<%BANNER%>

Perceived Relative Access to Health Care and Feelings of Control Over Health

Permanent Link: http://ufdc.ufl.edu/UFE0021028/00001

Material Information

Title: Perceived Relative Access to Health Care and Feelings of Control Over Health
Physical Description: 1 online resource (123 p.)
Language: english
Creator: Holzworth, Amanda N
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: access, control, discrimination, evaluation, health, perceptions, qualitative, quantitative, relative
Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A mixed methods exploratory case study was conducted. Twenty-three enrollees of a community health services program were interviewed. The Human Ecological Model, Theory of Planned Behavior, and Perceived Unfairness Model were used to elicit information about the participants perceptions of access to health care, relative access to health care, discrimination, and feelings of control. Many participants had relatively low access to health care, yet felt a lot of control over their health. Despite this, when participants were asked about others, they attributed little or no control to those who have the least access, and a lot or complete control to those with the most. Research should continue to explore relationships among social hierarchies, perceptions and emotions, and health outcomes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amanda N Holzworth.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Wilken, Carolyn S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021028:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021028/00001

Material Information

Title: Perceived Relative Access to Health Care and Feelings of Control Over Health
Physical Description: 1 online resource (123 p.)
Language: english
Creator: Holzworth, Amanda N
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: access, control, discrimination, evaluation, health, perceptions, qualitative, quantitative, relative
Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A mixed methods exploratory case study was conducted. Twenty-three enrollees of a community health services program were interviewed. The Human Ecological Model, Theory of Planned Behavior, and Perceived Unfairness Model were used to elicit information about the participants perceptions of access to health care, relative access to health care, discrimination, and feelings of control. Many participants had relatively low access to health care, yet felt a lot of control over their health. Despite this, when participants were asked about others, they attributed little or no control to those who have the least access, and a lot or complete control to those with the most. Research should continue to explore relationships among social hierarchies, perceptions and emotions, and health outcomes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amanda N Holzworth.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Wilken, Carolyn S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021028:00001


This item has the following downloads:


Full Text

PAGE 1

1 PERCEIVED RELATIVE ACCESS TO HEAL TH CARE AND FEELINGS OF CONTROL OVER HEALTH By AMANDA N. HOLZWORTH A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Amanda N. Holzworth

PAGE 3

3 For Mom, Nanny, and Papa Youve always been my greatest supporters, and Ill always be your Mander

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank a ll of those who directly contributed to the su ccess of this project: Carolyn Wilken, Eboni Baugh, Jason Storch, Ca ndice King, Alachua Count ys CHOICES Health Services, and a benefactor who provided most of the funding for the participants monetary incentive. I am indebted to the participants of this project, without whom it could not have happened. Their willingness to share their expe riences and thoughts strengthened my resolve to not only to complete the project, but also to share it with the public. I do not have the words to properly thank Br endan for all he has done for listening to seemingly endless hours of shop talk, for alwa ys making me smile, for reminding me of the bigger picture, and for being my best friend and life partner. There are also many others I am deeply grat eful for for their support, patience, and unending faith in me. In these regards I woul d like to recognize my parents, Nanny and Papa, Steven, Judith, and Amanda.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......10 LIST OF TERMS.................................................................................................................. .........11 ABSTRACT....................................................................................................................... ............13 CHAPTER 1 INTRODUCTION..................................................................................................................14 Purpose........................................................................................................................ ...........19 Limitations.................................................................................................................... ..........20 2 LITERATURE REVIEW.......................................................................................................22 Access......................................................................................................................... ............22 Relationship Between Health and Feelings of Control...........................................................24 Theoretical Framework.......................................................................................................... .25 Theories of Health Behavior...................................................................................................26 Health Disparities............................................................................................................. ......31 Race/Ethnicity.................................................................................................................31 Sex............................................................................................................................ .......33 Income......................................................................................................................... ....34 Differences in Measuring Income...................................................................................36 Socioeconomic Status......................................................................................................36 Insurance...................................................................................................................... ....37 Discrimination.................................................................................................................37 Relative Social Standing..................................................................................................40 Summary of Health Disparities.......................................................................................41 3 METHODS........................................................................................................................ .....43 Research Site Selection........................................................................................................ ...43 Research Design................................................................................................................ .....44 Sample Selection.............................................................................................................44 Instrumentation................................................................................................................44 Administration.................................................................................................................46 Analysis....................................................................................................................... ...........47 Concerns/Critiques............................................................................................................. ....48

PAGE 6

6 4 RESULTS........................................................................................................................ .......50 Response Rate and Demographics of Sample........................................................................50 Quantitative Analysis.......................................................................................................... ....51 Perceived Access to Health Care Services......................................................................51 Access chart..............................................................................................................51 Important access chart factors..................................................................................51 Ranked important access factors..............................................................................52 Access factors they have now..................................................................................52 Perceived Relative Access to Health Care Services........................................................53 Perceived Access.............................................................................................................54 Level of Control..............................................................................................................54 Perceived Discrimination Scenario.................................................................................55 CHOICES Health Questionnaire....................................................................................57 Quantitative Responses to Research Questions......................................................................61 RQ1. How do people with low incomes asso ciate feelings of control over their health with having access to health care services?......................................................61 RQ1a. Is perceived access to health care services associated with any form of discrimination?.............................................................................................................61 RQ1b. Is perceived relative access to health care services associ ated with feelings of control over health?.................................................................................................61 RQ2a. Are feelings of control associat ed with selected questions from the CHOICES questionnaire?...........................................................................................62 RQ2b. Is perceived access associated with selected questions from the CHOICES questionnaire?..............................................................................................................63 RQ2c. Is discrimination associated with selected questions from the CHOICES questionnaire?..............................................................................................................63 RQ2d. Is perceived relative access associat ed with selected questions from the CHOICES questionnaire?............................................................................................64 Summary of Quantitative Analysis.................................................................................64 Qualitative Analysis........................................................................................................... .....65 Themes: The Human Ecological Model.........................................................................67 Microsystem.............................................................................................................67 Mesosystem..............................................................................................................68 Exosystem................................................................................................................69 Macrosystem............................................................................................................70 Themes: The Theory of Planned Behavior......................................................................71 Control beliefs..........................................................................................................71 Power beliefs............................................................................................................72 Themes: The Perceive d Unfairness Model.....................................................................73 Perceived unfairness.................................................................................................74 Identity.....................................................................................................................76 Emotion....................................................................................................................77 Coping behavior.......................................................................................................78 Responses to Utilizing CHOICES Se rvices and Thoughts for Lawmakers...................79 Qualitative Responses to Research Questions........................................................................80

PAGE 7

7 RQ1. How do people with low incomes asso ciate feelings of control over their health with having access to health care services?...............................................80 RQ1a. Is perceived access to health care services associated with any form of discrimination?.....................................................................................................80 RQ1b. Is perceived relative access to h ealth care services associated with feelings of contro l over health?............................................................................81 RQ2. Do any of the above concepts, c ontrol over health, perceived access to health care, discrimination in health care, and perceived relative access to health care, show patterns of associa tion with responses from the CHOICES Questionnaire?......................................................................................................81 Summary of Qualitative Analysis...................................................................................81 5 FUTURE RESEARCH AND POLICY IMPLICATIONS.....................................................88 Review of the Results.......................................................................................................... ...88 Research Question 1........................................................................................................89 RQ1. How do people with low incomes asso ciate feelings of control over their health with having access to health care services?...............................................89 RQ1a. Is perceived access to health care services associated with any form of discrimination?.....................................................................................................89 RQ1b. Is perceived relative access to h ealth care services associated with feelings of contro l over health?............................................................................89 Research Question 2........................................................................................................89 RQ2a. Are feelings of control associat ed with selected questions from the CHOICES questionnaire?....................................................................................89 RQ2b. Is perceived access associated w ith selected questions from the CHOICES questionnaire?....................................................................................90 RQ2c. Is discrimination associated with selected questions from the CHOICES questionnaire?....................................................................................90 RQ2d. Is perceived relative access associat ed with selected questions from the CHOICES questionnaire?.....................................................................................90 Data in the Context of Theories..............................................................................................90 The Human Ecological Model.........................................................................................90 The Theory of Planned Behavior....................................................................................92 The Perceived Unfairness Model....................................................................................93 Future Research................................................................................................................ ......94 Methodological Considerations.......................................................................................94 Human Ecological Model................................................................................................94 The Theory of Planned Behavior....................................................................................95 The Perceived Unfairness Model....................................................................................95 Other Areas of Future Research......................................................................................96 Implications and Recommendations fo r Alachua Countys CHOICES Health Services....................................................................................................................... .97 Implications and Recommendations for Health Care Policy..........................................98 Conclusion..................................................................................................................... ..99

PAGE 8

8 APPENDIX A CONSTRUCT/THEORY/PURPOSE CHART....................................................................101 B CONCEPT PATTERN.........................................................................................................104 C INTERVIEW SCHEDULE AND KEY...............................................................................105 D ACCESS CHART.................................................................................................................111 E PERCEIVED RELATIVE HEAL TH CARE ACCESS SCALE.........................................114 F SAMPLE INTRODUCTORY LETTER..............................................................................115 LIST OF REFERENCES.............................................................................................................116 BIOGRAPHICAL SKETCH.......................................................................................................123

PAGE 9

9 LIST OF TABLES Table page A-1 Construct/theory/purpose chart........................................................................................101

PAGE 10

10 LIST OF FIGURES Figures page 4-1 Distribution of par ticipants by age group..........................................................................83 4-2 Distribution if particip ants by race and ethnicity...............................................................83 4-3 Distribution of access chart factor s that were marked as important..................................84 4-4 Distribution of participants in each quartile of the top fi ve factors they indicated they had which facilitate th eir access health care......................................................................85 4-5 Distribution of participants in each qu artile of the top five access factors they indicated they had if health insurance is taken away.........................................................86 4-6 Distribution of participants perceived relative ac cess to health care re lative to others in Alachua County.............................................................................................................87 5-1 Ecological view of factors affecting CHOICES enrollment..........................................100 B-1 Pattern of potential re lationships among concepts...........................................................104 D-1 Factors that contribute to a participan t accessing health care, how the participant ranks these factors, and whet her they have the factors....................................................111

PAGE 11

11 LIST OF TERMS CHOICES: Community Health Offering Innova tive Care and Educational Services: A community-based program that provides a link between primary health care services and workers with low-income. The basic eligibility criteria are as follows: must be a U.S. citizen, must be a resident of Alachua County, must not be eligible for benefits through the Veterans Association or Medicaid, must not have health insurance coverage (excludes Medicare for seniors), must not be offered affordable health insurance through employer (a formula is used), annual inco me must be between 100% of the federal poverty level, and must be working at least an average of 25 hours per work (exclusion for seniors) (Indigent Care and Trauma Center Surtax (2004); CHOICES internal documents) CONTROL: Over health for this research the de finition is left intentionally broad to allow for each participants definition; given the context of the questions regarding control over health, it could be implied to mean how much influence a person exerts over their current health status, how much influe nce a person feels they may be able to exert in the future, or how much influence a person feels they have to change their health status, and each of these could in turn be applied to health related behaviors (Adapt ed from Rotter, 1966) CONTROL BELIEFS: From the Theo ry of Planned Behavior; factors that impede or facilitate a behavior (Glanz et al., 1997; Terry, Gallois & McCamish, 1993) This research attempts to identify some external impediments. EXTERNAL FACTORS: F actors that are outside of the individual HEALTH CARE PROVIDER: For this research, a ny person that can give any type of health care service, i.e., a doctor, a nurse, a counselor, a dentist, a physicians assistant, a specialist, or a therapist HEALTH CONCERN: For this research, any state of health or health-related behavior that an individual identifies as important to them INTERNAL FACTORS: Factors co ntained within the individual LOW-INCOME: For this research, the income gui delines for enrollment in CHOICES is used; between 100% federal poverty leve l (CHOICES internal documents) MEDICAL EMERGENCY: For this research, a health problem that necessitates immediate medical attention PERCEIVED ACCESS TO HEALTH CARE: Whether a person thinks they can gain entry to health care services; Represents Perceived Behavioral Control; Combination of Power Beliefs and Control Beliefs given a specific context, does a person feel they can carry out a specific behavior (Adapted from Azjen & Madden, 1986 and Aday, Andersen & Fleming, 1980)

PAGE 12

12 PERCEIVED LEVEL OF ACCESS TO HEALTH CARE: A measure that compares the resources the participant identifies in what th ey need for access to health care with the resources the participant actually has (In ferred from Aday, Andersen & Fleming, 1980) PERCEIVED RELATIVE ACCESS TO HEALTH CARE : The participants rating of their own resources that contribute to health care access when they compare them with the resources of others in Alachua County (In ferred from Adler, Epel, Castellazzo & Ickovics, 2000) POWER BELIEFS: From the Theory of Planned Behavior; beliefs regarding the weight of each factor that impedes or facilitates a beha vior (Glanz et al., 1997; Terry, Gallois & McCamish, 1993) PROBLEM WITH DRUGS OR ALCOHOL: For this research, a point where the use of drugs or alcohol is clinically signifi cant. This may include the im pairment of daily activities, harm being done to the user or those around th em, and/or a situati on in which the user wishes to stop consuming but cannot.

