educ ated, a nd had an a vera g e ag e of 36. The y found five c omponents that supported their mode l of ada ptation: reckoning with death, life a ffirma tion, crea tion of meaning self-a ffirma tion, and rede fining rela tionships. All of these pr ocesse s they descr ibed re quire re flexivi ty and a le vel of a cce ptance of diag nosis that far e x cee d that documented in the ta rg et population of this paper Howeve r, the c oncept of un de rs ta nd ing pe rs on a l g ro wt h, a s a pr e c e de nt t o e ff e c tiv e c op ing is c on sis te nt w ith previously explored stress models. Ne ith e r o f t he se stu die s p ub lis he d c le a r d e sc ri pti on s o f t he ir da ta a na ly sis B oth of these studies sug g est a population that has a djusted to their diag nosis, presents for car e, ac ts in their own best interest, a nd is theref ore la rg ely incong ruent with the population of interest. I dentify ing the barr iers fa ced by minority HI V-positive adolesc ent mothers fr om this di ffe rently adjusted population may be of some va lue. Studies addressing barr iers to adhe renc e provide some explanation as to the contextual dif fe re nc e s e xpe ri e nc e d b y the se g ro up s. Mother ing b y HIV-Inf ect ed Wom en I ng ra m a nd Hu tc hin so n ( 19 99 b) de sc ri be d th e ir g ro un de d th e or y stu dy in another publication in gr eate r deta il. The purpose w as to descr ibe the mothering

PAGE 46

35 experience of HI V-positive women. The g oal was to unde rstand ac tions in t he conte x t of their belief s. A sample of 18 H I V-positive mothers ag ed 18-44, ha lf of whom wa s minority was subjec ts of extensive interviews. Using a g rounded the ory technique, a basic socia l psy cholog ical proc ess (B SPP ) of de fensive mother ing w as discover ed. The obvious connec tion to st igma is describe d in one woman s comment, Eve ry thing y ou do in y our life y ou have to wor ry about if someone is g oing to f ind out. I t is hard. You have to watch e very thing y ou say and do. T hey conclude d that stigma pr ovides the conte x t of HI V-mothering and sets the stag e for defe nsive mothering (I ng ram & Hutchinson, 1999b). Mothering consists primarily of wa y s to protect the c hild from both herself and society s pre judices. This rela tionship, t hat see ms to ex plain a mother s pref ere nce f or her c hilds welfa re ove r her own, is further explored in another less r igor ous qualitative study in the literature Andre ws, Wil liams, and Neil (1993) conducte d a ser ies of two qualitative int e rv ie ws : th e No rb e c k So c ia l Su pp or t sc re e n a nd a su bs ta nc e a bu se sc re e n to a sa mpl e of 80 HI Vinf e c te d mo the rs A we a kn e ss i n th is s tud y is t ha t no de sc ri pti on of da ta ev al u at i o n wa s i n cl u d ed No s i gn i fi ca n t t h em es we re i d en t i fi ed an d n o m et h o d o l o gy oth e r t ha n int e rv ie ws wa s g ive n. Qu ote s w e re dis c us se d g e ne ra lly b ut n ot t ru ly analy zed. The two tested sc ree ns that could be re ported on a n ordinal basis we re more fully descr ibed. When evalua ting soc ial support, it was impressive to re veal how ma ny of these mother s relied on y oung children f or emotional support (Andr ews e t al., 1993). I n r e g a rd s to the me dic a l c omm un ity mo st w ome n e xpre sse d d ist ru st. On e wo ma n, in commenting on car e off ere d to her HI V-positive son, said The y want to put him on DD I T he AZ T i s ma kin g him too a ne mic I m g oin g to t ry the DD I do se the y wa nt h im on first, befor e they g ive it to him. M y thinking is that he a nd I are almost the same, y ou

PAGE 47

36 k n o w w i t h t h e s a m e g e n e s s o I l l k n o w w h a t s i d e e f f e c t s h e l l g e t f r o m t h e D D I (I ng ram & Hutchinson, 1999a). T his statement shows a total lack of understanding of the rela tionship between body surfa ce a rea and side e ffe cts, but more ala rming is that her views ar e built on the foundation of mistrust. The re sear ch re viewed de scribes the women who a re likely to fail trea tment, or descr ibes the society at larg e that stig matizes and alienate s women fr om car ing f or the mse lve s. Am ple de sc ri pti ve da ta we re ide nti fi e d th a t de sc ri be d w ome n w ho fa il trea tment, but this data doe s not differ f rom the desc riptions of women most likely to be infec ted with HI V. A small but important section of the litera ture tried to de termine how women experienc ed living w ith HI V, and a llusions were made to perc eived judg ment and h o s ti li ty in th e ir w o r ld s a t l a r g e in c lu d in g in th e H I V tr e a tm e n t c o mm u n it y T h e n e xt logic al step is to ga ther qua ntitative data to desc ribe whe ther or not this perc eived differ ence in treatment is associa ted with differ ence s in outcomes, beca use if it is, then modify ing the trea tment environment will be funda mental to improving ma terna l ou tc ome s.

PAGE 48

37 CH APT ER 3 METHODS P ur pos e Th e pu rp os e of thi s st ud y wa s to de te rm ine if pr e vio us ly ide nti fi e d b a rr ie rs to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equally ; and to deter mine if the propose d constructs of struc tural fac ilitators or structural ba rrier s are associate d with HI V trea tment adher ence in the pe ri na ta l dy a d, the po stp a rt um m oth e r, or the e xpos e d in fa nt. Design This was a de scriptive and a naly tic epidemiolog ic ca se compa rison study The g oals wer e both to descr ibe fre quencie s of exposures and outcome s in the three g roups and to ana ly ze which independe nt variable s best predic ted outcomes. All data collecte d we re c oll e c te d th ro ug h r e tr os pe c tiv e c ha rt re vie ws D a ta we re a na ly zed thr ou g h b oth descr iptive statisti cs and log istic reg ression ana ly sis. A convenienc e sample w as obtained thr ou g h s y ste ma tic re vie w o f t he c ha rt s o f H I Vpo sit ive pr e g na nt w ome n in the hig h r isk obstetrical c linic (see F igur e 3-1) The study was c onstructed of 100 materna l case s and 200 compar isons. The 100 materna l case s consisted of postpartum HI V-positive women with the g roup name postpartum mothers. The 200 c omparisons wer e compr ised of two g roups. One hundr ed of the compar isons comprised the g roup dy ad a nd consisted of the same mothers as the c ase g roup during a diffe rent time interva l. For this g roup, da ta we re c oll e c te d d ur ing the pr e na ta l pe ri od ins te a d o f t he po stn a ta l pe ri od T he la st

PAGE 49

38 100 comparison subjects were in the infant group. This group was comprised of the babies born to the dyad group. As such, these infants shared maternal risk factors at the same point in time as the maternal group. The subjects were all patients of the same HIV subspecialty clinic and its affiliated high risk obstetrical clinic. Table 3-1 depicts the distribution of the independent variables for the maternal sample. Figure 3-1.Subject selection Setting This study was conducted at an urban outpatient HIV clinic and its affiliated high risk obstetrical clinic in the southeastern United States. Data were collected through chart extraction without contact with subjects. The clinics are located in an urban

PAGE 50

39 me dic a l c omp le x an d a re sta ff e d b y nu rs e s, ph y sic ia ns s tud e nts a nd re sid e nts Pa tie nts in the high r isk obstetrical clinic re ceive specia lty prena tal car e for a var iety of conditions including H I V seropositivity diabetes, a nd epilepsy The HI V clinic is locate d in the same c omplex and is part of the same he alth sy stem. I t is a fede rally funded ( Ry an Whit e) a nd state funde d (Childrens Medic al Service s/Medicaid) H I V sub-spec ialty car e c lin ic Ca re is p ro vid e d in ta nd e m by nu rs e pr a c tit ion e rs a nd inf e c tio us dis e a se ph y sic ia ns on a n o utp a tie nt b a sis H I V t re a tme nt, ps y c ho log ic a l su pp or t, d ie ta ry su pp or t, social work suppor t, and primary car e ar e all off ere d on site, unless specific ally excluded by ins ur a nc e or fu nd ing re g ula tio ns Sub jec ts Sam ple Selection Convenience sampling w as used to sele ct the sample f rom elig ible dy ads. B ased on a pr e dic tio n th a t 30 % o f t he su bje c ts w e re in a dh e re nc e to H I V c a re a pp ro xima te ly 229 subjects wer e re quired g iven a 0.05 leve l of signif icanc e, the 10 pr edictors, a nd 90% po we r o f t he te st. Inc lusi on C r it e r ia S u b j ec t s we re HIV ( +) fe m al es wh o s e p re n at al m ed i ca l ca re wa s p ro v i d ed b y t h e h i gh risk obstetrical c linic at the study site from 2000 to 2004. The mother a nd infant both survived throug h 18 months postpartum. The infa nt remaine d in the primary car e of the biologica l mother. Exc lusi on C r it e r ia Neona tal, infant, or mate rnal subjec ts who spent g rea ter than 2 w eeks inpa tient during the 18-month observa tion period. Neona tes and infa nts who spent g rea ter than 2 w eeks outside of the primary car e of their biolog ical mothers dur ing the observa tion period. The mother or the baby died bef ore the end of the observa tion period.