PAGE 13

13 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PERCEIVED RELATIVE ACCESS TO HEAL TH CARE AND FEELINGS OF CONTROL OVER HEALTH By Amanda N. Holzworth August 2007 Chair: Carolyn S. Wilken Major: Family, Youth, and Community Sciences A mixed methods exploratory case study was conducted. Twenty-three enrollees of a community health services program were interv iewed. The Human Ecological Model, Theory of Planned Behavior, and Perceived Unfairness Model were used to elicit information about the participants perceptions of access to health car e, relative access to hea lth care, discrimination, and feelings of control. Many pa rticipants had relatively low acce ss to health care, yet felt a lot of control over their health. Despite this, wh en participants were as ked about others, they attributed little or no control to those who have the l east access, and a lot or complete control to those with the most. Research should continue to explore relationships among social hierarchies, perceptions and emotions, and health outcomes.

PAGE 14

14 CHAPTER 1 INTRODUCTION With the increasing cost of medical care and the growing number of uninsured, many people are examining and questioning the structur e of our health care system. In the United States, access to health care is inextricably tied to having health insurance, which is in turn, highly correlated with employment (Collins, Sc hoen, Colasanto & Downey, 2003). According to the Kaiser Family Foundation, 54% of the populat ion of the United Stat es is insured through their employer, 26% through Medicaid or anothe r public program, and 5% are privately insured, leaving 16% uninsured ( StateHealthFacts.org 2004). In Florida, the picture is a bit dimmer : 47% of the population has employer-sponsored insurance, 29% have insurance through Medicaid or another p ublic program, 5% are privately insured, and 19% of the population is uninsured ( Individual State Profiles 2004) This translates to over 45.8 million people in the United States th at do not have health insurance, and over 3.2 million of them live in Florida. Not only does Fl orida have a higher than national average rate of uninsured people, it ties with four other st ates to rank fourth hi ghest in the nation ( 50 State Comparisons 2004). It is known that national and stat e-level problems manifest themselves at the local level. In Alachua County, a largely ru ral county with an urban center, there are about 38,000 people without health insura nce (A. J. Campo, presentation, CHOICES: Improving Access to Health Care Services 2005). Even if one is employed, three conditions mu st be met to have insurance through an employer: ) the workers company must offer insu rance; 2) the worker has to be eligible for the insurance offered; and 3) the worker has to be able to afford the insurance he or she is eligible for (Collins, Schoen, Colastanto, & Do wney, 2003, p. 2). Regarding the first condition, among people who are uninsured in the United Stat es, 60% of them are not offered insurance by

PAGE 15

15 their employer, 22% are offered insurance but do not participate, an other 11% are offered insurance but are ineligible, and 7% do not know if their employer offers insurance (Collins et al., 2003). Collins and colleagues (2003) conclude d that workers with low wages fare worse than other workers on all three conditions. Being uninsured in America is a large probl em for a number of reasons. People without health insurance, including those who alternate between having insurance and not having it, have higher morbidity and mortality rates than people who have consistent coverage. Many people who do not have health insurance delay looking for health care services, or do not seek them at all ( The Uninsured and Their Access to Health Care 2005). Many do not fill prescriptions because of their expense. Using path analysis Wa n and Soifer (1974) were able to establish the relative importance of various factors that influe nce utilization of health care services. Four factors had a direct influence on vi siting a doctor: 1) being in poor health, 2) as a response to an illness, 3) having health insurance, and 4) the average cost per visit.1 Tying all of this together, we can see that having health insurance not onl y contributes to a pers on accessing health care services, but also utilizing them Given the structure of our h ealth care system, the health of people without insurance suffers. It seems that the deleterious effects of not having health insurance might be partially mitigated if health ca re services could be re gularly utilized without insurance. In an ideal health care system, a health care in frastructure would be in place and one might take utilization for granted: people would recogn ize when they needed health care and know, or perceive, that they could access the health care they needed. However, there is no an ideal 1 Medicaid and Medicare were distinct from health insura nce in this analysis. The researchers found that participation in these programs did not have either a direct or indirect effect on using a doctors services (Wan & Soifer, 1974).

PAGE 16

16 system, and there is a difference between utilization, or realized access, and potential access, which will be discussed more fully in the lite rature review (Aday, Andersen, & Fleming, 1980). Our societys emphasis on individuality and responsibility also leads the debates surrounding health care. Individu als hold the keys to healthy liv ing; eating a proper diet and exercising are especially championed. Various programs exist, both in the private and public sectors, to raise awareness about healthy living. These programs vary in their focus. Some simply provide nutrition information while othe rs provide intensive one -on-one training about nutrition, exercise and managing stre ss. Others provide informati on about health screenings and immunizations and may also provide thes e services for free or reduced costs. There are a few widely-held assumptions behi nd many of these programs: 1) if people are provided information about the morbidity and mo rtality risks of living an unhealthy lifestyle, (i.e., not exercising regularly, not managing stress not eating properly and not receiving regular physical exams), they will change their unhealthy be haviors to avoid those risks, and 2) if people are provided information about the benefits of having a healthy lifestyle they will change their unhealthy behaviors to receive those benefits. Some examples of these include: Advice from Aetnas Living Well (2001) web site where the h eaders are Your Health, SelfManagement, and Wellness Products and Disc ounts. Particularly interesting is the description of Self-Management, How to achieve life's greates t giftthe best possible health. Topics include weight control, eating right, stayin g fit, overcoming depression and more ( Living Well 2001). Blue Cross and Blue Shield of Florida, Inc. also provides advice: Good health is good living, and th e more you know, the easier it is to stay healthy. That's why Blue Cross and Blue Shield of Florida provides members with helpful resources to keep them informed about health issues. Th ese programs, services and links are another way that we add value to our members' pr emium dollars and why Blue Cross and Blue Shield of Florida is a company to keep (Health and Wellness, 2004).

PAGE 17

17 This phenomenon goes beyond insurance companie s. The American Cancer Society, the American Diabetes Association, a nd the American Heart Association joined to create the website Everydaychoices.org (n.d.) where anyone can view information focused on the health benefits and risks associated with diet/ nutrition, exercise, a nd getting medical attention. For example, An Ounce of Prevention Not smoking (or quitting if you already smoke), making wise food choices, getting enough physic al activity, and main taining a healthy weight can help you reduce your disease risk, feel better and li ve longer. Theres a lot you can do on your own to prevent cancer, diabetes, heart diseas e and stroke. And your doctor can help you recognize unhealthy aspect s of your lifestyle. Your doctor is a great source of advice on how to make the right choices for li felong health (See Your Doctor, n.d.). From this site there is also a link to the Amer ican Diabetes Associati ons A Virtual Grocery Store Tour (n.d.). This site is promoted by the Centers for Medi care and Medicaid Services in their Staying Healthy: Medicares Preventive Services (2004). This is not to say that having knowledge is not an essential part of the behavioral change, but this simple, linear logic doe s not account for the reciprocal interaction of people and their environments. There are many fact ors that contribute to a person s behavior and changing that behavior. According to th e Transtheoretical Model, (Prochaska, 1979) providing information, or consciousness raising, is necessary to behavior change, but it is not sufficient (Glanz, Lewis and Rimer, 1997, p. 63). Research has supported ot her processes that are also needed; these include: self-reevaluation, envir onmental evaluation, helping relationships, counterconditioning, contingency management and soci al liberation. According to th e Theory of Planned Behavior (Ajzen, 1986), additional factors are: attitude toward the beha vior, subjective norms, intention, and perceived behavioral control. Research using the Theory of Planned Behavior has examined decision-making behavior in a number of settings (Some examples are exercise, Zizzi, Keeler & Watson, 2006; nutrition, De Vet, De Nooijer, De Vries & Brug, 2005; drug use, Naar-King,

PAGE 18

18 Wright, Parsons, Frey, Templin & Ondersma, 2006) and for an evaluation of its use as a health behavior intervention Bridle, Riemsma, Pa ttenden, Sowden, Mather, Watt & Walker, 2005. The focus of the present research draws from the control portion of a couple of theories that will be presented here and in Chapter 3. In the Theory of Planned Behavior (Glanz, Lewis & Rimer, 1997), perceived behavioral control is composed of two concep ts: control and power beliefs. Control beliefs can be either internal or external to a person; they are the perceptions that a person has about what facilitates or impedes the behavior (G lanz et al., 1997; Terry, Gallois & McCamish, 1993). Powe r beliefs are the perception of how much each facilitator or impediment affects whether or not the behavior is performed (Brown, 1999; Glanz et al., 1997; Terry, Gallois & McCamish, 1993). Control beliefs and power beli efs combine to yield the concept of perceived behavioral control, which is the perceived amount of control a person has in carrying out a specific behavior in a specific environment (Glanz et al., 1997) So, for example, a person who wishes to quit smoking may have control beliefs surrounding their physical addiction to nicotine and their daily routines that incorporate the behaviors of smoking, both internal impediments. Yet they may also perceive pendi ng social approval for success, an external fac ilitator, and possible health benefits, an internal facilitator. These control beliefs are then individua lly assigned weight, or a level of strength, which ar e the persons power beliefs. In th e balancing of these weights, one can then hypothesize about whether or not a pe rson will attempt to quit smoking. The likelihood for quitting is increased if the fac ilitators outweigh the impediments. From this we can see a recipro cal relationship between a person and their environment. On one hand there is a persons environment w ith its various facil itating and constraining factors, such as potential social approval, but there is also the potential to lose some social

PAGE 19

19 interactions from no longer engaging in smoking. And on the other hand, there is the person and their own facilitating or constraini ng factors, such as addiction, cu rrent physical health and goals for future health. A similar process unfolds with people without health insurance and/or other resources to adequately manage their health. There is interplay of the environment and the person at every level; however, there is the po ssibility that the environment restricts a persons scope of resources, both external and internal. Is it possibl e that a persons relatively low access to health care affects the amount of contro l they feel over their health? Purpose This paper will address what can be regarded as the beginning of the process to utilizing health care services2: accessing health care, or more spec ifically, perceived relative access to health care. This conception of access places emphasis on individuals experiences while keeping them in the context of their envir onment. Research has been gaining ground on understanding the relationship betw een relative income and healt h. It has also been making advances in explaining the health disparitie s that persist across races/ethnicities and socioeconomic statuses (SES). One way these av enues of research are re lated is in trying to establish the means by which people with lower soci al status have worse health outcomes. One possible explanation is through the perceived lack of control perpet uated by a lack of resources. This research will address these main research questions: 1. RQ1) How do people with low incomes associat e feelings of control over their health with having access to health care services? 1a) Is perceived access to health care se rvices associated with any form of discrimination? 2 This process will be discussed more in the Chapter 2, the Literature Review.

PAGE 20

20 1b) Is perceived relative access to health care se rvices associated with feelings of control over health? 2. RQ2) Do any of the above concepts show patterns of association with: a) perceptions of health status, b) diagnosis of any of the following conditi ons: asthma, diabetes, high blood pressure, congestive heart fail ure and depression, c) feeling of global sati sfaction with ones life, d) frequency of utilization of emergency rooms in the last two years, e) utilization of hospital inpatient services in the last two years, f) frequency of primar y health care visits, g) lapse of time since participant s last physical and dental exams, h) perceived primary barr ier(s) to utilizing he alth care services, or i) whether or not participants ha d a usual source of health care? Limitations Causal relationships cannot be es tablished in the current research because it is an exploratory case study. Knowledge of the inte rplay of the interdisciplinary concepts utilized is not to the point of allowing for a research design that c ould establish causality, such as an experiment or a longitudinal case study. The sample is limited to participants in the CHOICES program, and the research design precludes statistical generalizability. This research is limited in terms of its theoretical generalizability due to the lack of comparison group(s). This is an area for future research. Access chart does not include all possible inte rnal and external f acilitators and/or impediments. Examining emotions related to any form of discrimination or lack of control/power is beyond the scope of this research. This is a significant limitation as emotions may be a pathway for how the environment affects pe rceptions of health and/or biological functioning. Concept of control: wh ile the current study alludes to lo cus of control, it is beyond the scope of this study to identify wh ether the participants have an internal or external locus of control (Rotter, 1966). In this study, control wa s left intentionally broad to account for any

PAGE 21

21 sense of control the participant feels, regardless of its categori zation. However, this is an area for future research.

PAGE 22

22 CHAPTER 2 LITERATURE REVIEW The purpose of this research was to explore a potential relatio nship between perceptions of access to health care and perceptions of control ov er ones health. This literature review will establish the relevance and importance of this expl oratory research. First, access to health care will be discussed, followed by an examination of th e relationship between health and feelings of control. Next, the theoretical framework for this research will be reviewed, which includes descriptions of health behavior theories and an emphasis on th e Theory of Planned Behavior (Azjen & Madden, 1986). Finally, th ere will be a review of vari ous proposed causes for health disparities in the United States. Access One precursor to utilizing health care services is the perception that one could use them, or that one has access. Though the concept of access to health care is often used, there is a lack of clarity about what access means (Ch itwood, Comerford & McCoy, 2002; Lillie-Blanton & Hoffman, 2005; and Williams & Jackson, 2005, for examples). Some authors conceptualize it solely as a cost issue (For examples, Fede rman, Vladeck & Siu, 2005 and Frist, 2005), while others approach it as a multidimensional concept, including cost, language, cultural barriers, and literacy (Kennedy, 2005). One of the most inclusiv e definitions includes cost, the number of providers in the community, the travel time to th e nearest provider, if there are any language barriers, if there are any cultural barriers, if one has a regular pe rson or place to receive health care, and whether or not a person has seen a physician in the past year if they have been ill at any point, or in the past two years if they report good h ealth (Brown, Ojeda, Wyn & Levan, 2000). A widely used concept of access takes account of th e, likelihood of a usua l source of care, having an unmet medical need or postponing getting care in the past year the likelihood of an

PAGE 23

23 ambulatory medical care visit, a vi sit for a routine health problem or general exam, and a visit to a hospital emergency department (Cunningham & Hadley, 2004, p. 235). Gulliford and Morgan (2003) described two th emes of medical care access identified through reviewing previous research: 1) the poten tial or actual entry of a given individual or population group into the health car e delivery system (p. 5), and 2) gaining access refers to the initiation into the process of utili zing a service (p. 5). In other words, health care access can be characterized as accounting for such resources in the community, a persons external resources, and a history of utilizing the services. The first type of access identified by Gulliford and Morgans (2003) focuses on the health care resour ces in the community, or the lack of them. The present research will not include this type of access since the resear ch site, Alachua County, FL, USA, has a number of locations that provide medical care, including a number of hospitals. This research will use the second type of access, which focuses on utilization of the available services. However, this research will take a step back from actual utiliz ation of medical services to look at the perception of being able to use th em. This approach does not assume that once the medical infrastructure is in plac e people will use it. There are various factors that could affect the decision to use medical services; one being th e perception of being able to use them. To clarify this, a more thorough disc ussion of defining access follows. Generally, the literature defines access by referr ing to a proxy, utilization, or realized access (Aday, Andersen & Fleming, 1980). Decades ago Aday and others made a distinction between realized and potential access. Realized access can be measured by accounting for utilization, as in counting the fre quency or rate of services that are used, or by patient satisfaction with these services (Aday et al., 1980). On the other hand, po tential access in cludes, the characteristics of the delivery system (the avai lability and distribution of health care providers

PAGE 24

24 and facilities, for example) and characteristics of the population-at -risk in an area (their age, health status, insurance coverage, and income le vels, for example) (Aday et al., 1980, p. 25). This way of defining and discussing access does no t fit neatly into the two types described by Gulliford and Morgan (2003), which signals that although researchers recognize access as a multi-faceted concept, they have not been cons istent in identifying and accounting for these facets. Upon examining Aday and colleagues (1980) definitions of access and Gulliford and Morgans (2003) descriptions, we see that the populations pe rceptions is not explicitly accounted for, and neither is the impact the health care infrastructu re can have on these perceptions. The present research attempts to fill a gap in the literature by examining the possible interaction of the health care infrastructure and a persons percepti ons of their own health and control of it. Relationship Between Health and Feelings of Control Research has suggested that feelings of c ontrol may be a moderator and/or pathway for external factors to affect an individuals hea lth. In a recent article Moradi & Hasan (2004) surveyed an under-researched group, Arab American s, for their experiences of discrimination and self-reported mental health status. More sp ecifically, they were interested in the possible mediating effect of personal control. Through path analysis they found that discrimination was significantly correlated to psychol ogical distress, though causality cannot be determined. They also found that personal control partially mediated the link between discrimination events and psychological distress (Moradi & Hasa n, 2004, p. 423) (Ruggiero & Taylor, 1997; Link, Lennon & Dohrenwend, 1993). Bailis, Segall, Mahon, Chipperfield, and Dunn ( 2001) linked the constructs used in the present research, perceived contro l and self-rated health status Using a national probability sample of Canadians, they found, the indir ect effect of SES through perceived control

PAGE 25

25 approximately equaled or exceeded the direct e ffect of SES on physical and mental health for women and men (Bailis et al., 2001, p. 1670). Though there is strong evidence for relationships between SES, perceived control, and self-report ed physical health, the authors closed with cautioning against causally interpreting the results. They also caution against over-generalizing the perceived control results be cause the index used measured control over important life events; it was not a global measure or a measure related to control over any aspect of health (Bailis et al., 2001, p. 1674). Both Moradi and Hasan (2004) and Bailis and colleagues (2001) examined specific stressors: discrimination and socioeconomic st atus. Gallo, Bogart, Vranceanu and Matthews (2005) conducted a broader examination of stress They found that, Women with lower SES reported less perceived control and more social strain, and less positive emotion when compared with their higher SES counterparts (Gallo, et al., 2005, p. 394). While these results provide further support for the idea that perceived contro l has a relationship with socioeconomic status, they should be generalized carefully given the small sample of women. This section has reviewed research about the relationship between feelings of control and physical and mental health. Re search results lead to questio ns regarding other factors, specifically other sources of enviro nmental stress, could also be rela ted to feelings of control, as may be measured globally or context-specific. Ag ain, the purpose of the present research is to inquire about whether people with low income perceive themselves as having low levels of control in health care settings; and furthermore, if either or both of these perceptions are associated with their relative level of access. Theoretical Framework Several theoretical models will be used to guide this exploratory and interdisciplinary research. The Human Ecological Model (Bronfe nbrenner, 1979) serves as the framework for

PAGE 26

26 this research. This model is useful because it uses interacting ecological levels to study human behavior. This research l ooks at the link between the e xosystem and the individual. Specifically, it examines the potential relationshi p between perceptions of the exosystem and perceptions of the self. This model also provides a way to critically examine the assumptions of health behavior theories, which also exist in society at large. In general, our society assu mes an individual is the starting point and that an individual will seek resources from their environment to fulfill their needs. This assumption is the foundation for many social programs, including health programs. However, according to the Human Ecological Model there is a reciprocal relationship between a person and the environment, a relationship withou t a clear starting point (Bronfenbrenner, 1979). Furthermore, this model does not assume an indivi dual will have the intention to seek resources, or even that the environment will have the resources. This model is useful because it is critical it demands a holistic examination of huma n behavior and societal assumptions. Theories of Health Behavior The health behavior theories pr esented here will be used in varying levels of involvement. The Theory of Planned Behavior (Azjen & Madd en, 1986) is the primary guiding theory since it was developed to better explai n behaviors in which influences outside of a persons control affect their behavior (Azjen & Madden, 1986). Though a dependent behavior, such as healthcare seeking behavior, will not be examined in this research, there are a number of reasons for its usefulness in the model. First, it fits with the Human Ecological Model (Bronfenbrenner, 1979) it accounts for environmental (and internal) fact ors that affect behavi or. Second, it recognizes intention as one of the direct fa ctors affecting behavior Finally, this theory is general enough to be applied to various health programs.

PAGE 27

27 The Theory of Planned Behavior (Azjen & Madd en, 1986) posits that in tention is a direct factor to actual behavior, with other factors affecting intention in turn. The present research challenges this linear relationship and asks if a reciprocal interaction is a more appropriate characterization of the process le ading to health care behavior. The other theories that will be reviewed are: the Perceived Unfairness Model (Jackson et al., 2006), Social Cognitive Theory (Bandura, 1986) the Health Belief Model (Glanz et al., 1997), and the Transtheoretical Mo del and Stages of Change (Prochaska, 1979). The Perceived Unfairness Model (Jackson et al., 2006) will be used in data analysis, along with the Theory of Planned Behavior (Azjen & Madden, 1986). The remaining three theories will be briefly discussed because of their prevalence in research examining health behaviors. The Theory of Planned Behavior (Azjen & Madden, 1986) was developed from the Theory of Reasoned Action (Fishbein & Azjen, 1975). It pr oposes that there are tw o direct influences on behavior: intention and perceived behavioral control. In turn, perceived behavioral control, subjective norms and attitudes toward the behavior also influence intention (Glanz et al., 1997). Intention is the likelihood of th e behavior being performed (Glanz et al., 1997). This research will not review or examine subjective norms or attitudes surrounding any behavior. The process of sample selection will account for intention. Perceived behavioral control is comprised of two constructs: control beliefs and power beliefs. Control beliefs are the perceived influe nce of the environment; th ey are the perceptions of whether or not a person will be able to engage in a particular behavior in a specific context. Power beliefs are the perceptions of the degree to which the environment will impede or facilitate the behavior (Glanz et al., 1997).