PAGE 51

40 Da ta Ana lys is The Statistical Analy sis Sy stem (SAS) (Ver sion 9.1) was used f or all statistical analy ses and f or writing the scientific r eport of the quantitative data. D escr iptive statistics were use d to obtain the summary measure s for a ll data including a desc ription of the sa mpl e c ha ra c te ri sti c s. De sc ri pti ve sta tis tic s in c lud e d me a ns me dia ns mo de s, rang es, interqua rtile rang es, and standa rd devia tions for continuous varia bles. Categ orica l v ar i ab l es we re s t at i s t i ca l l y re p re s en t ed i n fr eq u en cy d i s t ri b u t i o n s p er ce n t age distributions and g raphic al illustrations. A p-value of less than 0.05 was c onsidered sta tis tic a lly sig nif ic a nt. To a dd re ss t he ma jor hy po the sis of the stu dy s te pw ise log ist ic re g re ssi on wi th fo rw a rd e lim ina tio n te c hn iqu e s w e re us e d to bo th c on tr ol f or c on fo un de rs a nd e sti ma te ind e pe nd e nt r e la tio ns hip s b e tw e e n th e pr e dic tiv e va ri a ble s a nd the ou tc ome L og ist ic reg ression ana ly sis was also used to explore potential diffe renc es in predic tor varia bles be tw e e n w ho pa rt ic ipa te d in re c omm e nd e d H I V c a re fr om t ho se wh o d id n ot. Al l predic tor varia bles wer e included tha t had either at least a marg inal bivariate association with the outcome var iables or f or which the re w as some ra tionale that the var iable may ha ve be e n a c on fo un de r o r e ff e c t mo dif ie r f or oth e r v a ri a ble s. Th is i nit ia l mo de l a lso inc lud e d h y po the size d in te ra c tio n te rm s f or wh ic h s tr a tif ic a tio n a na ly se s su g g e st po te nti a l in te ra c tiv e e ff e c ts. The point and interva l estimates of the odd ra tios of the cate g orica l predictor varia bles wer e re ported. The hy pothesis of homog eneity was a ddresse d by running fr e qu e nc ie s. I n mo st c a se s, Chi -s qu a re sta tis tic s w e re pr e se nte d. I n c a se s w he re a ny c e ll in t he 2x2 ta ble or 3x2 ta ble wa s v a lue d a t le ss t ha n 5 th e F ish e r s E xac t T e st f or sma ll samples re sults was instead used for incre ased va lidity

PAGE 52

41 P roce dure F oll ow ing g ra du a te c omm itt e e a pp ro va l, p a pe rw or k w a s p ro c e sse d to g a in a pp ro va l of the Un ive rs ity of F lor ida I ns tit uti on a l Re vie w B oa rd (I RB -0 1) I nc lud e d in the pac ket was a letter of suppor t from the administration of the study site. Ex pedited pr oc e ssi ng wa s r e qu e ste d a nd a pp ro ve d, a s su bje c ts w e re fu lly de -i de nti fi e d f or the ir protec tion (in compliance w ith the Health I nformation Portability and Ac countability Act of 1996). This option was c hosen bec ause the identify ing da ta omitted was not essential to study outcomes. Subjects wer e dee med vulnera ble bec ause da ta on mental illness and su bs ta nc e a bu se we re c oll e c te d f ro m pr e g na nt, HI Vpo sit ive a nd po te nti a lly pe dia tr ic su bje c ts. Af te r c le a ri ng the Un ive rs ity of F lor ida I RB -0 1, a pp ro va l of the stu dy sit e I RB wa s a lso re qu e ste d a nd ob ta ine d u nd e r t he sa me c on dit ion s. Data Collec tion Upo n ap pro val of t he U ni vers it y of Fl ori da I RB-0 1 an d s tu dy si te I RB, de -i de nti fi e d d a ta we re c oll e c te d th ro ug h c ha rt e xtra c tio n o ns ite a t th e c lin ic T he da ta collection for m was construc ted using Snap softwar e. The principal investig ator persona lly conducte d or over saw da ta collec tion at all times. Each subjec t was assig ned a unique numer ic identifier not tra cea ble bac k to the subject at a ny time and not rec orded a ny pla c e tha t c on ta ins pr ote c te d h e a lth inf or ma tio n ( PHI ). As the sit e is a fe de ra lly funded H I V clinic, super confide ntial require ments were alre ady in place onsite. O ffsite, de-ide ntified data w ere transporte d via passwor d protec ted laptop computer and transfe rre d to the secur e ser ver of the Univer sity of F lorida Colleg e of N ursing A bac kup copy of the de -identified da ta re mains in the possession of the Principal I nvestig ator a n d i s s t o r e d s e c u r e l y.

PAGE 53

42 P ot e nt ial He alt h Ri sks There wer e no phy sical or psy cholog ical risks to subjects of this study There was no interve ntion and no contac t at any point with subj ects. All data collecte d wer e d e i d e n t i f i e d f r o m t h e o u t s e t e l i m i n a t i n g t h e r i s k a s s o c i a t e d w i t h d i s c l o s u r e o f H IV sta tus P otential Health Benefits There wer e no dire ct hea lth benefits from pa rticipation in this st udy as there was no contac t with subjects and no interve ntion. There w ere potential benef its to society as knowledg e wa s g ained to explain rea sons HI V-positive mothers ar e lost to follow up car e at a dispropor tionate ra te. Considering the potential impact pre mature pa renta l death has on children, f amilies, and society the bene fit of dec rea sing pa renta l morbidity and mor ta lit y ou tw e ig hs the ri sk to s ub je c ts. P ot e nt ial F ina nc ial Ri sk Th e re wa s n o c os t to su bje c ts, no inf or ma tio n th a t w ou ld t ra c e ba c k to su bje c ts, and no conta ct with subjects. Ther efor e, ther e wa s no potential financia l risk. P otential F inan cial Benefits There was no re imbursement or c ompensation of any kind to subjects. There fore the re wa s n o p ote nti a l f ina nc ia l be ne fi t to su bje c ts. Co nflic ts of Int e r e st The princ ipal investiga tor and subinvestig ators did not have a ny conflict of int e re st r e g a rd ing thi s p ro toc ol. No be ne fi t, b e y on d p ro fe ssi on a l a nd a c a de mic de ve lop me nt w a s d e ri ve d f ro m pa rt ic ipa tio n in thi s p ro je c t.

PAGE 54

43 CH APT ER 4 RESUL TS P ur pos e Th e pu rp os e of thi s st ud y wa s to de te rm ine if pr e vio us ly ide nti fi e d b a rr ie rs to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equally ; and to deter mine if the propose d constructs of struc tural fac ilitators or structural ba rrier s are associate d with HI V trea tment adher ence in the pe ri na ta l dy a d, the po stp a rt um m oth e r, or the e xpos e d in fa nt. De sc r ipt ion of Sa m ple The c onvenienc e sample in this case compar ison study was c onstructed of 100 materna l case s and 200 compa risons. The 100 mater nal ca ses consisted of postpar tum HI V-positive women with the g roup name postpartum mothers. The 200 c omparisons wer e compr ised of two g roups. One hundr ed of the compar isons comprised the g roup dy a d a nd c on sis te d o f t he sa me mot he rs a s th e c a se g ro up du ri ng a dif fe re nt t ime interval. F or this gr oup, data we re c ollected dur ing the prena tal period instead of the po stn a ta l pe ri od T he la st 1 00 c omp a ri so n s ub je c ts w e re in t he inf a nt g ro up T his g roup wa s comprised of the babies born to the dy ad g roup. As such, the se infants shar ed materna l risk fac tors at the same point in ti me as the ma terna l gr oup. The subjec ts were all patients of the sa me HI V subspecia lty clinic and its aff iliated high r isk obstetrical clinic. Table s 4-1 and 42 depict the distribution of the indepe ndent var iables for the materna l sample.

PAGE 55

44 Table 41. Number a nd perc ent distribution of maternal sa mple by ag e and r ace Ag e 14-18 19-21 22-24 >24 Total (%) 4 13 21 62 100 Race AA Carribe an I sland Hispanic Cauca sian Other Total 40 21 16 12 11 100 Table 42. Number a nd perc ent distribution of maternal sa mple by dichotomous varia bles Var iable Yes No Total (%) Mental illness 23 77 100 New dia g nosis 62 38 100 Poverty 67 33 100 HS/GED 61 39 100 Drug s or alcohol 4 96 100 Pill burden 0 100 100 Variables Th e ind e pe nd e nt v a ri a ble s a re ma te rn a l de pr e ssi on or oth e r m e nta l il lne ss, proxi mity (new ness) of H I V diag nosis, race income, adole scenc e (a g e), hig h school educa tion, substance abuse pill burden, structura l barrie rs to acc ess, and struc tural fac ilitators to acce ss. The depe ndent var iables ar e consistent, but def ined diffe rently for e a c h g ro up be c a us e the sta nd a rd of c a re is d if fe re nt d ur ing e a c h p e ri od T he se ou tc ome measure s include labora tory values, pe rce ntag e of missed medic ation doses, and de lay ed or missed appointments for c are Hyp ot he sis The following two-tailed hy pothesis was tested: Mate rnal de pression or other mental illness, prox imity of HI V diag nosis, race income, adole scenc e, hig h school

PAGE 56

45 educa tion, substance abuse pill burden, structura l barrie rs to acc ess, and struc tural fac ilitators to care cause a sig nificant diff ere nce in a dhere nce a s measure d by the la bo ra tor y re su lts mi sse d me dic a tio n d os e s, a nd de la y e d o r m iss e d c lin ic a pp oin tme nts of HI Vpo sit ive pr e g na nt w ome n, po stp a rt um m oth e rs o r t he ir inf a nts I n o rd e r t o te st the hy pothesis, the following four r esea rch que stions were a sked and a nswere d. Re se ar c h Que st ion One Ar e the thr e e g ro up s ( po stp a rt um m oth e rs in fa nts a nd dy a ds ) h omo g e ne ou s w ith respe ct to the three outcome mea sures (la boratory results, medica tion adhere nce, a nd appointment adher ence )? An ana ly sis of fre quency was used to e valuate this question. Chi-Square statistics and p-va lues wer e computed by cross-c lassify ing the varia bles and g roups with eac h outcome var iable. The re sults presented in Ta ble 4-3 indica te that the thre e g roups wer e not homoge neous with respe ct to labora tory results (Chi-Square =98.82, p=0.0001) L abora tory results wer e sig nificantly less likely to be in the ac cepta ble ra ng e for postpartum mothers (47%) than for inf ants (96%) or dy ads (96% ). The r esult indicated that the three g roups wer e not homog eneous w ith respec t to ART medication adher ence (Chi-Square= 74.12, p=0.0001). Me dication adhe renc e wa s signific antly less likely to be in the acc eptable r ang e for postpartum mothers (36%) than for inf ants (90%) or dy ads (92%) The thre e g roups wer e also not homog eneous w ith respec t to appointment a dh e re nc e (C hiSqu a re = 57 .1 6, p= 0. 00 01 ). Ap po int me nt a dh e re nc e wa s si g nif ic a ntl y le ss likely to be in the ac cepta ble ra ng e for postpartum mothers (52%) than for ba bies (80%) or dy ads (97% ). Table 4-3 depic ts the distributi on of outcomes in ea ch g roup.