PAGE 28

28 The Theory of Planned Behavior (Azjen & Madden, 1986) has not been used widely in health research, but it has been used extensivel y to examine exercise behavior (For examples Motl, Dishman, Ward, Saunders, Dowda, Felton & Pate, 2005; Payne, Jones, & Harris, 2002; and Nguyen, Potvin & Otis, 1997), and also to study sexual health a nd condom use (Bogart, Cecil, & Pinkerton, 2000; De Wit, Stroebe, De Vroome, Sandfort, & Griensven, 2000; Montano, Kasprzyk & Fishbein, 1996, for examples) smoki ng (For examples, Bledsoe, 2006; Collins & Ellickson, 2004; Wilkinson & Abraham, 2004), nutrition (Robinson & Smith, 2002 or Paisley, Lloyd, Sparks, & Mela, 1995 for examples ), health screening and treatmen t adherence (Orbell, Hagger, Brown, & Tidy, 2006; Steadman & Rutter, 2004; Abraham, Clift, & Grabowski, 1999, for examples), alcohol and drug use (McMillan, Conner, 2003; or Kuther, 2002, for examples), and a myriad of other behaviors, including br eastfeeding (Dodgson, Henly, Duckett & Tarrant, 2003, for example), sun protection (For an exam ple Martin, Jacobsen, Lucas, Branch, & Ferron, 1999) and bike helmet use (Quine, Rutter, & Arnold, 2001, for example). Its use is beneficial for a number of reasons. First, it is desi gned for use with a particular behavior, as opposed to measuri ng a persons global capacity to engage in desirous behaviors (Garcia & Mann, 2003). It has been argued that mu ltidimensional health locus of control should be used to assess global capaci ty instead of perceived behavi oral control (Luszczynska & Schwarzer, 2005). Second, and perhaps more importan tly, it is used to ex plain behaviors where a persons exerts little control over the situation in which the behavior is to occur. Finally, it provides a way to examine both internal and external influences on a persons behavior. For this research, control and power beliefs will represent access to health care. If it is found that perceptions of relative access to health care influence perceptions of control, then the Theory of Planned Behavior (Azjen & Madden, 1986) may not be the appropriate framework for

PAGE 29

29 research or interventions with populations that perceive themselves as lacking necessary resources and/or perceiving discrimination. One weaknesses of the Theory of Planned Behavior (Azjen & Madden, 1986) is that it does not account for possible differences that may occur across socioeconomic statuses, r aces/ethnicities, or between se xes. The absence of such concepts is a serious concern in light of much re search pointing to the social weight attached to these and their relationshi ps with health outcomes. In usi ng the Theory of Planned Behavior (Azjen & Madden, 1986) in research, one should rema in open to the possibilities of these factors affecting and interacting at various points with a ny of the already-included factors of the theory. In addition to the Theory of Planned Beha vior (Azjen & Madden, 1986), the Perceived Unfairness Model (Jackson et al., 2006) will be used to interpret the findings of this research. This model emphasizes the impact of subjectiv e experience on physical health (Jackson et al., 2006, p. 24). Constructs in the model include: iden tity, perceived unfairness, helplessness, stress response to perceived unfairness, allostatic load, coping behavi ors, morbidity and mortality. Rather than focusing on the individual, it may be more usefulto cons ider how the social environment influences health and to recognize the importance of social context in shaping health outcomes (Jackson et al., 2006, p. 32). It is important to note that this is the first model to link the constructs laid out for this research. Social Cognitive Theory (SCT) (Bandura, 1986) has a number of concepts that cover perception of the self and the environment, the ac tual environment, emotions, and behavior. It accounts for the interactive relationship between an individual and the environment through the assumption of reciprocal determinism. One critique of SCT is that it seems infallible because it acts more like a model in that it cannot be dispr oved and it more of a lens through which to view behavior (Glanz, Lewis & Rimer, 1997). Another critique is it s assumption that once a person

PAGE 30

30 knows the benefits of a behavior an d that they can perform that be havior, they will indeed carry out the behavior if the perceived costs do not outweigh the perceived benefits. So, while many see SCT as being able to explain any outcome, th e explanation may not be plausible because of this assumption. A large number of factors c ould influence performing the behavior, such as lack of resources or discrimi nation. A person could know they are capable of carrying out a particular action from an internal perspective, but could feel restricted by external factors. Future research using SCT could challeng e this assumption and its infallibility. The Health Belief Model (Rosenstock, 1966) has many of the same constructs as Social Cognitive Theory. It can be described as a very rational model that accounts for personal-level factors such as personality and SES; however the crux of the model is in determining the likelihood of behavioral change. This likeli hood is directly influe nced by combining the perceived costs and benefits of th e behavior change and by the perc eived threat of illness. The likelihood is indirectly affected by the personal factors, and the perceived susceptibility and/or severity of an illness (Glanz et al., 1997). This model is dependent on a persons knowledge of at least illnesses existence and symptoms, particul arly those they may be especially at-risk for. One of the assumptions of the Transtheoretical Model (Prochaska, 1979) is that one theory will not be able to explain the process of changes in behavior (Glanz et al., 1997). This model is particularly relevant because, like the Human Ecological Model, it asks us to examine our assumptions. Its constructs include numerous pr ocesses of behavioral change, five stages of change, self-efficacy, and the costs and benefits of the behavioral change. Information gathered during the current research will be compared to some of these constructs. The purpose of these comparisons will be to see if results are consistent with the predictions of the model. Future

PAGE 31

31 research could use this model to examine if en rollment in a health care program that provides greater access to hea lth care services plays a role in the process of change. The Human Ecological Model (B ronfenbrenner, 1979), the Theo ry of Planned Behavior (Azjen & Madden, 1986) and the Perceived Unfair ness Model (Jackson et al., 2006) are central to the present research. The Human Ecological Model (Bronfenbrenner, 1979) is the backdrop; it is used to critically examine the possible relationships between a persons perceptions of themselves and their environment. The Theory of Planned Behavior (Azjen & Madden, 1986) and the Perceived Unfairness Model (Jackson et al., 2006) take this cr itical stance a step further by outlining the perceptions of inte rest: relative access to health care, discrimination and feelings of control over health. The latter two will be used to interpret the research findings. Health Disparities As discussed in Chapter One, health dispar ities exist in the United States. Various explanations for these disparities have been proposed and researched. The remaining body of this literature review is dedicated to discussing so me of this research. Explanations included are: race/ethnicity, sex, income, health insurance, socioeconomic status, discrimination and relative social standing. Common themes in the literature will be emphasized with their relevance to the present research. Race/Ethnicity Most researchers agree that raci al disparities in health are no t due to any innate, biological differences in races or ethnicities. One prominen t reason is that race is a socially constructed category and therefore it is not linked to genetics (Wu and Schimmele, 2005). This leads to examining various environmental factors for possibl e explanations of racial and ethnic disparities in health. Studies have indicated that racial residential segrega tion is one important factor when examining utilization rates of medical services (Skinner, We instein, Sporer, and Wennberg,

PAGE 32

32 2003; Williams and Collins, 2001). In a review of literature, Williams and Collins (2001) cited various reasons why residential se gregation is an important fact or. These reasons include: the influence of community resources and perceptions of safety on peoples ability to engage in health behaviors, the link between residentia l segregation and educational and employment opportunities, the quality of th e neighborhood and housing, and the quality of medical care. There are also effects of place white people with low income tend to live in areas with a lower concentration of poverty than Afri can Americans with low income. In addition to finding that racial residential segregation is link ed to health care utilization, Skinner and colleagues (2003) found that geographical region was a factor. In a more in-depth analysis Baicker, Chandra, Skinner, and Wennberg (2004) found that not only did racial disparities exist by medical pro cedure within a geographic region, but also across regions. While these results provide useful information, it is im portant to consider that many of the studies looking at regional differences use Medicare and/ or Medicaid data, which does not necessarily control for income, socioeconomic status, the uninsured population or the privately insured population, which could each have varying impacts. Aside from the relationship found between race and geographic characteristics, race and ethnicity are related to income and socioeconom ic status (Skinner et al., 2003; Williams & Collins, 2001; Williams & Jackson, 2005). Retros pective longitudinal mortality data from 1950 to 2000 provided comparisons that are not genera lly found in research. Williams and Jackson (2005) examined mortality rates for five causes of death: suicide, pneumonia and flu, homicide, heart disease, and cancer. Regarding suicide, the data did not show signif icant racial disparities at any point in time. Pneumonia and flu are cas es where large racial di sparities existed in 1950, but no longer exist. They attributed the decrea se to the distribution of medical care through

PAGE 33

33 Medicare and Medicaid, an ar ea which social variations in motivation, knowledge, and resources play a small role (Williams and J ackson, 2005, p. 331). Racial disparities found in rates of homicide, hear t disease, and cancer were often less than the disp arities found among socioeconomic groups. Racial differences in SE S contribute to reduced levels of health insurance coverage for African American, and limited access to medical care plays a role in racial differences in disease (Williams and J ackson, 2005, p. 329). They concluded the article with policy implications. Such policies incl ude decreasing the income gaps and providing greater access to health care among races and ethnicities. As th ese studies show, it is not race or ethnicity that causes health disp arities, but possibly a number of interacting contextual factors (Borrell, Dallo & White, 2006; Lasser, Himmelst ein & Woolhandler, 2006; Wu and Schimmele, 2005). Sex Physical and mental health disparities can be found between the sexes. Two ecological studies examined the potential relationships between womens st atus and autonomy at the statelevel and mortality and morbidity rates and depressive symptoms, respectively (Chen, Subramanian, Acevedo-Garcia & Kawachi, 2005; Kawachi, Kennedy, Gupta & Prothrow-Smith, 1999). Both studies found a correlation between th ese. While the results of Chen and colleagues (2005) held constant across raci al and socioeconomic groups, Kaw achi and others point out that, certain types of gender inequality, such as redu ced female earnings, are themselves patterned by race, class and other socioeconomic characteristic s. In other words, the ill health burden of gender inequality is dispr oportionately borne by women and men with preexisting socioeconomic disadvantage (1999, p.31).

PAGE 34

34 Income There is a burgeoning field of research that li nks income inequities to health outcomes. Kawachi, Kennedy, and Wilkinson (1999) focused on two possible avenues in their recent book. The first is the link between income inequality and the disruption of social cohesion and second the growing evidence on ho w social status gets under th e skin to produce deleterious health effects (p. 158). This latter avenue is taken in this literature re view and serves as one of the springboards of the research at hand. Subramanian, Kawachi, and Kennedy (2001) used an ecological framework to analyze peoples self-reported health status with individual and state-level factors. Results indicated that people with middle-income ($15,000 $35,000) are less likely to report poor or fair health than people in the low-income category, and individu als with high-income are even less likely to report such health status than individuals with middle-income. Through Chi-square analyses the researchers found a significant relationship betwee n self-reported health and state-level data, even after controlling for indivi dual-level characteristics. This finding lends support to the idea that an ecological approach is not only appropriate, but needed, in health research. However, the ecological factors do not dismiss the importance of individual-level factors. Subramanian et al. (2001) also found that self-rated poor health was significantly correlated with each low-income and being African American. Kennedy, Kawachi, and Prothrow-Smith (1996) c oncluded that the rela tive distribution of wealth, not just absolute wealth or lack of it, affects mortality ra tes. They used 1990 Census data to calculate each states average household income and mortality data of the same year from the CDC. They used the Robin Hood index to estimate income inequali ty. This index estimates the percentage of total income that would need to be taken from the richer half of households and given to the poorer half of house holds to attain an equal distri bution of wealth (Kennedy et al.,

PAGE 35

35 1996, p. 1004). The data showed that even after a ccounting for poverty levels, income inequality was positively correlated with mortality rates. In consideration of the criticisms levied on many ecological studies of populations, Lochner, Pamuk, Makuc, Kennedy & Kawachi ( 2001) examined whether individual-level mortality risk is associated with state-level income inequality. Individual characteristics to assess risk included age, sex, ra ce/ethnicity, marital status an d annual income. Their findings indicated that state-level variab les might affect socioeconomic gr oups differently. Lochner et al. (2001) did not find a relationship between mortal ity and income inequalit y for people with high income ($50,000 or more annually). However, th e relationship did exist for people with middle and low incomes (less than 200% of the federal poverty level (FPL). Near-poor persons living in states in the lowest inco me inequality category had significantly lower death rates than nearpoor persons residing in other statesThe same pattern was evident for middle-income persons, although the differences were smaller (Lochner et al., 2001, p. 388). In general, people with low income had twice th e mortality risk of people with high income. Interestingly, they did not find a relationship for those in the fourth inco me category, those living in poverty according to the FPL. The authors suggested, for such persons, it is possible that poverty is a more important determinant of morta lity than state-level ineq uality (Lochner et al., 2001, p. 390). While it may be argued that indi vidual-level factors play a greater role in determining mortality risk than do statelevel data, the auth ors remind us that because income inequality is related to the socioeconomic gradient, it a ffects mortality risk through SES (Lochner et al., 2001, p. 390). Similar to the study conducted by Lochner et al. (2001), Kennedy, Kawachi, Glass, and Prothrow-Smith (1998) examined whet her state-level income inequalities predicted individual-level morbidity. Fair or poor h ealth was described more by women than men, and

PAGE 36

36 blacks more than whites. Income was strongly associ ated with fair or poor health with five times as many people with low income (less than $10,0 00) reporting it than people with high income ($35,000 or more). There is an increased risk of about 30% for fair or poor health for individuals living in states w ith greater inequalities in in come (Kennedy et al., 1998, p. 923). Though this risk lessened when individual factor s were added to the model, the relationship remained significant. This study lends more ev idence to the idea that contextual factors far removed from the individual nevert heless affect individual health. Differences in Measuring Income Some researchers suggested that the varying indices used to measure income inequality make it difficult to relate it to health measures in meaningful way. Kawachi and Kennedy (1997) addressed this issue using data from the 1990 Census. They us ed eight indicators of income inequality and analyzed them with mortality rates. They concluded that each of the inequality indicators were highly correlated with each othe r, poverty rates, and increased mortality rates, respectively. They concluded that the choice of indicator be based on the type of income transfer policies that are in place in the area under research. Socioeconomic Status Though socioeconomic status (SES) was not relate d to functional health status in Wu and Schimmele (2005), it is important to note, as they did, that polic y differences between the United States and Canada may be an inte rvening factor. As stated before people in Canada do not face the same difficulties in accessing health care because of the universal health care system. Research conducted in the United States points to socioeconomic status again and again as a plausible explanation for health disparities. A few pertinent articles will be described here.

PAGE 37

37 Mechanic (2000) defined class and SES. In trying to measure social class, SES is generally used. Typically, it is measured by education, in come, occupation, and sometimes residential location. These measures are highly a ssociated but commonly explain different components of the relationship between SES a nd health, suggesting th at each acts through somewhat different pathways (paragraph 6). In their article, Health disp arities by race and class: Why bot h matter, Kawachi, Daniels, and Robinson (2005) discussed divergent explana tions for the relations hip between race and health. The first explanation is that race is biol ogically meaningful. The second is that race is a proxy for class, and the third: ra ce is a distinct construct, akin to caste (Kawachi et al., 2005, p. 343). They argue for treating race and class as se parate constructs because then it, still leaves open the possibility that race independently influences he alth through pathways such as personal experiences of discrimina tion or cultural differences in lifestyles (Kawachi et al., 2005, p. 347). While they should be re garded as distinct constructs, it is important for research to analyze them together, which has been a rare occurrence. Insurance Lillie-Blanton and Hoffman (2005) reviewed liter ature to address the question of whether, and to what extent does race/ethnicity interact w ith health insurance to a ffect a persons access to health care. Among the four articles they review ed, they concluded that racial/ethnic disparities in health care access could be reduced by greater equity in health insurance coverage among races/ethnicities (La sser et al., 2006). Discrimination Belle and Doucet (2003) reviewed research on women, particularly with their relation to poverty, povertys relation with depression, ineq uality and well-being, and discrimination and well-being. All of the research concurred that women have less status than men in income,

PAGE 38

38 social mobility, social status, a nd life opportunities, which are in tu rn associated with increased morbidity and mortality rates for women. The aut hors urge researchers to focus attention on an issue that is more fundamental than the welfare system: the unequal distribution of resources and the economic, political, and psychological pro cesses that maintain it (Bullock & Lott, 2001 in Belle & Doucet, 2003, p. 110). In a literature review, W illiams, Neighbors, and Jacks on (2003) found fifty-three population-based empirical studies that examin ed the association be tween perceptions of racial/ethnic discrimination and a particular i ndicator of health (Williams et al., 2003, p. 200). None of the studies that examin ed mental health showed a nega tive correlation, a nd all but four reported positive association between discrimination and the outcome examined, for example, psychological distress, anger, major depression, a nd generalized anxiety disorder. There were various studies that examined discrimination with physical health status. Of these, the majority showed a positive association, and a couple repor ted either a conditional association or no association. The authors urge re aders to consider the following in light of their results, The literature on stress and health i ndicates that stressors influence physical illness primarily through causing negative emotional states su ch as anxiety and depression, whic h in turn can have a direct effects on biological processes or patterns of beha vior that affect disease risk (Williams et al., 2003, p. 205). Karisen and Nazroo (2002) analy zed the relationship between racism, social position, and health among ethnic minority groups (Karisen & Nazroo, 2002, p. 625). Concurring with the results from Kennedy et al. (1998), they found statistically signifi cant associations between selfassessed fair or poor health, the experience or per ception of racism, social class, age, and sex (Karisen & Nazroo, 2002, p. 626). Additionally, wo men were 60% more likely than men to

PAGE 39

39 report fair or poor health. The analyses done by Karisen and Nazroo ( 2002) provide evidence more that perceptions of lower social st atus are associated with health status. Harrell, Hall, and Taliaferro (2003) reviewed studies that studied physiological responses to discrimination. While laborator y studies that looked at possible moderating variables, such as coping style, have had inconsis tent results, studies examining physiological response and cultural orientation are relatively consis tent. Other laboratory experi ments are inconsistent. Some studies found increased physiolo gical response in a discrimina tory context, while others concluded that the physiologica l response was similar under disc rimination as other stressors. The combined findings of self-re porting surveys are generally in consistent as well. Some have shown an inverse relationship between disc rimination and health outcomes, others have shown a positive correlation, and still others concl ude that no relationship ex ists. Harrell et al. (2003) concluded, It is not possi ble to argue from the existi ng findings that the different responses to racism are unique ( p. 247). However, researchers have been generally successful in demonstrating that racist ma terialelicits physiol ogical arousal (Harrell et al., 2003, p. 247). They postulated that racism increases the volu me of stress one experi ences and may contribute directly to the physiological arous al that is a marker of stress -related diseases (Harrell et al., 2003, p. 247). Kessler, Mickelson, and Williams (1999) eval uated the prevalence, distribution and mental health correlates of pe rceived discrimination in the U. S. (Kessler et al., 1999, p. 208). Although perceived discrimination did not link mental health pr oblems with race/ethnicity, gender or education, it did account for part of th e association between mental health problems and people with low income. The authors provid e two possible explanati ons for the failure to

PAGE 40

40 link mental health problems with most of the disadvantaged statuses (Kessler et al., 1999, p. 223). First, although reported exposure was gene rally higher among people in disadvantaged than advantaged subsamples, these associati ons were not generally strong. And, second, while there was some variation in the streng th of the association between discrimination and the mental health outcomes across subsam ples this variation wa s neither consistent not strong (Kessler et al., 1999). Although the data did not support th eir hypothesis, they concluded that perceived discrimination may actually be a larger problem than they ha d previously thought because of its prevalence among the participants; one-third had experienced at least one ma jor discriminatory event and over 60% experience day-to-d ay perceived discrimination (Kessler et al., 1999, p. 224). Finally, Schulz, Gravlee, Williams, Israel Mentz, and Rowe (2006) conducted a longitudinal study of the relationship between everyday discrimination and symptoms of depression and general self-report ed health among African Ameri can women in Detroit, Mich (p. 1265). The findings were very clear and si gnificant. There was a positive relationship between experiences of discrimination over time and depressive symptoms, and there is a negative relationship between a cha nge in discrimination over time and a change in self-reported health status (Schulz et al., 2006, p. 1267). Both of these relati onships held significance after controlling for the possible conf ounding variables of age, income education, discrimination and health status at baselin e (Schulz et al., 2006, p. 1267). Relative Social Standing Adler, Epel, Castellazzo, and Ickovics (2000) introduced an instrument for measuring subjective SES. It is a drawing of a ten-rung ladd er given with instructions to place an X on the rung that best represents where they think they stand on the ladder while imagining the people at the top being the best off and the people at the bottom being the worst off (Adler et al., 2000, p. 587). Additionally, measures of objective SES, self-rated hea lth, negative affect,

PAGE 41

41 pessimism, and perceived control over life were collected. Adler et al. (2000) found that subjective SES was significantly related objective SES. Subjective SES was positively correlated with self-rated health, and was associat ed with all of the psychological variables as well. The higher the women placed themselves on the ladder, the lower was their chronic stress, subjective stress, negativ e affect, pessimism and passive coping and the greater was their perceived control over life and active coping (Adler et al., 2000, p. 589). On the other hand, objective SES was only associated with decreased pessimism and a passive coping style. Some limitations to note however are that all of the wo men were white, all of them had at least a high school degree, they were recruited for a main st udy to be healthy, and most of the subjective SES responses were toward the top of the ladder. Summary of Health Disparities As the research has shown, none of the factors that contributes to he alth disparities acts alone. Research supports the idea that various fa ctors that contribute to a persons physical health contribute to whether or not a person ha s regular access to health care services. The majority of research presented here can be br ought together under one common theme: there are hierarchical differences between various groups, and the groups lower in the hierarchy tend to have poorer health outcomes. In extrapolating th is theme and applying it to health care we have the general idea of some groups receiving less health care, which coul d occur in quality or quantity. Drawing from the above liter ature the following relations hips are proposed: greater personal and environmental resour ces are correlated with better he alth, as is regular medical care. Greater personal and envir onmental resources are correlated with regular medical care. While both quality and quantity ar e important factors in describi ng health care, a person must receive some form of health care prior to evalua ting its quality. Yet ther e are steps even before

PAGE 42

42 thisthe environment must contain health care pr oviders and a person must perceive that they can access them. The concept of perceived relative access has been inferred from this research. Given the interdisciplinary nature of this li terature review, a number of areas can be identified for further research. One research pa th, and the one taken in this project, leads to questions of whether socioeconom ic status and self-reported hea lth are related to perceived control over health or a health behavior. This path of research may provide some explanations for the findings that changes in these health practices over time are patterned by social status. Disadvantaged racial groups and those with low SES are less likely to re duce high-risk behavior or to initiate new health-e nhancing practices (Williams & Jackson, 2005, p. 327). While the current research project will not be able to dr aw causal relationships, it will ask the following: whether people with low income perceive themse lves as having low levels of control over their health, and if this is broadly associated with pe rceptions of their access to health care resources in the community, as compared with their perc eptions of others acce ss to such resources.