PAGE 57

46 Table 43. Distribution of cases a nd compar ison gr oups by independe nt variable s of exposure to barrie rs and e x posure to fa cilitators, and by outcome mea sures of fa vorable labora tory results, medica tion adhere nce, a nd appointment adher ence Cases Comparison g roups Postpartum mot he rs Dy a ds I nf a nts Yes No Total % Yes No Total % Yes No Total %x p 2 Expo su re to barr iers 31 69 100 5 95 100 31 69 100 87.70 .0001 Expo su re to fac ilitators 6 94 100 100 0 100 97 3 100 177.36 .0001 F a vo ra ble labora tory results 47 53 100 96 4 100 96 4 100 98.80 .0001 Medica tion adher ence >90% 46 54 100 90 10 100 92 8 100 74.10 .0001 Appointment adher ence 52 48 100 80 20 100 97 3 100 57.20 .0001 R e s e a r c h Q ue s t io n T wo I s there a signific ant re lationship between the outcome var iables (labor atory results, appointment adhe renc e, and me dication adhe renc e) a nd demog raphic varia bles (ag e, ra ce, me ntal illness, poverty new diag nosis, pill burden, hig h school educ ation, su bs ta nc e a bu se ) a mon g the thr e e g ro up s? An a ly sis of fr e qu e nc y a nd Coc hr a nMa nte lHa e ns zel Sta tis tic we re us e d to a dd re ss t his qu e sti on T he re su lts sh ow n in Ta ble 44 in dic a te d th a t th e re la tio ns hip betwee n labora tory results and a g e, mater nal ra ce, me ntal illness, poverty pill burden, a nd ne w d ia g no sis we re sta tis tic a lly no ns ig nif ic a nt f or e a c h g ro up T he re la tio ns hip be tw e e n la bo ra tor y re su lts a nd e du c a tio n w e re sta tis tic a lly no ns ig nif ic a nt f or inf a nts and for postpar tum mothers (Chi-Square= 4.79, p=0.03). The CochranMantel-Ha enszel Statist ic indicated a n overa ll significa nt relationship betwee n labora tory results and educa tion (Chi-Square=4.92, p= 0.03). Mothers with a hig h school educ ation wer e 2.24 t i m e s m o r e l i k e l y t o h a v e f a v o r a b l e l a b o r a t o r y r e s u l t s ( C I: 1.1-4.6). The re w as also a

PAGE 58

47 signific ant associa tion between la boratory results and substanc e abuse for postpar tum m o t h er s (C h i -S q u ar e= 4 9 7 p =0 0 3 ) a n d fo r d y ad s (C h i -S q u ar e= 4 7 8 p =0 0 3 ), t h o u gh there was no sig nificant re lationship to i nfant outcome s. The over all association wa s sta tis tic a lly no ns ig nif ic a nt. Table 44. Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of la bo ra tor y re su lts Cases Comparison g roups Postpartum mot he rs Dy a ds I nf a nts x p x p x p 22 2 Odds ra tio 95 % C I Ag e 0.79 0.85 0.85 0.84 2.55 0.47 Race 3.02 0.55 6.25 0.18 1.72 0.79 Mental illness 1.09 0.30 1.72 0.19 0.01 0.92 Poverty 0.40 0.53 2.05 0.15 0.54 0.46 New dia g nosis 0.22 0.64 0.25 0.61 0.25 0.61 HS/GED 4.79 0.03 0.21 0.65 0.21 0.65 2.24 1.1-4.6 Substance a buse 4.70 0.05 4.79 0.15 0.13 0.72 The re sults shown in Table 4-5 indicate d that the re lationship between me dication a dh e re nc e a nd a g e ma te rn a l r a c e me nta l il lne ss, po ve rt y p ill bu rd e n, a nd ne w d ia g no sis we re sta tis tic a lly no ns ig nif ic a nt f or e a c h g ro up T he re wa s a sig nif ic a nt r e la tio ns hip betwee n materna l educa tion and medication adhe renc e for infants (Chi-Square =4.48, p=0.03) a nd postpartum mothers (Chi-Square =5.97, p=0.01) though the re w as no sig nif ic a nt a sso c ia tio n w ith me dic a tio n a dh e re nc e a nd e du c a tio n f or dy a ds T he ov e ra ll association of me dication adhe renc e and e ducation for g roups was statistically signific ant (C hiSqu a re = 8. 07 p = 0. 00 5) M e dic a tio n a dh e re nc e wa s 2 .4 9 ti me s mo re lik e ly to b e in the ac cepta ble ra ng e if mothers ha d a hig h school educ ation (C.I .: 1.32-4.7). The re w as a sig nificant re lationship between me dication adhe renc e and substanc e abuse for mother s (C hiSqu a re = 4. 89 p = 0. 03 ), bu t no t f or dy a ds or inf a nts T he g e ne ra l a sso c ia tio n te st be tw e e n me dic a tio n a dh e re nc e a nd su bs ta nc e a bu se fo r t he g ro up s w a s st a tis tic a lly

PAGE 59

48 signific ant (Chi-Square =5.11, p=0.02) There was no sig nificant re lationship between the independe nt variable s and appointment adhe renc e, as shown in Ta ble 4-6. Table 45. Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of medica tion adhere nce Cases Comparison g roups Postpartum mot he rs Dy a ds I nf a nts x p x p x p 22 2 Odds ra tio 95 % C I Ag e 0.52 0.91 0.42 0.94 0.96 0.81 Race 1.74 0.78 4.33 0.36 1.11 0.89 Mental illness 1.33 0.25 7.66 0.01 1.06 0.30 Poverty 0.60 0.44 1.65 0.20 0.05 0.83 New dia g nosis 0.39 0.53 0.01 0.98 1.53 0.22 HS/GED 5.97 0.01 0.01 0.93 4.49 0.04 2.49 1.3-4.7 Substance a buse 4.89 0.04 0.36 0.55 0.34 0.56 Table 46. Relationship between ma terna l ag e, ra ce, me ntal illness, poverty educa tion level, substance abuse, a nd objective pill burden a nd the outcome mea sure of appointment adhe renc e Cases Comparison g roups Pos tpa rt um m oth e rs Dy a ds I nf a nts x p x p x p 22 2 Ag e 4.91 0.18 8.70 0.07 2.68 0.44 Race 4.09 0.39 3.20 0.52 1.98 0.74 Mental illness 2.09 0.15 0.92 0.34 0.13 0.72 Poverty 1.46 0.23 0.00 0.99 0.55 0.46 New dia g nosis 0.26 0.61 1.08 0.20 0.09 0.76 HS/GED 1.81 0.18 0.99 0.32 0.38 0.54 Substance a buse 3.85 0.12 0.13 0.72 0.77 0.38 Rese arch Question Th ree I s there a signific ant associa tion between the outcome var iables (medic ation adher ence labora tory results, and a ppointment adhere nce) and structur al fa cilitators and str uc tur a l ba rr ie rs ? An a ly sis of fr e qu e nc y a nd Coc hr a nMa nte lHa e ns zel Sta tis tic we re us e d to a dd re ss t his qu e sti on T he re su lts sh ow n in Ta ble 47 in dic a te d th a t th e re la tio ns hip betwee n labora tory results and struc tural bar riers ( insurance companies, inc luding

PAGE 60

49 Me dic a id H MO s, tha t li mit a c c e ss t o c omp re he ns ive c a re in t he su bs pe c ia lty HI V c lin ic or re quirement of a n individual to li nk self with follow-up c are including the require ment of selfinitiated case manag ement bef ore c linic appointment is given) wer e statistically nonsignific ant for ba bies and dy ads but wer e sig nificant for postpartum mothers (Chi-Square =5.53, p=0.02) The ove rall assoc iation acc ording to CochranMa nte lHa e ns zel Sta tis tic wa s n ot s ta tis tic a lly sig nif ic a nt. I n a dd iti on th e re su lt indicated that the r elationship betwee n labora tory results and struc tural fa cilitators (t ra ns po rt a tio n to c lin ic T a rg e te d O utr e a c h to Pr e g na nt Wo me n A c t in vo lve me nt [t o provide tra nsportation, emotional support, and/or e x pedited pape rwork], HI V nurse a nd s o c ia l w o r k e r in v o lv e me n t, me d ic a ti o n d is p e n s e d [i n s te a d o f p r e s c r ip ti o n s g iv e n ], pr ima ry c a re a va ila ble on sit e in ta ke pa pe rw or k n ot n e c e ssa ry a nd /or pr e vio us ly established and ong oing r elationship with case ma nag er) was not statistically signific ant for e ach g roup. Table 47. Relationship between struc tural bar riers a nd structura l fac iliti es and the outcome mea sures of la boratory results, medica tion adhere nce, a nd appointment adher ence Cases Comparison g roups Postpartum mot he rs Dy a ds I nf a nts x p x p x p 22 2 Odds ra tio 95 % C I L a bo ra tor y re su lts Structural bar riers 5.53 0.02 1.87 0.17 0.64 0.22 Structural fa cilitators 3.38 0.07 0.13 0.72 Medica tion adhere nce Structural bar riers 4.23 0.04 0.15 0.70 0.58 0.44 Structural fa cilitators 3.58 0.06 0.27 0.60 Appointment adher ence Structural bar riers 4.46 0.03 1.84 0.18 1.32 0.25 0.36 0.17-0.78 Structural fa cilitators 5.90 0.03 0.09 0.76 32 1-1034

PAGE 61

50 Re se ar c h Que st ion F our Ar e ma te rn a l de pr e ssi on or oth e r m e nta l il lne ss, pr oximi ty of HI V d ia g no sis rac e, income, a dolesce nce, hig h school educ ation, poverty substance a buse, pill burden, structura l barrie rs to acc ess, and struc tural fa cilitators to car e sig nificantly differ ent as measure d by the labora tory results, medica tion adhere nce, a nd appointment adher ence of HI Vpo sit ive pr e g na nt w ome n, po stp a rt um m oth e rs o r t he ir inf a nts ? L og istic reg ression wa s used to addre ss this hy pothesis. The for war d selec tion proce dure w as utiliz ed to obtain the optimal model for e ach outc ome var iable. As shown in Table 4-8, the final ana ly sis of Maxi mum L ikelihood Estimates shows the following varia bles wer e sig nificant to labora tory results: ag e (p = 0.04), mater nal educ ation (p = 0.008) membership in the baby g roup (p = 0.0001), or the dy ad g roup (p = 0.0001). The a ssociation betwee n positive laboratory results and y oung materna l ag e wa s inversely rela ted, with an odds ra tio of 0.6 (95% C. I 0.40-0.98). Ma terna l educa tion provided a n odds ratio of 2.8 (95% C. I 1.31-6.06). Comparing membership in the baby g roup or the dy ad g roup to that of the postpar tum mother g roup had a n odds ratio of 31.7 (9 5 % C I : 1 0 4 9 -9 5 9 4 ). T h i s m ea n s t h at y o u n g m at er n al age m at er n al l ac k o f a h i gh school educ ation or its equivalent, and simply being in the postpartum materna l gr oup wer e most closely associate d with poor labora tory finding s. Ba bies wer e 31 times more lik e ly to h a ve the ra pe uti c la bo ra tor y re su lts tha n f or tho se po stp a rt um m oth e rs or dy a ds Table 48. Summary of log istic reg ression ana ly sis predicting favor able labor atory re su lts Va ri a ble B SE Wal d sta tis tic p Odds ra tio 95 % Wa ld confide nce lim its Ag e -0.46 0.23 4.21 0.04 0.63 0.4-0.9 Educa tion 1.04 0.39 7.03 0.008 2.80 1.3-6.1 Group: B aby 3.46 0.56 37.48 0.0001 31.72 10.49-95.94 Group: Dy ad 3.46 0.56 37.48 0.0001 31.72 10.49-95.94