PAGE 43

43 CHAPTER 3 METHODS This chapter describes the methods used to explore the potentia l relationship between perceptions of access to health care and percep tions of control over ones own health. The primary investigator was not aw are of any research linking thes e constructs, however, recently Jackson and colleagues (2006) published the Pe rceived Unfairness Model, which proposes a relationship. A case study design with mixed met hods was used for this exploratory research. The details of this project follow. Research Site Selection Alachua County, Florida, was an appropriate pla ce to conduct this research due to the large gap in the distribution of resources among areas of Alachua County, even within its one urban area, Gainesville. Gainesville, a large urban area of Alachua County, houses the University of Florida, Santa Fe Community College, and a la rge student population which contribute to these disparities. In addition to the social and material inequities here, Alachua County offered a unique site to conduct this research because th e voters expressed a commitment to addressing health care disparities. In 2004, voters approve d a seven-year 0.25% sale s tax to fund a local program that would provide uninsured workers w ith low-income access to primary health care services; this program has a component with fewer benefits for citizens 65 years of age or older that does not necessitate being uni nsured. This ballot initiative was passed after years of many concerned citizens raising awaren ess of rising health care cost s, the increasing numbers of uninsured, and the lack of health care associated with being un insured. This community-based program, Community Health Offering Innovative Care and Educational Services (CHOICES) Health Services, began in October 2005, and ha s over 460 enrollees to date (As of 10/27/06).

PAGE 44

44 Research Design Given the communitys initiative to provide a ccess to health care to workers with lowincome, this was a prime location to explore the idea that perceived access to health care is related to an individual s perception of their ab ility to manage their health. A case study design was suited for this exploratory research. In a case study, participants can be chosen on a common outcome, homogeneity, just as in experiments, so intern al validity is strengthened (de Vaus, 2001). In this research, all of participants had a low-income and were enrolled in CHOICES. This case study was used to build and explore the Theory of Planned Behavior (Azjen & Madden, 1986) and the Perceived Un fairness Model (Jackson Kubansky & Wright, 2006) by examining whether or not acce ss to health care and/or soci al class are associated with feelings of control over health. Sample Selection The theoretical population for this research was all people with low-income. The theoretical sample included all of the people en rolled in CHOICES Health Services Program. Randomized sampling techniques were not a concer n for this study since cases are used for theoretical rather than statis tical generalization (deVaus, 2001, p. 240). Due to the rate of enrollment in CHOICES and the sample size goal of 15, ongoing sampling with new groups of enrollees was required. Subjects were inte rviewed within three weeks of enrollment in CHOICES to minimize the possibility that subjec ts perceptions and responses might change as result of participation in the program. Instrumentation The interview was designed to study two main variables: pe rceived relative access to health care and perceived feelings of control over health. In a ddition to the interview schedule

PAGE 45

45 designed specifically for this research, the CHOICES Enrollee Health Questionnaire was administered to each participant. In addressing access to health care it was im portant to balance the existing research knowledge with attempting to learn the experiences and perceptions of the participants in this study. In other words, it was important to frame questions according to previous research, but also to allow the participants to use their own definitions in responding. The open-response questions inquired about their perceived leve l of access and were supplemented by questions about their perceived relative access. Perceived relative access was, in part, measured by the ladder instrument adapted from Adler and co lleagues (2000). Although the depiction of the instrument was different and the in structions were altered to specifically address access to health care and the population of Alachua County, the concept of social relativity was the same. The remaining questions regarding perceived relative access were created for this project and asked about feelings of discriminati on in health care and were cons tructed with the foundation of perceived behavioral control from the Theory of Planned Behavior (Azjen & Madden, 1986). Perceived control over health was the other important variable in this research. Prior research has indicated that peopl e with low-income feel less cont rol over their liv es than people with higher incomes (Adler et al., 2001; Bailis et al., 2001; and Gallo, et al., 2005 for examples). However, no research has been identified that examined the possible relationship between feelings of control over health and perceived relative access to health care. CHOICES Enrollee Health Questionnaire: As previously mentioned, the CHOICES Enrollee Health Questionnaire was administered with the interview schedule. This multiplechoice/short answer questionnaire elicits the following information: 1) perceptions of health status, 2) diagnoses of the following condi tions: asthma, diabetes, high blood pressure,

PAGE 46

46 congestive heart failure and depr ession, 3) global satisfaction with life, 4) frequency of utilization of emergency rooms in the last two year s, 5) utilization of hos pital inpatient services in the last two years, 6) freque ncy of primary health care visi ts, 7) length of time since last physical and dental exams, 8) prim ary barrier to utilizing health ca re services, and 9) whether of not they had a usual source of health care prio r to enrolling in CHOICES (CHOICES internal documents, 2006). The instrument is comprised of items used in the Behavioral Risk Factors Surveillance Survey (BRFSS), the National Health Insurance Survey (NHIS), and the JaxCare, Inc. Needs Assessment. Although there has been re search on the validity and reliability for the BRFSS and the NICH, neither of these instruments was used in their entirety in the present research. The author was unaware of any data regarding the validity or reliability for the JaxCare, Inc Needs Assessment, but it was not used in its entirety either. Administration Once data collection began, each week an introduc tory letter was sent to all new enrollees that fit the criteria for this study, people that enrolled in the previous week. This letter was shortly followed by a phone call inviti ng the subject to participate and to arrange a meeting time if the enrollee was willing to participate. At least 3 postal days transpired after the letter had been mailed and an enrollee was contacted vi a telephone. Based on standard recruitment protocol and after weighing time and resources, it was decided that each enrollee would be contacted by phone five times (Lavrakas, 1993). Interviews were held at a public location of the participants choosing. The day before the interview, each participant received a reminder call. Prior to beginning each interview, the informed consent form was reviewed. Each part icipant was asked if they had any questions and written consent was obtained. A ll participants were asked if they consented to being taperecorded. In the few cases were the participant did not consent, the inte rview was not recorded.

PAGE 47

47 The interviews that were tape-recorded were later transcribed for analysis. The tape recorder was not turned off until the participant had left since interviewees sometimes open up at the end of the interview(Bryman, 2004, p. 333). For th e interviews that we re not tape-recorded, responses were typed out, as if transcribed, for later analysis. Analysis Some of the interview data were analyzed using thematic analysis of the transcribed interviews (Kossak, 2005). This analysis included constant comparison of the responses in order to identify concepts, categories, and themes th at emerge from the data (Marshall & Rossman, 2006; Strauss & Corbin, 1998). The first step in thematic analysis was to describe concepts, which, in the case of this research, was the phrases or statements that stand out in a participants response (Strauss & Corbin, 1998). Next, categories were form ed by combining concepts. Strauss and Corbin (1998) provide two reasons for this: first, it pares down the amount of material; and second, it allows th e researcher to begin explaining and predicting the data. After categories were created, they were then collapsed into themes. Themes were then, the product of this inductive process (Marsha ll & Rossman, 2006). Analyses were done manually to allow the researcher complete immersion in the data. Though qualitative data was the focus, frequencies were calculated of the emergent concepts, categories, and themes. Each participants codes were compared with their responses to the scalar instrument and the CHOICES Enrollee Health Questionnaire. Th e results from this process were compiled and reviewed for trends. The trends were then comp ared to the concept patte rn developed prior to data collection and is similar to that described in Bailis and colleagues (2001). Chi-square analyses were performed for va rious combinations of the constructs and demographics. Frequencies of the responses and demographics were calcul ated. The results of these and the qualitative analyses ar e presented in the next chapter.

PAGE 48

48 Concerns/Critiques There was a concern regarding the amount of time spent administering the instrument package. The length of time could affect respondents and the in terviewer by prompting respondent fatigue or instrument decay which would decrease the inte rnal validity of the research (deVaus, 2001). This concern was so mewhat abated after pre-testing showed an average of 20 minutes for each participant. There was a concern about response rate. Fo r this reason a moneta ry incentive of $15.00 was offered to each participant. This was st ated in the introductor y letter (Appendix F). People who were contacted for participation we re eligible to receive services since the Alachua County Office of Social Services (OSS) had approved th em for the CHOICES program. The Office of Social Services does the elig ibility screening for CHOICES; however, the approved applicants have not nece ssarily completed the enrollment process. After a person has been approved by OSS, their name, contact inform ation and demographics ar e sent to CHOICES. At this point CHOICES mails them an enrollment package, which includes forms that must be completed, signed and returned in order for some services to be rece ived and the enrollment process to be complete. The decision to contac t enrollees after OSS has determined eligibility, rather than waiting until the enrollment process has been completed, was based on the following: o Once they have been approved by OSS, they can use the prescription benefit; o Once they are approved by OSS, they are sent information that informs them they are enrolled in CHOICES, which could have a psychological impact. In an attempt to obtain a more homogenous sample, a time limit needed to be set for how long a person could be in the program and still be contacted for participation; and Four weeks was chosen after considering th e amount of time allowed for mailing letters prior to contacting enrollees vi a phone, the amount of time allotted for attempting and making

PAGE 49

49 phone contact, allowing for the in terviews to be scheduled up to one week in advance of the phone contact when a person agreed to participate. This time was seen as short enough in that the chances of a person returning the necessary forms to CHOICES, seeing a doctor and receiving a prescription and obt aining any type of medical test was fairly small.

PAGE 50

50 CHAPTER 4 RESULTS This chapter presents the results of the exploratory case study, which inquired about perceptions of access to health care and perceptions of control over ones health. Twenty-three people who are enrolled in CHOICES Health Servi ces participated in this research. First, demographics of the participants will be desc ribed. Next, the quantit ative findings will be presented, followed by the qualitative results. The chapter will conclude with a summary of a few of the most significant results. Response Rate and Demographics of Sample The response rate was 29.11%. The characte ristics of the sample are as follows: The mean age was 37.6 years, with a range fr om 21 years. (Distri bution in Figure 4-1 on p. 81) Twelve women (52.2%) and eleven men (47.8%) participated. The distribution of race/ethnicity of the sample is (Figure 4-2 on p. 81): African American: 34.8% Caucasian: 60.9% Latino1: 4.3% Aside from those who were excluded, ther e were three categor ies in addition to participants: those that refused participation, those that agreed to participate, but did not come to the interview, and those who were unable to be c ontacted via telephone. A t-test was run with a Bonferroni adjustment to see if any of these gr oups significantly differed from the participants regarding age. Chi-square analyses were run to compare these groups to see if there were 1At the time data was collected, the O ffice of Social Services (OSS), whic h collects the demogr aphic information from enrollees for CHOICES, did not in clude Latino as an option. Other races and ethnicities that OSS specified at that time included Asian, Native Am erican, Bi-Racial, and Other. The Latino participant in this study was documented as Caucasian by OSS. For th is reason, they have been categorized as Caucasian for the analyses that follow.

PAGE 51

51 significant differences in race/ethni city, sex, or marital status. Th ese analyses revealed that none of the groups were significantly different from the participants on the measured demographics. Quantitative Analysis Quantitative analysis was used to explore perceived access to health care services, perceived relative access to health care serv ices, and the CHOICES questionnaire, as well as trend within these. The results of these analyses follow. Perceived Access to Health Care Services Access chart There were eighteen items on the Access Chart, most of them external resources. Participants were asked to place a mark next to each factor they considered important to being seen by a doctor or other hea lth care professional. Extra sp aces were offered and it was suggested that they could add anyt hing else they deemed important. After this, they were asked to rank the top five factors from those they initially marked as important. Finally, the participants were asked if they ha d each factor on the list two m onths ago and if they had it now. The responses to each of these follow. Important access chart factors Participants identified an aver age of 12 factors as important. Five participants added one f actor to the original list. More than half of the participants indicated the following were important: having a regular doctor (95.7%), trusting the doctor (95.7%), having health insurance (87.0%), knowing about how much it will cost ahead of time ( 87.0%), knowing they could see a specialist if the doctor referred them (87.0%), knowing th ey could get lab work done if the doctor suggested it (87.0%), knowing th ey could get prescriptions if the doctor prescribed them (82.0%), a doctor with a payment plan (78.3%), being able to get off of work (73.9%), having cash or credit (69.6%), transportation (69.6%), part icipating in a government program (56.5%), a doctors office open at nigh ts or on weekends (56.5%), and being able to walk-in without an appointment (52.2%) (Figure 4-3 on p. 82).

PAGE 52

52 The remaining factors and percenta ge of participants that incl uded them as important are as follows: knowing they could cha nge health-related behavior if the doctor recommended it (47.8%), knowing someone that goes to the doctor they are thinking of seeing (26.0%), babysitter (21.7%), and an item added by a participant (21.7%). Ranked important access factors Two factors were included in the top five ranked important factor s: health insurance (65.1%) and having a regular doctor (60.7%). Hea lth insurance was generally ranked first, while having a regular doctor was generally ranked third. Three additional factors were in cluded regularly by participants at 43% of the sample each: trusting the doctor, being able to get prescriptions if the doctor prescribes them, and being able to see a specialist if the doctor refers them. Access factors they have now Regardless of the importance placed on each factor by the participant, less than half of the participants had the following factors from the Access Chart: cash or credit, health insurance, participation in a government pr ogram, babysitter, knowing someone that sees the doctor they are thinking of seeing, a doctors office ope n nights or weekends, being able to walk-in, a regular doctor, trusting the doctor, being able to see a specialist if the doctor refers them, being able to get lab wo rk if the doctor recommends it, knowing they could change health-related behavior if the doctor recommends it, and items added by the participant. Regardless of the importance placed on each fctor by the participant, more than half of the participants had the following items from the Access Chart: transportation, can get off work, knowing the cost of a visit ahead of time, and being able to get prescriptions if the doctor prescribes them. On average, participants had 40.54% of the factors they marked as important and 42.63% of the factors they included in their top five important factors. Most of the participants had ha lf or less than half of either the factors they marked as important (69.6%). Another 21.7% had betw een 51% of the factors they indicated were important. The remaining 8.7% had ove r 75% of the factor s they considered important to seeing a doctor. There were no significant diffe rences on any of the above measures between sexes or races, or among ages or age groups. Participants ranged from 0% for the percenta ge of factors they have now that were included in their top five importa nt access factors. Most had ha lf or less of the items they included in their top five factor s (65.2%). The distribution is as follows (Figure 4-4 on p. 83):

PAGE 53

53 o 0% of resources: 43.5% of participants o 25% of resources: 21.7% of participants o 50% of resources: 0% of participants o 75% of resources: 34.8% of participants Of the participants that repor ted they had 75% of the top fi ve factors they marked as important the majority of them (75%) indicated they had health insurance. Since one of CHOICES eligibility requirements is that an applicant not be insu red, one could presume the participants were not insured, but thought that CHOICES was a type of insurance, if this were the case, the distribution wo uld be as follows (Figure 4-5 on p. 84): o 0%: of resources: 43.5% of participants o 25% of resources: 21.7% of participants o 50% of resources: 26.1% of participants o 75% of resources: 8.7% of participants Perceived Relative Access to Health Care Services After completing the Access Char t, participants were asked to place a mark on the step where they would stand of a ten-rung ladder. Th e participants were then asked to imagine that everyone in Alachua County got to stand on this ladder. The pe ople standing on the top step had everything they needed to see any doctor or health care provider, wh ile the people standing on the bottom step did not have what they needed. The top and bottom steps included possible examples of people who could be standing there, the coaches at the University of Florida, doctors themselves, and people that are homele ss or cannot find a job right now, for each step respectively. These examples were included af ter pilot testing and we re chosen over providing examples of actual resources the people would have, or not ha ve, at each step. Participants placed themselves on steps onesev en. The median was four. Steps one and five received the highest number of participants with five and six respectiv ely (Figure 4-6 on p. 85). Ninetyone percent of the participants placed themselves on the bottom half of the ladder (steps one through five). The remaining 8.7% placed themselves on the top half. There were no significant diffe rences between sexes or races, or among ages or age groups.

PAGE 54

54 Perceived Access Participants were asked a series of four questions about whet her or not they could get the care they needed if they were slightly sick, like if they had a cold or the flu, if they had a medical emergency, if they were depressed, and if they ha d a problem with drugs or alcohol. At least half of the participants felt th ey could get care if they were slightly sick, had a medical emergency, or if they had a problem with drugs or alcohol (56.5%, 60.9%, and 56.5%, respectively). Just below half of the participants (47.8 %) indicated they could get care if they were depressed. In each circumstance there were some that fe lt they would not get the care they needed (slightly sick 17.4%, emergency 17.4%, depresse d 21.7%, and problem with drugs or alcohol 26.1%). However, the responses were not categori cally yes or no. For e ach question there were participants who responded that they would not take that concern to a doctor or medical professional (26.1%, 8.7%, 13.0%, and 8.7%, re spectively). Additionally, there were participants who were not sure or did not know if they could get care if they had a medical emergency (8.7%), were depressed (13.0%), or ha d a problem with drugs or alcohol (8.7%). Finally there were two depends responses fr om two participants for either a medical emergency or if they were depressed. These responses will be included in the qualitative analysis section later in the ch apter. There were no significant differences between sexes or ages, or among ages or age groups. Level of Control Participants were asked to rate the level of control they felt they had over their health from a list five responses. The multiple choice responses were a) no control, b) that their health was out of their hands, c) a little cont rol, but not a lot, d) a lot of control, but not complete, or e) complete control. Over half of the participants (65.2%) agreed th at they had a lot of control over

PAGE 55

55 their health. The remaining participants were split almost evenly between having no control (13.0%) and a little control (17.4%). One particip ant (4.3%) felt they had complete control. There were no significant diffe rences between sexes or ages, or among ages or age groups. Participants were asked to pick which stat ement the people standing on the top and bottom steps of the Relative Access La dder might agree with. The re sponse options were the same, however worded in third person instead of first. The results were very clear. The majority of participants attributed either a lot (56.5%) or complete (30.4%) c ontrol to those standing at the top step, cumulatively 87.0%. The remaining 13.0% indicated those on the top step have a little control over their health. Again, a majority of 87.0% attributed th e lesser levels of control to those on the bottom step: no control (26.1%), their health was out of their hands (43.5%), or little control (17.4%). Of the remaining responses, 8. 7% indicated that those on the bottom step had a lot of control and 4.3% did not choose a step on the ladder, but said that it depends. There were no significant differences between sexes or ages, or among ages or age groups. Perceived Discrimination Scenario Participants were asked a series of questions about whether or not they felt certain groups would have an easier time getting in to be seen by a doctor given that the example person in each group had been having trouble breathi ng lately and wanted to see a doctor about it. They were asked if there were differences between sexes, whether or not someone has health insurance, whether or not someone has the ability to pay cash up front, if there we re differences between African Americans and Caucasia ns, and if there were differe nces between other races or ethnicities. Participants were asked to limit their responses to what they perceived in Alachua County. If the participant indi cated that there was a differen ce within each section of the scenario, they were asked a follow-up question to confirm or indicate which group they thought was favored. Additional follow-up questions as ked them how the disadvantaged person (the

PAGE 56

56 term disadvantaged was not used; they were as ked how the person might feel knowing they dont have or are not (insert variable) given that they know their situa tion might be different if they had or were (insert variable). They were asked what advice th ey might give the disadvantaged person. Finally, after all the variables were revi ewed, participants were asked which variable would weigh more than the others, of those they had indicated would make a difference. (The question posed to them included listing back the variables they had previo usly indicated.) The responses to most of these questi ons will be described in the qualit ative analysis section of this chapter. Briefly, the general responses of whethe r or not there were diffe rences, and what groups were favored, are described below. Does sex makes a difference? Most partic ipants (69.6%) stated that there is not a difference between women and men. However, 4.3% said that is de pends, and 26.1% said that there is a difference. Of those who indicated that there w ould be a difference or that it depends, 71.4% identified fema les as having an easier time and 28.6 identified males. Does insurance make a difference? The majority of participants indicated that it does make a difference (82.6%). The remaining re sponses were distributed over the responses that it would not make a difference (4.3%), that they did not know if it would make a difference (4.3%), and that it would depend (8.7 %). Of those who indicated that insurance would make a difference, it was unanimous that the person with insurance would have an easier time. Does the ability to pay cash make a differenc e? Over half (65.2%) stated that it would make a difference, 26.1% said that it would not make a difference, a nd 8.7% said that it depends. Of those that indi cated it would make a differenc e, 60.9% were clear that the ability to pay cash would be favored. Is there a difference between African Americans and Caucasians? There was not majority in agreement in responses. Almost half (43.5 %) of participants indi cated that there was not a difference, and an equal number (26.1%) e ach indicated that it depends or that they hope not/it should not. One participan t (4.3%) stated they did not know. Is there a difference between other races or et hnicities? Just over ha lf (56.5%) stated that there was not a difference. Anot her 17.4% responded either that it depends or that they did not know/hadnt had the experien ce. One participant (4.3%) replied that they hoped not and another (4.3%) replied th at there was a difference. The most important variable? Almost half of the participants (47.8%) indicated that health insurance alone was the most important. Another 43.5% indicated another reason that

PAGE 57

57 generally included both health insurance and th e ability to pay cash. The remaining 8.7% indicated cash alone was the most important. There were not any significant differences be tween races or ages for any of the above questions. However, through chi-square analys is, it was found that there was a significant difference between age groups when asked if there was a difference between sexes ( =18.847, significant at 0.016 with 95% conf idence). Twothirds (66.7%) of the participants between 45 years of age indicated that there wa s a difference. There were two variables that differed significantly by sex: ability to pay cash and if there was a difference between African Americans and Cau casians. Regarding the ability to pay cash, 90.9% of the men indicated that it made a di fference, whereas 35.7% of women stated that it did ( =6.302 significant at 0.043 with 95% conf idence). Finally, regarding difference between African Americans and Caucasians, 72.7% of men indicated that it did not make a difference, but only 14.3% of women said the same ( =7.905 significant at 0.048 with 95% confidence). CHOICES Health Questionnaire The CHOICES Health Questionnaire was the last instrument to be given to the participants. The results are as follows: a. Perceptions of health status Participants were asked to ra te their health as excellent, very good, good, fair, or poor. Most participants reported they had either very good or good health (39.1% each). 13.0% indicated that they had fair health and 8.7% said their h ealth was poor. There was not any significant difference between sexes or races, or among ages or age groups. b. Number of poor physical health days Participants were asked how many days during the past thirty had they had poor physical health, which included injury or illness. The range was zero to thirty days with an average of five days. There was not any signi ficant difference between sexes or races, or among ages or age groups. c. Number of poor mental health days Participants were asked how many days duri ng the past thirty days had their mental health been poor; this include d stress, depression, and problems with emotions. The range was zero to thirty days with and aver age of 9.5 days. There was not any significant difference between sexes or races, or among ages or age groups. d. Number of days poor physical or ment al health hindered daily activities Participants were asked how many days duri ng the past thirty days had either poor physical or mental health stopped them from doing their usual activities, which included going to work, self-care, and recreation. Th e range was zero to thirty days with an