PAGE 62

51 The fina l analy sis of Maxi mum L ikelihood Estimates shows the following varia bles (Ta ble 4-9) w ere signific ant to medication adhe renc e: mater nal educ ation (p = 0.005) being in the baby g roup (p = 0.0001), and be ing in the dy ad g roup (p = 0.0001) The odds ra tio estimates showed that mothers with a hig h school educ ation or its e qu iva le nt h a d a 2. 5 ti me s g re a te r c ha nc e (9 5% C. I .: 1 .3 24. 72 ) o f a dh e ri ng to their antire troviral medica tions. Members of the ba by g roup we re a lmost 12 t imes more likely to rec eive their medications (95% C. I .: 5.29-25.39), and me mbers of the dy ad g roup we re a lmost 15 t imes more likely to adher e to their medic ations (95% C. I 6.434.58). Table 49. Summary of log istic reg ression ana ly sis predicting favor able medic ation adher ence Va ri a ble B SE Wal d sta tis tic p Odds ra tio 95 % Wa ld confide nce lim its Educa tion 0.91 0.33 7.85 0.0050 2.49 1.3-4.7 Group: B aby 2.45 0.4 37.5 0.0001 11.59 5.29-25.39 Group: Dy ad 2.7 0.43 39.34 0.0001 14.87 6.4-34.6 Table 410 shows that the final a naly sis of Maxi mum L ikelihood Estimates ind ic a te s th e fo llo wi ng va ri a ble s w e re sig nif ic a nt t o a dh e re nc e to t he re c omm e nd e d c lin ic a pp oin tme nts : ma te rn a l a g e (p = 0. 04 ), str uc tur a l ba rr ie rs (p = 0. 00 37 ) a nd me mbe rs hip i n t h e d y ad gro u p (p = 0 0 0 0 1 ). T h e o d d s ra t i o es t i m at es (0 6 6 ) s h o we d t h at y o u n ger age had a ne g ative re lationship to appointm ent adhe renc e (95% C. I 0.45-0.97), a s did the prese nce of structura l barrie rs odds ratio 0.32 (95% C. I 0.15-0.69). The dy ad g roup wa s 23 times more likely (95% C. I 6.71-79.02) to have favor able a ppointment adhere nce. Suppo r t fo r Hyp ot he sis Ba sed on the re sults presented a bove, the hy pothesis: Maternal de pression or other menta l illness, prox imity of HI V diag nosis, race income, adole scenc e, hig h school

PAGE 63

52 educa tion, substance abuse pill burden, structura l barrie rs to acc ess, and struc tural fac ilitators to care are associate d with a sig nificant diff ere nce in a dhere nce a s measure d by the la bo ra tor y re su lts mi sse d me dic a tio n d os e s, a nd de la y e d o r m iss e d c lin ic a pp oin tme nts of HI Vpo sit ive pr e g na nt w ome n, po stp a rt um m oth e rs o r t he ir inf a nts is su pp or te d a nd sta tis tic a lly sig nif ic a nt. Table 410. Summary of log istic reg ression ana ly sis predicting favor able a ppointment adher ence Va ri a ble B SE Wal d sta tis tic p Odds ra tio 95 % Wa ld confide nce lim its Ag e -0.41 0.20 4.37 0.04 0.66 0.4-0.9 Educa tion -1.15 0.39 8.45 0.004 0.32 0.15-0.67 Group: B aby 0.62 0.41 2.28 0.13 1.86 0.83-4.15 Group: Dy ad 3.14 0.63 24.85 0.0001 23.03 6.7-79

PAGE 64

53 CH APT ER 5 DI SCUS SI ON AND I MPL I CATI ONS P ur pos e Th e pu rp os e of thi s st ud y wa s to de te rm ine if pr e vio us ly ide nti fi e d b a rr ie rs to HI V trea tment adher ence aff ect the pe rinatal dy ad, the postpar tum mother, or the exposed infant equally ; and to deter mine if the propose d constructs of struc tural fac ilitators or structural ba rrier s are associate d with HI V trea tment adher ence in the perina tal dy ad, the postpar tum mother, or the exposed infant. The depende nt variable s wer e oper ationalized as follows: adher ence to antiretrovira l medication = se lf repor t of missing <10% of doses; adhe renc e to appointment sche dule = a ll primary and su bs pe c ia lty c a re re c e ive d w ith in t he re c omm e nd e d ti me fr a me ; a nd the ra pe uti c la bo ra tor y re su lts = a sta ble or fa lli ng HI V v ir a l lo a d, a nd a sta ble or ri sin g CD 4 c ou nt. The re sults showed that all outcome mea sures we re sig nificantly differ ent betwee n the infant a nd the postpartum mother g roups. Fur ther, e x posure to the teste d constructs of struc tural fa cilitators and structur al bar riers a ccounte d for the ma jority of the diffe renc es. The use of the wome n and babie s as their own c omparisons (at diff ere nt po int s in the c a re sy ste m) a ssu re d th a t th e e xpos ur e s to ma te rn a l f a c tor s w e re the sa me for a ll three g roups. This allows more c onfidenc e that the diff ere nces e x presse d wer e likely associate d with the exposures to structural bar riers a nd structura l fac ilitators.

PAGE 65

54 Figure 5-1.Stigma theory in this study, as adapted from Goffman (1963)

PAGE 66

55 Desc riptive Statistics Str uc tur a l ba rr ie rs we re pr e se nt f or 69 % o f p os tpa rt um m oth e rs 3 1% of dy a ds and 5% of babies. This shows hete rog eneity betwee n g roups (p< .0001), with postpartum mothers sig nificantly more likely to experience structura l barrie rs to car e than ba bies or dy ads. Structural fa cilitators also appe are d heter og eneous be tween g roups (p< .0001), being prese nt for 100% of babies, 97% of dy ads, and 6% of postpartum mothers. CD4 c ou nts a nd HI V v ir a l lo a ds we re sig nif ic a ntl y le ss l ike ly (p < .0 00 1) to b e in t he a c c e pta ble rang e for postpartum mothers (47%) than for ba bies (96%) or dy ads (96% ). Gre ater than 10% of medic ation doses wer e missed in 54% of postpar tum mothers, while only 10% of babies a nd 8% of dy ads re ported missing > 10% of dose s. This shows a diffe renc e at the (p<.001) level of sig nificanc e. The last outcome tha t differ ed betwe en g roups was missed or de lay ed a pp oin tme nts A g a in, po stp a rt um m oth e rs we re le ss l ike ly (p < .0 00 1) to a tte nd c lin ic appointments (48% missed) than w ere their babie s (20% missed) or the same mother s during their pre g nancy (3% missed). Laboratory Re sults Thera peutic labor atory results consisted of stable or dec rea sing vira l loads and sta ble or inc re a sin g CD 4 c ou nts F or po stp a rt um m oth e rs th e ra pe uti c la bo ra tor y re su lts wer e positively impacted by materna l achieve ment of a hig h school diploma or e qu iva le nt ( p= 0. 02 86 ). I n f a c t, e ve n a ft e r a dju sti ng fo r g ro up me mbe rs hip s ub je c ts i n a ll g roups wer e 2.24 times more likely to have positive labora tory finding s if the mothers had a hig h school educ ation or its equivalent.

PAGE 67

56 Thera peutic labor atory results for postpar tum mothers were neg atively aff ecte d by pr e na ta l ma te rn a l su bs ta nc e a bu se a t di a g no sis a c c or din g to t he F ish e r s E xac t T e st (p=0.046) This effe ct is limi ted to the postpartum mothers, ther e wa s no g ener al association (p= 0.21). Ther e we re f ew subjec ts who were positive for illicit subst ance s at the onset of their preg nancy (n=3) and those who w ere enter ed re sidential treatment prog rams to maintain custody of their ba bies. These residential pr og rams re quired routine drug -testing so ong oing use was not in question. Since there was not ong oing su bs ta nc e a bu se th e c lin ic a l si g nif ic a nc e of thi s f ind ing is n ot r e a dil y a pp a re nt. Materna l CD4 counts and viral loads we re a lso neg atively aff ecte d by the prese nce of structura l barrie rs to car e (p= 0.019). Ag ain, no sig nificant g ener al association wa s found (p= 0.11) to sug g est these ba rrier s aff ecte d the labora tory results of infants or dy ads. This means that y oung materna l ag e, mater nal lack of a hig h school educa tion or its equivalent, and simply being in the postpartum materna l gr oup wer e most closely associate d with increa sed viral loads a nd decr ease d CD4 counts. Medicat ion Adh ere nce Medica tion adhere nce dur ing pr eg nancy (g roup=dy ad) w as neg atively aff ecte d by materna l mental illness accor ding to the F ishers Exact Test (p= 0.015), thoug h materna l mental illness was not shown to have a g ener al associa tion once adjustment wa s made for g roup member ship (p=0.84). A g ain, this was a r elatively small gr oup of women ( n=21), a nd the c lin ic a l si g nif ic a nc e is n ot r e a dil y a pp a re nt. I nfant medic ation adher ence was c orre lated with mater nal educ ation using the F ish e r s E xac t T e st ( p= 0. 04 5) Po stp a rt um m a te rn a l me dic a tio n a dh e re nc e wa s a lso positively corr elated w ith maternal e ducation (p= 0.015). Afte r adjusting for g roup

PAGE 68

57 membership, a g ener al associa tion (p=0.005) e x isted. Subjects in all gr oups were 2.49 times more likely to report medic ation adher ence if mothers had r ece ived a hig h school e du c a tio n o r i ts e qu iva le nt. Postpartum maternal medic ation adher ence was ne g atively associate d with prior materna l substance a buse ac cording to analy sis using the F ishers Exact Test (p= 0.04). I n fac t, there w as a g ener al associa tion (p=0.024) a fter a djustment for g roup member ship. This means subjects in all g roups wer e twice as likely to miss m edica tions if the mother had a history of prior substanc e abuse Finally there was a corr elation betwe en medica tion adhere nce a nd the pre sence of structura l barrie rs noted (p= 0.04) only for the postpa rtum mothers g roup. This means that only subjects in the postpartum mother g roup we re likely to experience structura l barr iers that af fec ted their medic ation adher ence The log istic reg ression model showed tha t the best predic tors of medica tion adher ence wer e mater nal educ ation (p=.005) membership in the infa nt gr oup (p<.0001) and member ship in the dy ad g roup (p< .0001). The odds r atio estimates show that postpartum mothers with a hig h school educ ation or its equivalent have a 2.5 times g rea ter c hance of succ essful adhe renc e to their a ntiretrovira l medications. Simpl y being in t he inf a nt g ro up me a nt s ub je c ts w e re a lmo st 1 2 ti me s mo re lik e ly to r e c e ive the ir me dic a tio ns B e ing in t he dy a d g ro up me a nt s ub je c ts w e re a lmo st 1 5 ti me s mo re lik e ly to adher e to their medic ations. Th is s ho ws tha t so me thi ng a bo ut t he inf a nt a nd pr e na ta l dy a d g ro up s si g nif ic a ntl y enhanc ed medica tion adhere nce. F or postpartum mothers, the be st predictor of thera peutic medica tion adhere nce w as educ ation.

PAGE 69

58 Appoin tm ent Adhere nce Postpartum maternal a ppointment adhere nce w as neg atively corr elated w ith the prese nce of structura l barrie rs (p= 0.03). Statistical testing a lso showed a g ener al association (p= 0.008). I n this case, the inf luence of structura l barrie rs was so strong that afte r adjusting for g roup member ship subjects were about 3 times less likely to attend clinic appointments if there wer e structur al bar riers to doing so. I n contra st, postpartum maternal a ppointment adhere nce w as positively aff ecte d by the pre sence of structura l fac ilitators to care (p=0.03) There was a lso a g ener al association af ter a djustment for g roup member ship (p=0.02). Reg ardle ss of the g roup, subjects wer e 32 times more likely to adher e to the re commended c linic appointment schedule if structura l fac ilitators were prese nt. The log istic reg ression model showed tha t older mater nal ag e (p= .0365), pre sence of str uc tur a l ba rr ie rs (p = .0 03 7) a nd me mbe rs hip in t he dy a d g ro up (p < 0. 00 01 ) b e st explained appointment adher ence The odds ra tio estimates showed that y oung ag e had a neg ative re lationship to appointm ent adhe renc e (O R=0.66), as did the pr esenc e of structura l barrie rs (OR=0.32) Membership in the dy ad g roup incre ased the likelihood of a dh e re nc e to a pp oin tme nts I n f a c t, w ome n w e re 23 tim e s mo re lik e ly to a tte nd c lin ic v i s i t s d u r i n g t h e p r e n a t a l p e r i o d t h a n t h e y w e r e t o c o n t i n u e t h e i r c a r e a f t e r d e l i v e r y. Summ ary of Findings Str uc tur a l ba rr ie rs we re pr e se nt f or 69 % o f p os tpa rt um m oth e rs 3 1% of dy a ds a nd 5% of inf a nts (p < .0 00 1) St ru c tur a l f a c ili ta tor s w e re pr e se nt f or 10 0% of inf a nts 97% of dy ads, and only 6% of mother s (p<.0001) This means that extrinsic structural ba rr ie rs in c lud ing ins ur a nc e lim ita tio ns a nd c omp lic a te d c lin ic a dmi ssi on pr oc e du re s, we re pr e se nt f or a ma jor ity of mot he rs a nd a lmo st n o in fa nts Co nv e rs e ly e xtri ns ic

PAGE 70

59 structura l fac ilitators, including compr ehensive funding and strea mlined clinic admission pr oc e du re s w e re pr e se nt f or a ll o f t he inf a nts a nd a lmo st n on e of the mot he rs T his shows that policies favor infant outcomes by almost never pr esenting barr iers to infant c a re a nd a lmo st a lw a y s g ivi ng mot he rs the too ls n e e de d to su c c e e d w ith the c a re of the ir infants. On the other hand, mothers we re r are ly the re cipients of structur al help for themselves, and e x perie nced struc tural bar riers ove r two-thirds of the time. As for outc ome measur es, 96% of infants and dy ads had a cce ptable labor atory results. This is in contrast to only 47% of postpar tum mothers. This showed a sig nif ic a nt d if fe re nc e (p < .0 00 1) M e dic a tio n a dh e re nc e ha d a sim ila r d ist ri bu tio n, wi th only 10% of infa nts and 8% of dy ads re porting poor adhe renc e. Ove r half ( 54%) of postpartum mothers re ported poor a dhere nce, a signific ant diffe renc e ag ain (p< .0001). Gi ve n th e se fi g ur e s, the la st o utc ome a dh e re nc e to r e c omm e nd e d c lin ic a pp oin tme nts is not surprising. D y ads we re r are ly nonadher ent (3%) infants (20%) and postpartum mothers almost half of the time (48%). Postpartum mothers who ha d alre ady demonstrated the ability to adher e to trea tment during their pre g nancie s experience d the majority of nonthera peutic labor atory results, missed medications, and missed a pp oin tme nts Missing an d Exclu ded Data Complete data sets we re obta ined from 300 subjec ts. No data points wer e mis sin g a t th e tim e of a na ly sis du e to a va ila bil ity of re du nd a nt c ha rt ing via pa pe r c ha rt s, computer pr og rams, and a cross disciplines within the clinic. The indepe ndent var iable pill burde n wa s opera tionaliz ed to measur e number and times of pills per day ag ainst the trea tment norm or standar d at the time of ther apy Be cause most subjects beg an medica tion therapy during their pre g nancy (trea tment

PAGE 71

60 nave) prog ression to burdensome reg imens due to drug resistanc e wa s not found in any instance. H ad this not been a char t review self-r eporte d pill burden as per ceive d by subjects would likely have be en a va lid dichotomous m easur e to add to the model. F or purposes of this ana ly sis, pill burden is shown a s fre quency only I t was not opera tionaliz ed in this study in a manner that showed he terog eneity Materna l substance a buse is underr epre sented in this sample bec ause of the study purpose a nd its ex clusion criter ia. The pur pose wa s to ex amine outcomes of mothers and babies whe n challeng ed with the same materna l environment. This prohibited ana ly sis of babies that we re not in mater nal custody beca use the e nvironment then diffe red f rom that of the mot he r. Mo the rs tha t ut ili zed ill ic it d ru g s d ur ing pr e g na nc y we re lik e ly to l os e c us tod y of the ir inf a nts a t le a st t e mpo ra ri ly a nd thu s e xclu sio n c ri te ri a e lim ina te d th e se infants, their mothers, a nd the dy ad fr om the sample. Et hic al C ons ide r at ion s As previously stated, postpartum mothers who ha d alre ady demonstrated the ability to adher e during their pre g nancie s experience d the majority of missed medications, missed appointments, and a r esulting inc rea se in their vira l loads and de c re a se in t he ir CD 4 c ou nts (i mmu ne fu nc tio n) Wit h th e e xce pti on of e du c a tio n, the se results wer e not found to be a ssociated with wa y s of desc ribing these mothers, ( i.e., black, poor drug -addic ted, or de presse d). What did appea r to influence outcomes wa s the structure of the he alth car e sy stem in which they wer e expected to f unction. The complexit y of the sy stem in fact, may be why an incr ease in materna l educa tion was associate d with succe ssful treatment. F or example, the mother of a neona te is g iven an actua l phy sical supply of Retrovir f or her baby and re quired to re turn-demonstra te prope r administration by a HI V nurse spe cialist. The same nurse g ives the mother a n

PAGE 72

61 a pp oin tme nt t ime a nd da te fo r t he inf a nt t o f oll ow -u p w ith the su bs pe c ia lty nu rs e prac titioner in the HI V clinic af ter discha rg e home. The postpar tum mother, on the other ha nd, has no mec hanism in place to he lp her rec eive or continue her own medica tions, and is in fact, not allowed to make an a pp oin tme nt f or su bs pe c ia lty c a re un til a ft e r h e r 6 we e k v isi t in the ob ste tr ic a l c lin ic postpartum; a clinic that no long er a ddresse s her HI V-re lated nee ds once she has de liv e re d h e r b a by A n a dd iti on a l r e qu ir e me nt i s th a t mo the rs c on ne c t wi th c a se manag ement prior to e ntering car e in the subspec ialty clinic. Fur ther, she likely loses acc ess to medications in the interim, bec ause the Medica id that cover ed the pr eg nancy us ua lly e xpir e s b e fo re (o r i f) sh e lin ks wi th c a se ma na g e me nt t o b e g in t he pr oc e ss t o acc ess the AI DS Drug Assistance Prog ram (A DAP). I t does appe ar tha t decisions made about who will rec eive wha t ty pe of he lp throug h the maze of c are is important to the overa ll health of HI V-positive women. The pe ople who dec ide what c onstitut es a sc ientific problem sha pe the wor ld, beca use the wor ld is shaped by rese arc h (Ka ne & Thomas, 2000). The development of kn ow le dg e is a po lit ic a l e nte rp ri se tha t r e fl e c ts s oc ie ty s c ur re nt d omi na nt v a lue s, including the deter mination of which g roups and subjec ts should be resea rche d and by whom (B rowne 2001). Since re sear ch is the founda tion for best pra ctice, it is important to include mater nal outcome mea sures in trea tment protocol. The r esults of this study sug g est that a ne ed for a more inclusive approa ch to mater nal hea lth in the perinatal a nd postnatal periods is justified by basic e thical conc epts, including personhood a nd value of mot he rs a nd fe tus e s. Ac c or din g to B e a uc ha mp a nd Chi ldr e ss ( 19 94 ), the fi rs t pr inc ipl e in b ioe thi c s is the re sp e c t f or a uto no my T his me a ns pe rs on s w ho a re a ble to m a ke de c isi on s a bo ut t he ir

PAGE 73

62 own ca re must be a llowed to, and re sear cher s and provider s must respect these decisions. This does not happen w hen HI V-positive mothers ar e judg ed inca pable of understanding risk and tre atment options, or when tre atment counse ling c onsists of discussing fe tal health in the abse nce of discussions of the impact of that tre atment on mater nal hea lth. Anti-retr oviral prophy laxi s g iven to the fe tus via the mother is not fre e fr om impact on the mothers he alth. Viral mutation and re sultant drug resistanc e ar e docume nted conseque nces of even shor t-term prophy laxi s during the per inatal per iod (Toni et al., 2005). Moreove r, it is important to i nclude mate rnal outcome s as a mea sure in any study that uses the bodies of mother s to impact the hea lth of others. Succe ssful adher ence overa ll is predicated on pa tients understanding and ac cepting their diag nosis, and expression of a willingne ss to take medica tions (Roge rs et al., 2001) Guidelines ar e e vo lvi ng ra pid ly to a llo w d e la y in t he int ro du c tio n o f a nti re tr ov ir a l me dic a tio ns un til patients are rea dy The a im is t o minimi ze self and c ommunity harm by minimi zing multipl e drug resistant viral mutations and the w aste of r esourc es. Preg nant women a re the exception to this rule, as the protec tion of the fetus is an immediate ne ed, and w aiting mon ths or y e a rs fo r m a te rn a l r e a din e ss i s n ot c on sid e re d a via ble op tio n. Gu ide lin e s st a te that antiretrovir al thera py must be started by wee k 14 of pre g nancy or as soon ther eaf ter as the woma n is identified as HI V-positive (Steinhart, Or rick, & Simps on, 2002). B ut ethicists caution not to overrule the dec isions of mothers reg arding their babie s hea lth, as the g ener al view that be ing H I V-positive is the worst possible outcome for a baby may no t be a c c ur a te wh e n v ie we d in the fu ll c on te xt of a mot he r s w or ld ( F uCha ng Ts a i, 2001).

PAGE 74

63 Reinforc ement of the position t hat being HI V-positive is the worst possible thing that can ha ppen deva lues the lives of the mother s, whether or not that is the intention. With ou t a fo c us on the he a lth ou tc ome s o f a ba by a nd its mot he r, the me ssa g e is re inf or c e d a t e a c h e nc ou nte r t ha t th e ba by is t he tr ue pa tie nt. One a pproac h is to consider Clarke s (1999) a ssertion that the mother a nd the fetus should be c onsidered jointly as the person when making decisions about the person s future This appea rs to be the most sound answe r to the question of whic h is of more va lue; to simpl y state that the que stion is voi d beca use they are the same individual. This is beneficia l to the fetus, bec ause the only rela tionship t he fe tus can for m is one of biologica l depende nce on the mother. The mother s psy cholog ical and e motional depende nce on the fetus is also honore d. This positi on values the dy adic a nd interdepe ndent nature of the mother a nd fetus, and a s such, serve d as the e thical perspe ctive of this re sear ch study The e thical position remains that to compromise either person would ha rm both. Since the mother a nd the fe tus are both e ntitled to moral standing, ther e ar e duties owe d to the dy ad by the medica l and re sear ch community The a bsence of trea tment for the mil lio ns of wo me n in fe c te d w ith HI V w hil e the pr e ve nti on of mot he r t o c hil d transmission (MTCT) rema ins the focus of r esea rche rs and c linicians worldwide, use s the justification of the utilitarian ar g ument that protec ting e x posed babie s does the most g ood for the most people with limi ted alloca ted re source s. After all, the mothers ar e alr eady infec ted. Alleviation of suff ering in mothers is ignore d in the name of cost-ef fec tiveness. This is a contrived pove rty beca use the re source s exi st to do both i n the United States. I t is essential to remember that whether in the United States or else wher e in the wo rl d, imb ue d in the va lue of the mot he r i s th e fa c t th a t sh e is a pe rs on wi th H I V. Ha rr is

PAGE 75

64 (1 99 9) a dd re sse d th e c os t of a c kn ow le dg ing so me on e s pe rs on ho od it br ing s h e r i nto the sa me mor a l c a te g or y a s th os e wh o a re un inf e c te d. Judg ing so me on e a s a no np e rs on is a wa y of dis ta nc ing he r r e a lit y fr om e ve ry on e e lse s T o a c kn ow le dg e he r s a me ne ss wo uld be to a dmi t th a t e ve ry on e is v uln e ra ble to H I Vinf e c tio n. Th is c a n b e inc re dib ly unsettling for the uninfec ted world. I t is the duty of policy maker s and provider s to keep actua l medical and r esea rch pr actice within the confines of moral beha vior. This means that reinfor cing the stigma place d on HI V-positive mothers throug h prac tices that e xce ssi ve ly we ig ht f e ta l ou tc ome is b oth imm or a l a nd un e thi c a l. Sign if icance Providers and policy maker s focus on the c ause most proxim al to the sy mptom the y a re tr e a tin g (K ri e g e r, 19 94 ). I n th e c a se of pe ri na ta l H I V t ra ns mis sio n, the c a us e is exposure to the mother. B ut the more distal exposures that rende red the mother susceptible to infec tion go unc hecke d. Fa ilure to addre ss the larg er dispar ities makes the preve ntion effor t shortsighted. The same infa nts who benef ited by the PACTG 076 protocol in the mid-1980s have now g rown into impoverished, a t-risk adolesc ents who e n g a g e i n a c t i v i t i e s ( o t h e r t h a n e x p o s u r e t o t h e i r m o t h e r s ) t h a t c o u l d r e s u l t i n H IV inf e c tio n. Chi ldr e n o f m ino ri ty g ro up s w ho se c on te mpo ra ri e s w e re sp a re d b y thi s protocol ar e now c oming of ag e to bea r their own e x posed infants. T h e v a l u e l a d e n t e r m m o t h e r t o c h i l d t r a n s m i s s i o n ( M T C T ) i s t h e a c r o n ym u s e d b y t h e s c i e n t i f i c c o m m u n i t y w o r l d w i d e t o d e s c r i b e p e r i n a t a l t r a n s m i s s i o n It assumes that the fe tus is already born (c hild, not fetus, though the ma jority of transmissions do not occur af ter birth), a nd it assigns blame to the mother for g iving the baby HI V. Thoug h mothers often r eport primar y risk for HI V infec tion to be unprotecte d

PAGE 76

65 se x with the fa the rs of the se ba bie s, a ny a ssi g nme nt o f r e sp on sib ili ty to t he fa the r i s a bs e nt. Mother-to-c hild-transmission assumes the mother is exclusively to blame if her c hil d is bo rn wi th H I V. I t ma ke s h e r t he e xec uto r o f a n im mor a l a c t up on a de fe ns e le ss c hil d, a nd jus tif ie s h os til ity tow a rd he r. Ve rt ic a l tr a ns mis sio n a lso de no te s a dir e c t r ou te from mother to c hild. No value-ne utral term e x ists (i.e., pare nt to child transmission). Since the tra jectory of infec tion is m ore a ccur ately fathe r to child via mother, ve rtical transmission could more ac cura tely be desc ribed a s ang ular. This is supported by McNa ir and Prathe rs (2004) rec ent finding s indicating that a par tners r isk behaviors exert more influenc e on a w omans HI V status than her own beha vior. I t follows that if a mothers he alth is predicted by the beha vior of the f ather so then is the baby s. I nte re sti ng ly b oth the me dic a l c omm un ity a nd so c ie ty pla c e the so le responsibility and blame f or the highrisk pre g nancy on mothers. An example is the messag e conve y ed in the Hig h Risk Obstetrical Clinic (HROB) wher e the pr edominant messag e to HI V-positive mothers is one of sa ving the babies fr om the highrisk situations in w hic h th e ir mot he rs ha ve pla c e d th e m. I n o ne qu a lit a tiv e stu dy (I ng ra m & Hutchinson, 1999a), a black woma n spoke of the hostility she fe lt every time she walke d i n t o t h e c l i n i c t o s ee a b ea u t i fu l b l ac k b ab y o n a s t at es p o n s o re d p re v en t i o n s i gn s ay i n g, She has he r daddy s ey es and he r mother s AI DS (F igur e 5-2) She states, When y ou are positive, y ou lear n undertone s. These undertone s, even in the he althcar e community may be a f undamental ba rrier to continued trea tment for mothers a fter the y have sa fely delivere d their babie s.

PAGE 77

66 Figure 5-2.Florida Department of Health announcement, circa 1995 A mother whose social worth is reinforced by her ability to procreate must confront the knowledge that she will never have a nonpathological pregnancy. This distortion potentially impacts decisions about engagement in self-care, and encourages

PAGE 78

67 f u r t h e r d e v e l o p m e n t o f t h e a l r e a d y l o w e x p e c t a t i o n s o f t h e m e d i c a l c o m m u n i t y In eff ect, their fate s in the biomedical equa tion are a lrea dy cast. Thus is crea ted the new risk cate g ory for unde r-tre ated HI Vthat of pr eg nant women. B ut it also crea tes the risk ca teg ory of fe tus of a minority mother, whe n ag ain; the re al risk ca teg ory is sex ual par tner or of fspring of a ma le who has se x with males. Providers spea k candidly of the ne ed to cur tail reproduc tive activity in these mothers beca use of their presumptive tre atment fa ilure and the burden the ir orphane d or infec ted babies will place on society The simple fa ct is, despite the incr ease d number of pr e g na nc ie s in wh ic h b a bie s a re e xpos e d to HI V, le ss t ha n 1 0 w e re re po rt e d a s in fe c te d in Florida in 2004 (CDC, 2005). Im plications f or F uture Re searc h Ex isting literatur e does not re flec t the rea l risk that without chang ing soc ial and economic c ircumstance s, these same HI V-ne g ative babie s and their f riends would g row into adolesce nts and ac quire their ow n HI V infec tions because the same soc ial and e c o n o m i c c h a l l e n g e s t h a t p l a g u e d t h e i r m o t h e r s n o w p l a c e t h e m a t r i s k P e r i n a t a l H IV literature has an ove rwhe lming foc us on the preve ntion of HI V-infe ction in the fetus/infant. I t can be arg ued that the r eal disea se to be ba ttled is hopelessness and a lack of ag ency not just HI V infec tion. I f this is so, then the public health establishment is not a d d r e s s i n g t h e r o o t p r o b l e m s a n d w i l l l i k e l y n o t m a k e a d e q u a t e g a i n s i n r e d u c i n g H IV infec tions in m inority populations. HI V-positive mothers see m in full recog nition of the c ir c ums ta nc e s o f t he ir liv e s. Pr ov ide rs h ow e ve r w e ll i nte nti on e d, a re un a ble to a dd re ss what re ally needs to be chang ed to lower risk; namely the introduction of the hope for a

PAGE 79

68 fu tur e a nd a g e nc y in i nd ivi du a l li ve s. Th e se ba rr ie rs mus t be lif te d th ro ug h p oli c y sh if ts at least a t a community level. This is well out of the re ach of individual providers. I n the literature disparity within communiti es is fre quently noted to aff ect he alth. I t is not si mply the abse nce of a we alth, but the also the pre sence of jea lousy or envy of oth e rs w e a lth tha t a ff e c ts h e a lth St ud ie s in dic a te tha t po or so c ie tie s w ith ou t in c ome stratifica tion have better health than we althy societies with stratifica tion (Fa rmer, 2003) I n the most ge nerous individual, exist ing be neath a nothers le vel must crea te selfloathing. F arme r (2003) ref ers to the e qualization of this di sparity as cr eating fr eedom from wa nt. I t can be descr ibed as more than a f ree dom from want, but also the f ree dom fr o m fe el i n g l es s t h an o t h er s I n p o o r n ei gh b o rh o o d s d ep ri v ed o f h o p e, an d i n t h e H i gh Ris k O bs te tr ic a l Cli nic (H RO B ), de pr ive d o f j oy tha t ot he r m oth e rs -t obe e xpe ri e nc e th is fee ling of inadequa cy is reinforc ed. The economy of wa nt and despa ir thrive in the min or ity po pu la tio ns in t he Un ite d St a te s. B e c a us e so c ia l se g re g a tio n is sti ll t he no rm minority people e ng ag e in most behaviors, including highrisk behavior s, with someone who is of the same minority beca use that is who is closest and re adily available Pr e ve nti on e ff or ts i n F lor ida c on sis t pr ima ri ly of me ssa g e s a ime d a t mi no ri ty communities, a prac tice known a s social marke ting ( Fig ure 53). People a re e ncoura g ed to know their individual HI V status beca use knowledg e is power But without the powe r to decide w hen and w ith whom to enga g e in hig h-risk beha viors, and the soc ial and str uc tur a l su pp or t to se e k tr e a tme nt i f i nf e c te d, the e xpe ns ive pr e ve nti on me ssa g e is po int le ss, be c a us e kn ow le dg e a lon e is n oth ing Wit ho ut t he po we r t o d e ma nd the ho ne st disclosure of, a nd protec tion from infec ted par tners, no pre vention is likely to occur Wha t is ne e de d is the po we r o f m ino ri ty fe ma le s to de ma nd to k no w t he sta tus of the ir pa rt ne rs a nd on c e inf e c te d, to e xpe c t a de qu a te me dic a l tr e a tme nt t o e xten d th e ir liv e s to

PAGE 80

69 see their children grow. This begins with economic power and extends to social power, and is not adequately addressed by the prevalent message of know your status. A requirement for survival with HIV is the presence of enough social and economic capital to withstand the structural violence in place, including that in place in the health care system. Figure 5-3.Florida Department of Health announcement, circa 2005 Research must begin with why infected individuals were disempowered even before they became infected, and what keeps them so now. These factors are what

PAGE 81

70 Krieg er ( 1994) re fer s to as the social a nd political determinants of he alth. Rece nt attention has bee n focuse d on the ef fec ts that gr oups, or member ships in groups, c onvey on the a g e nc y of ind ivi du a ls. Whe n th e re is t he la c k o f d ist inc tio n b e tw e e n in div idu a ls and populations, and a tendenc y to view populations as the sum of the individuals that comprise them; the pr actice is termed biomedica l individualis m (Krieg er 1994) Any approa ch to under standing HI V-positive mothers and the decisions they make a s individuals mus t be rooted in a n understanding of the c ontext ual eleme nts of the communities that both support and constrain them. He alth and beha vior dec isions made by these wome n reg arding both themselves and their children must also be e valuated with this knowledge prior to labeling them unrea chable or noncompliant. Eff ective interventions will not be derive d and implemented without an unde rstanding of the environment a nd constructs unde r which individuals oper ate. Of ten survivalnot healthis the g oal of these women; hea lth decisions will alway s be subjug ated by s u rv i v al d ec i s i o n s Un d er s t an d i n g t h e c o n t ex t o f d ec i s i o n s wi l l h el p ch an ge e n v i r o n m e n t s t h a t c o n s t r a i n i n d i v i d u a l a g e n c y. Suggestions for Furt her Re searc h Th is s tud y so ug ht t o d e te rm ine if mot he rs a nd inf a nts ha ve dif fe re nt h e a lth outcomes during the per inatal and postnata l periods. I t measure d whether they experience d structura l barrie rs and f acilitators to ca re, a nd tried to deter mine if these or other e x posures e x plained some of the differ ence s in the outcomes they experience d. Certainly this work has demonstra ted that the hea lth outcomes measure d differ signific antly betwee n postpartum mothers and both g roups containing their childre n (d y a d a nd inf a nt) I t a lso ind ic a te s th a t po stp a rt um m oth e rs a re muc h mo re lik e ly to encounte r bar riers, a nd much less likely to encounter fac ilitators. The inverse is true for

PAGE 82

71 dy a ds a nd inf a nts th e y a re muc h mo re lik e ly to e xpe ri e nc e fa c ili ta tor s a nd hig hly un lik e ly to e nc ou nte r a ny ba rr ie rs Th is r e se a rc h s ug g e sts tha t th e se fi nd ing s a re a na tur a l r e su lt o f a ssi g nin g bla me a nd sti g ma A n e thi c a l or ie nta tio n a rg ue s f or inc lus ion of ma te rn a l he a lth a s a n o utc ome measure in future studies. L eft una nswere d is the question of whethe r or not equa l fac ilitation of maternal hea lth care would in fac t lead to improved mate rnal outcome s. Ho we ve r, thi s r e se a rc h s ug g e sts tha t H I Vpo sit ive mot he rs wo uld g o to a ny le ng th t o pr ote c t th e he a lth of the ir ba bie s. Stu die s th a t in tr od uc e int e rv e nti on s to inc re a se ma te rn a l r ole e ff ic a c y ma y sh ow mot he rs the imp or ta nc e of su rv ivi ng to s e e the ir children thrive Expa ns ion of the c ur re nt p e ri na ta l pr e ve nti on pr oto c ol t o in c lud e ma te rn a l he a lth outcomes as mea sures of suc cess may prompt more interve ntions. The cre ation of fede ral re quirements to link infant and mate rnal outcome s at fede rally funded c linics w o u l d a l s o i m p r o v e m a t e r n a l h e a l t h a t t h e c l i n i c l e v e l E d u c a t i o n a l e f f o r t s a n d s u r v e ys direc ted at provide r attitudes may provide a dditional insi g ht into dis crimination at the clinic level. And importa ntly qualitative studies are neede d to describe how women experience the HI V trea tment sy stem, and to interpre t how the messag es g iven ar e hea rd. Fur ther qua litative study of the role women fe el they play in the lives of their ba bies wo uld a dd tr e me nd ou sly to a mixe d me tho ds de sig n th a t in c lud e d a n in te rv e nti on to incre ase the ir role e ffica cy and to dec rea se their leve ls of depre ssion.

PAGE 83

72 APPENDIX A DATA COLLECTION INSTRUMENT

PAGE 84

73 APPENDI X B EXPECTED VAL UES FOR DAT A COL L ECTI ON Mate rnal Age at Tim e of Delivery 14 y ear s to 18 y ear s of ag e 19 y ear s to 21 y ear s of ag e 22 y ear s to 24 y ear s of ag e Ov er 2 4 y ea rs o f a ge Mate rnal Race Afric an Amer ican Caribbea n I slander Hi sp a nic Cauca sian Other P ill B urden Infant dose is four times daily for six weeks, then two times daily three day s a we ek until si x months of ag e or until instructed to stop Ad ult (1998-1999) dose is 3 tabs three times a day and 1 tab two times a da y Ad ult (2000-2004) dose is five tabs twice a day Structural Barr iers Re qu ir e me nt t o o bta in o wn c a se ma na g e r p ri or to t ra ns fe r o f c a re fr om H RO B to HUG Me clinic Re qu ir e me nt o f a dd iti on a l in ta ke pr oc e ss p ri or to t ra ns fe r o f c a re fr om H RO B to HU G M e c lin ic I nsuranc e cove rag e that prohibits mother or infa nt from re ceiving primary and sp e c ia lty c a re a t th e sa me sit e Structural F acilitators Automatic transf er of paper work without intake a ppointment TO PWA o r s oc ia l w or k a ssi sta nc e to a rr a ng e tr a ns po rt a tio n to a pp oin tme nts a nd to assist with financial pape rwork ART medication dispensed instea d of pre scription g iven Nurse spe cialist and social wor ker a ssigne d to attend clinic, follow prog ress a nd provide tea ching Lab Valu es Materna l HI V RNA PCR value (viral load) is stable or fa lling Materna l CD4 count (immune sy stem) is stable or incr easing I nfant HI V DNA PCR remains neg ative I nf a nt r e ma ins ne g a tiv e fo r P CP

PAGE 85

74 Re c om m e nde d P r im ar y a nd Subs pe c ial ty Ca r e Sc he dule Adult postpartum HI V clinic visits scheduled a nd maintained at l e as t e v e r y 9 0 d a ys Pr e na ta l vi sit s k e pt a c c or din g to p e ri od ic ity re c omm e nd e d in c ha rt on pr ior vis its I nfant HI V testing and tre atment visits performe d within one month of rec ommended time o DN A P C R # 1 p er fo rm ed b y 4 we ek s o f a ge o DN A P C R # 2 p er fo rm ed b y 8 we ek s o f a ge o Ba ct ri m o b t ai n ed b y 1 0 we ek s o f a ge o DN A P C R # 3 p er fo rm ed b y 5 m o n t h s o f a ge o E I A p er fo rm ed b y 1 9 m o n t h s o f a ge I nfant primar y car e per formed w ithin one month of recommende d time, whether at HU G M e or oth e r p e dia tr ic sit e o 2 m o n t h E P S DT p er fo rm ed b y 3 m o n t h s o f a ge o 4 m o n t h E P S DT p er fo rm ed b y 5 m o n t h s o f a ge o 6 m o n t h E P S DT p er fo rm ed b y 7 m o n t h s o f a ge o 1 2 m o n t h E P S DT p er fo rm ed b y 1 3 m o n t h s o f a ge o 1 8 m o n t h E P S DT p er fo rm ed b y 1 9 m o n t h s o f a ge F ederal P overt y Guidelines Yea r (one person, two pe rson, three person, f our per son, eac h addtl per son) 1 2 3 4 +ea 1998 (8050, 10850, 13650, 16450, + 2800 ) 1999 (8240, 11060, 13880, 16700, + 2820 ) 2000 (8350, 11250, 14150, 17050, + 2900 ) 2001 (8590, 11610, 14630, 17650, + 3020 ) 2002 (8860, 11940, 15020, 18100, + 3080 ) 2003 (8980, 12120, 15260, 18400, + 3140 ) 2004 (9310, 12490, 15670, 18850, + 3180 )

PAGE 86

75 REFERENCES Ander sen, R. M. (1995). Revisiting the behavior al model and a cce ss to medical ca re: D o e s i t m a t t e r ? Journal of Health and Social Behav ior, 36 1-10. Andre ws, S., W illiams, A. B., & Neil, K. (1993) The motherchild rela tionship i n the HI V-1 positive fa mily Image : Journal of Nursing Scholarship, 25 (3), 193198. Ay nalem, G., Me ndoza, P., Frede rick, T., & Mascola, L (2004). Who and why ? HI Vt e s t i n g r e f u s a l d u r i n g p r e g n a n c y: Im p l i c a t i o n f o r p e d i a t r i c H IV e p i d e m i c d i s p a r i t y. AIDS and Be havior, 8 (1), 2531. Ba su, S. (2004). AI DS, empire, a nd public health beha viorism. In te rn ati on al J ou rn al o f Health Serv ices, 34 (1), 155167. B e a tty L .A ., Whe e le r, D. & Ga ite r, J. (2 00 4) H I V p re ve nti on fo r A fr ic a n A me ri c a ns : Current a nd future dir ections. Journal of Black Psy chology, 30 (1), 4058. Be aucha mp, T. L ., & Childress, J. F. (1994). Principles of biomedical ethics (4th ed). Ne w Y or k: O xfor d U niv e rs ity Pr e ss. Be lzer, M. E., Fuc hs, D. N., L uftman, G. S., & Tucke r, D. J. (1999). Antiretrovira l adher ence issues among HI V-positive adolesc ents and y oung adults. Journal of Adolesce nt Health, 25 (5), 316319. Br owne, A. J. (2001). The inf luence of libera l politi cal ideolog y on nursing scienc e. Nursing Inquiry, 8 (2), 118129. Bunting S. M. (1996). Sources of stigma a ssociated with women w ith HI V. Di v e rsi ty in Health Care, 19 (2), 6474. Carg ill, V. A., Stone, V. E., & Robinson, M. R. (2004). HI V trea tment in Africa n Americ ans: Challeng es and oppor tunities. Journal of Black Psy chology, 30 (1), 24-39. Centers for Disease Control and Prevention (CDC) (2004) HIV/AID S surveillance re po rt (Vol. 16). Atlanta : Author. Centers for Disease Control and Prevention (CDC) (2005) W ISQARS le ading causes of death reports, 1999 2003. Retrieved F ebrua ry 25, 2006, from htt p:/ /w e ba pp a .c dc .g ov /Sa sw e b/n c ipc /le a dc a us 10 .h tml

PAGE 87

76 Clarke, L (1999). The person in abor tion. Nursing Ethi cs 6 (1), 3747. Dunbar H. T., Mueller C. W ., Medina, C., & Wolf, T. (1998). Psy cholog ical and spiritual gr owth in women living with HI V. Social W ork, 43 (2), 144155. Eb ra him S. H. A nd e rs on J. E ., Wei dle P. & Pur c e ll, D. W. ( 20 04 ). Ra c e /E thn ic disparities in HI V testing and knowledg e about tre atment for H I V/AI DS: United States, 2001. AIDS Patient Care and STDs, 18 (1), 2733. Fa rmer, P. (2003) Pathologies of power: Health, human rights, and the new war on the poor B e rk e le y : U niv e rs ity of Ca lif or nia Pr e ss. FuChang Tsai, D. (2001) How should doctors a pproac h patients? A Confucian ref lection on per sonhood. Journal of Medical Ethics, 27 (1), 4451. Glanz, K., Rimer, B. K ., & L ewis, F M. (Eds.) (2002) He alt h b e ha v ior an d h e alt h education (3 rd e d. ). Sa n F ra nc isc o: Jos se y -B a ss. Goffma n, E. (1963). Stigma: Notes on the managem ent of spoiled identity. New Y ork: Simon & Schuster. Hall, W. A., & Caller y P. (2001). Enhanc ing the rig or of g rounded the ory : I ncorpor ating ref lexi vity and re lationality Qualitative Health Rese arch, 11 (2), 257273. Har ris, J (1999). The conce pt of personhood a nd the value of life. Ke nn e dy In sti tut e of Ethics Journal, 9 (4), 293308. Henr y J Kaiser Fa mily Founda tion. (1999). Hearing their voice s: A qualitative researc h stu dy on HI V t e sti ng an d h igh e rris k te e ns Me nlo Pa rk CA : A uth or Re tr ie ve d A pr il 10 2 00 6, fr om http:// www.kf f.org /y outhhivstds/ 1492-index.cfm hooks, b.(2003). Rock my soul: Blac k people and self-estee m New Y ork: Atria. I ng ram, D., & Hutchinson, S. A. (1999a) HI V-positive mothers and stig ma. He alt h Ca re for W omen I nternational, 20 (1), 93104. I ng ra m, D ., & Hu tc hin so n, S. A ( 19 99 b) D e fe ns ive mot he ri ng in H I Vpo sit ive mot he rs Qualitative Health Rese arch, 9 (2), 243258. Kane D., & Thomas, B (2000). Nur sing a nd the F wor d. Nursing Forum, 35 (2), 17-24. K r i e g e r N ( 1 9 9 4 ) E p i d e m i o l o g y a n d t h e w e b o f c a u s a t i o n : h a s a n y o n e s e e n t h e s p i d e r ? Social Scienc e & Medic ine, 39 (7), 887903.

PAGE 88

77 L y nch, J. W ., Kaplan, G A., & Salonen, J T. (1997). Why do poor people behave p o o r l y? V a r i a t i o n i n a d u l t h e a l t h b e h a v i o u r s a n d p s yc h o s o c i a l c h a r a c t e r i s t i c s b y stag es of the soc ioeconomic life course Social Scienc e & Medic ine, 44 (6), 809819. Mc Na ir L D ., & Pr a the r, C. M ( 20 04 ). Af ri c a n A me ri c a n w ome n a nd AI DS: F a c tor s influencing risk and re action to HI V disease Journal of Black Psy chology, 30 (1), 106-123. Mellins, C. A., Ehrhardt, A. A., Ra pkin, B., & Have ns, J F. ( 2000). Psy chosocia l fac tors associate d with adaptation in HI V-infe cted mother s. AIDS and Be havior, 4 (4), 317-328. Murphy D. A., Wilson, C M., Durako, S. J., Muenz L R., & B elzer, M. (2001). Antiretrovira l medication adhe renc e among the REACH HI V-infe cted a dolesce nt cohort in the U.S.A. AIDS Care, 13 (1), 2741. Nazroo, J. Y. (2003). The struc turing of ethnic ine qualities in health: Economic position, rac ial discrimination, and ra cism. American J ournal of Public Health, 93 (2), 277285. National I nstitut es of He alth/National I nstitut e of A llerg y and I nfec tions Disease (1999) Reassuring finding s about infants exposed to zi dovudine NI H N e ws R e le as e J an ua ry 12, 1999. Retrieved Ma rch 15, 2005, f rom htt p:/ /w ww 3. nia id. nih .g ov /ne ws /ne ws re le a se s/1 99 9/p a c tg 21 9. htm Pin c h, W. J. ( 19 94 ). Ve rt ic a l tr a ns mis sio n in HI V i nf e c tio n/A I DS: A f e min ist perspe ctive. Journal of Advance d Nursing, 19, 36-44. Poindex ter, C. C., & L insk, N. L (1999). HI V-re lated stig ma in a sample of HI Vaff ecte d older fe male Af rica n Americ an ca reg ivers. Social W ork, 44 (1), 4654. Roge rs, A. S. (2001). HI V re sear ch in Amer ican y outh. J ou rn al o f A do le sc e nt H e alt h 29 (3S), 1-4. Roge rs, A. S., Miller, S., Murphy D. A., Ta nney M., & F ortune, T. ( 2001). The T REAT (thera peutic re g imens enhanc ing a dhere nce in tee ns) prog ram: Theor y and pr e lim ina ry re su lts Journal of Adolesce nt Health, 29 (3S), 30-38. Smit h, L ., Fe aster, D J ., Prado, G., Ka min, M., Blaney N., & Sz apocznik, J (2001). The psy chosocia l functioning of HI V+ a nd HI VAf rica n Americ an re cent mother s. AIDS and Behav ior, 5 (3), 219231. Speige l, K., & Schr imshaw, E. W. (2001). Women with HI V/AI DS describe d sever al forms of positive cha ng es re sulting fr om their illness. Evidenc e-Base d Nursing, 4 (4), 126-129.

PAGE 89

78 Steinhart, C., Orrick, J. J ., & Simpson, K., Eds. (2002). HIV/AID S primary care guide. Gainesville: Univer sity of F lorida. Tay lor, B. ( 2001). HI V, stigma and hea lth: I nteg ration of theor etical c oncepts. Journal of Advance d Nursing, 35 (5), 792798. Todd, T. J ., & Miller, S. (2000) I ncre asing adher ence to antiretrovira l therapy in adolesc ent patients with HI V via novel c ompliance tools. J ou rn al o f P e dia tri c Pharmacy Practice, 5 (5), 229234. Toni, T. D., Masquelier B., L azaro, E., Dor e-Mba mi, M., BaGomeis, F. O ., & T e a D i o p Y ( 2 0 0 5 ) C h a r a c t e r i z a t i o n o f N e v i r a p i n e r e s i s t a n c e m u t a t i o n s a n d H IV ty pe 1 subty pe in women f rom Abidjan af ter Ne virapine sing le dose prophy laxi s of HI V ty pe 1 mother to c hild transmission. AI DS Re se ar c h a nd Hu ma n R e tro v iru se s, 21 (12), 10311034. Tross, S. (2001). Women at heter osexual risk for HI V in inner-c ity New Y ork: Reac hing the har d to rea ch. AIDS and Be havior, 5 (2), 131140. Why te, J ., Standing, T., & Madig an, E. (2004) The re lationship between H I V-re lated knowledg e and sa fe se x ual beha vior in Afric an Amer ican wome n dwelling in the rura l southeast. JANAC, 15 (2), 5158. Wil kinson, R. G. (1999). Putting the picture toge ther: Prosperity redistribution, health, and we lfare I n M. Marmot & R. G. Wi lkinson (Eds.), So c ial de te rm ina nts of h e alt h (p p. 26 526 6) O xfor d, UK : O xfor d U niv e rs ity Pr e ss. Wil ley C., Redding, C., Staffor d, J ., Gar field, F ., Geletko, S., Fla niga n, T., Melbourne K., Mitty J ., & Ca ro, J. J (2000). Stag es of c hang e for adher ence with medication reg imens for c hronic disea se: Deve lopment and validation of a measure Clini cal Therapeutics, 22 (7), 858872. Y o d e r S D ( 2 0 0 2 ) In d i v i d u a l r e s p o n s i b i l i t y f o r h e a l t h : D e c i s i o n n o t d i s c o v e r y. Hastings Center Report 32 (2), 2231.

PAGE 90

79 B I OG RA PHI CA L SKE TCH Pa tr ic ia Ste a rn e s Ro bin so n g ra du a te d f ro m th e Un ive rs ity of Ce ntr a l F lor ida wi th he r B a c he lor of Sc ie nc e in N ur sin g in 1 99 4. She wo rk e d a s a re g ist e re d n ur se wi th pediatric brain-injure d patients until she moved to Gainesville to pursue he r Master of Science in Nursing Whil e in Gaine sville, she worke d at Shands Hospital as a r eg istered nu rs e wi th b oth pe dia tr ic on c olo g y pa tie nts a nd la te r, in t he pe dia tr ic int e ns ive c a re un it. She ear ned her Master of Science from the Unive rsity of F lorida in 2000. Since that time, she has wor ked as a pediatric nurse pr actitioner in a n HI V clinic in Central F lorida. She also taug ht both underg radua te and g radua te nursing students at the Univer sity of F lorida while ea rning her doc torate. H er minor c ourse of study for he r doctora te was e p i d e m i o l o g y.