PAGE 58

58 average of 8.5 days. There was not any signi ficant difference between sexes or races, or among ages or age groups. e. Diagnosis of any of the following condi tions: asthma, diabetes, high blood pressure, congestive heart failure and depression, Participants were asked if they had any of the four health conditions targeted by CHOICES, and depression, which is curre ntly not covered under the program. Most of the participants (60.87%) ha d at least one of the conditions. o 13.04% reported having asthma o 13.04% reported having diabetes o 21.74% reported having high blood pressure o 4.35% reported having conge stive heart failure o 34.78% reported having depres sion, though not all of them had been clinically diagnosed There was not any significant difference between sexes or races, or among ages or age groups. f. Global satisfaction with life Choosing from a list of responses, participants were asked how satisfied they were with their life, very satisfied, satisfi ed, dissatisfied, or very diss atisfied. The majority (87.0%) of the participants were either very sa tisfied (26.1%) or satisfied (60.9%). One participant (4.3%) was very diss atisfied, and two participants (8.7%) indicated they were between being satisfied and dissatisfied. There was not any significant difference between sexes or race, or among ages or age groups. g. Number of emergency room visi ts in the past two years Participants were asked how many times they had visited the emergency room in the past two years. Responses choices were 0, 1, 2, 3, or 4 or more. While it is a conservative estimate, if four visits are used as the uppe r bound, the range was zero to four visits with an average of 1.48 visits. There was not a ny significant difference between sexes or among ages or age groups. A significa nt difference was found between races2. A chisquare analysis revealed signi ficance of 0.034 at 95% confidence ( =8.673). Subsequently a t-test was conducted, whic h resulted in significance of 0.021 at 95% confidence (t=2.503). African Americans had a total of 20 visits and Caucasians had a total of 14 visits, which represent 8 and 15 participants respectively. h. Number of hospitalizations in the past two years 2 For the purposes of analysis, the racial categories included African Am erican and Caucasian. The Latino participant was included in the Caucasian category sin ce this matched how race and ethnicity was being recorded with the Office of Social Services at the time of data collection.

PAGE 59

59 Participants were asked how many times they ha d been hospitalized in the past two years. Response choices were 0, 1, 2, 3, or 4 or more The responses range d the answer choices, however only one participant (4.3%) responded w ith four or more. The majority (73.9%) of participants had not been hospitalized in the past two years. The remaining responses are that two participants (8.7 %) had been hospitalized once and three (13. 0%) were twice. There was not any significant difference betw een sexes or race, or among ages or age groups. i. Number of health care visits in the past year Participants were asked how many times they ha d been to a doctors office or other clinic in the past twelve months, excluding the em ergency room visits and hospitalizations. Answer choices were none, 0, 1, 2, 4, 6, 8, 10, 13, and 16 or more. The number of visits ranged from zero to 13. Th e majority (87.0%) of participants had 5 or less visits. The average, median, and mode we re two visits. The complete distribution is as follows: o 0 visits: 17.4% (4 participants) o 1 visit: 17.4% (4 participants) o 2 visits: 30.4% (7 participants) o 4 visits: 21.7% (5 participants) o 6 visits: 0% (0 participants) o 8 visits: 4.3% (1 participant) o 10 visits: 4.3% (1 participant) o 13 visits: 4.3% (1 participant) There was not any significant difference betw een sexes or race, or among ages or age groups. j. Length of time since last physical Participants were asked how long it had b een since they had had a routine physical, which was not a check-up regard ing a specific injury or c ondition. The response choices were within the past year, with in the past two years, within the past five years, or more than 5 years. The responses, showing a bi-modal distribution, were as follows: o Within the past year: 43.5% (10 participants) o Within the past two years: 4.3% (1 participant) o Within the past five years: 13.0 % (3 participants) o More than five years: 39.1% (9 participants) There was not any significant difference betw een sexes or race, or among ages or age groups. k. Length of time since last dental exam

PAGE 60

60 Respondents were asked how long it had been since they had b een to a dentist or dental clinic for any reason. The response choices were within the past year within the past two years, within the past five years, or more than 5 years. Half ( 52.2%)of the participants had seen a dentist within the past year, but for many (39.1%) it had been at least two years. The distribution is as follows: o Within the past year: 52.2% (12 participants) o Within the past two years: 8.7% (2 participants) o Within the past five years: 21.7% (5 participants) o More than five years: 17.4% (4 participants) There was not any significant difference between sexes or race, or among ages or age groups. l. Barriers to utilizing health care ervices Participants were asked if any of the following had stopped them from going to the doctor in the past six months: cost, transporta tion, office hours, or language barriers. The majority (73.9%) of participants indicated th at cost was a barrier, 26.1% indicated that it was not. Substantially fewer participants pe rceived barriers of inconvenient office hours and transportation, 13.0% and 8.7%, respectiv ely, most did not (inconvenient hours 87.0% and transportation 91.3%). There wa s not any significant difference between sexes or race, or among ages or age groups. None of the participan ts perceived language as a barrier, however all of the participants were fluent in English and it cannot be assumed that this is not a barrier for a nyone participating in CHOICES or living in Alachua County who might be eligible for the program. m. Usual source of health care? Participants were asked whether or not they had a usual doctors office, clinic, or other place they go to if they are sick or need advi ce about their health. Ha lf of the participants (56.5%) indicated that they did have a usual source of care, and slightly less than half (43.5%) said they did not. There were not any significant differences between sexes or among ages or age groups. However, a signi ficant difference betw een races was found. All of the African Americans indicated they had a usual source of care, while 38.5% of the Caucasians indicated the same. ( =9.436 at 0.002 significance at 95% confidence). There is a discrepancy between the number of participants that res ponded affirmatively to this question and the percent that indicated they had a regular doctor while completing the Access Chart. It is likely that while there would be some overlap, having a regular doctor is not the same as having a usual place to go when one is sick. Taken together the results of this question and the one inquiring about the number of emergency room visits in the past two years could have implications for health care usage and program planning and outreach activities.

PAGE 61

61 Quantitative Responses to Research Questions RQ1. How do people with low incomes associate feelings of control over their health with having access to health care services? Chi-square analysis was run to see if there was an association between participants level of control over their own health and their pe rceived access to health care, which included the percent of important Access Char t resources they have now, th e percent of the five ranked Access Chart resources they have now, and the quartile they fit in for the five ranked resources. Statistical significance was not found between th e participants level of control and their perceived access to health care. RQ1a. Is perceived access to health care services associated with any form of discrimination? Chi-square analysis was run to see if ther e was an association between participants perceived access to health care and the various va riables included in the S cenario, principally if the participants perceived differences in abili ty to access services be tween sexes, health insurance status, ability to pay cash, and among races and ethnicitie s. Statistical significance was not found among any of these variables. RQ1b. Is perceived relative access to health care services associated with feelings of control over health? Chi-square analysis was run to see if there was an association between the participants perceived relative access, which included where they placed themselves on the ladder, their ladder quartile, whether they placed themselves on the bottom or top half of the ladder, and whether or not they felt they could get the care they needed under various circumstances, and the level of control they felt over their health. Statistical si gnificance was not found between perceived relative access to health ca re services and control over health.

PAGE 62

62 The second research question inquired about possible relationships between the concepts examined in the first research question and th e responses to the CHOI CES Questionnaire. A few of the questions from the questionnaire were chosen for chi-square analysis either because the concept is prominent in prominent in the litera ture or because it might be particularly useful information to the CHOICES program. The select ed questions addressed health status, if the participant had any of the speci fied conditions, life satisfactio n, number of emergency room visits in the past two years, number of doctors vis its in the past year, and if the participant had a usual source of care. RQ2a. Are feelings of control associated wi th selected questions from the CHOICES questionnaire? Only one question from the questionnaire was significantly associated with participants level of control, the numbe r of emergency room visits in the past two years ( =17.868 at 0.037 with 95% confidence). Participants that had not been to the emergency room in the past two years felt they either had a lot of control over their health (88.9%) or comp lete control (11.1%), which represents 39.1% of the participants. Partic ipants that had been to the emergency room at least once in the past two year s (60.9% of the partic ipants) ranged in responses among having no control (21.4%), a little c ontrol (28.6%), or a lot cont rol (50.0%) over their health. One other question approached statistic al significance, life satisfaction ( =16.502 at 0.057 with 95% confidence). Over half of the participants ( 65.2%) indicated that th ey had either a lot or complete of control over their health and that they were satisfied or ve ry satisfied with their life. The remaining combinations are as follows: 8.7% indicated that they were very satisfied with their life and had no control over their heal th; 13.0% stated that they were satisfied with their life and had a little contro l over their health; 4.3% stated th at they were very dissatisfied with their life and that they had no control over their health; and the remaining two participants

PAGE 63

63 felt they were between being satisfied and dissati sfied with one (4.3%) fe eling that they had a little control over their health, a nd one (4.3%) feeling that they ha d a lot of control over their health. RQ2b. Is perceived access associated with selected questions from the CHOICES questionnaire? No statistical significance was found among a ny of the questions addressing perceived access to health care and any of the selected questions from the CHOICES questionnaire. RQ2c. Is discrimination associated with selected questions from the CHOICES questionnaire? Chi-square analysis was run to see if ther e was an association be tween whether or not participants perceived differences between any of the groups presen ted in the Scenario, and if so, what status was favored. Statis tical significance was found when asking if there was a difference between sexes when compared to the participants health status ( =19.912 at 0.003 with 95% confidence). Another chi-square analysis was then done with these and compared by races and were found to be significant for African Americans ( =8.000 at 0.018 with 95% significance) and Caucasians ( =13.200 at 0.040 with 95% signi ficance). It was found to be significant when comparing women ( =13.440 at 0.009 with 95% confidence) and men ( = 7.975 at 0.047 with 95% confidence). All of the participants that indicated there was not a difference between sexes reported that they were in very good or good health, whereas those that reported there was a difference between sexes, or that it depended, ranged from good to poor hea lth (28.6% reported good health, 42.9% reported fair healt h, and 28.6% reported poor health).

PAGE 64

64 RQ2d. Is perceived relative access associated with selected questions from the CHOICES questionnaire? No statistical significance was found among th e questions addressing perceived relative access and any of the select ed questions from the CHOICES questionnaire. After the above results were found, additional chi-s quare analyses were r un to see if there was a relationship between any of the following: Access Chart if they have cas h or credit now, if they have health insurance now and if they have a regular doctor now and Step on the ladder, ladder quintile, whether the participant was on the top or bottom half of the ladder, if they felt they could get the care they needed under various circumstances, level of control over their ow n health, and from the CHOI CES questionnaire: health status, if they had any of th e specified conditions, the number of emergency room visits in the past two years, the number of doctors offi ce visits in the past ye ar, and if they had a usual source of care. There were three statistically significant asso ciations, all with havi ng cash or credit now: step on the ladder ( =36.827 at 0.006 with 95% conf idence), ladder quintile ( =27.371 at 0.001 with 95% confidence), and whethe r the participant was on the top or bottom half of the ladder ( =12.204 at 0.007 with 95% confidence). Summary of Quantitative Analysis Given that the majority of que stions asked of the particip ants for the above analyses focused on external facilitators of perceived access to health care it is not surprising that these were identified as important. It is noteworthy, howe ver, that one salient potential internal barrier on the Access Chart, knowing that one could make health-related behavior changes if the doctor recommends it, was not included as important by over half of th e participants. Yet it was included by a substantial number of participants so it s hould be further explored.

PAGE 65

65 It is notable that there were not any significant differences between sexes on the majority of the above analyses. While there was one significant difference among age groups in the Perceived Discrimination Scenari o, this should be interpreted car efully since there were not many people within each category. As seen, ther e were two health care utilization questions where statistical significance wa s seen between races. Though it is understood the sample size prohibits statistical generalizability, the si gnificance found within each question could be partially explained by the correspon ding question. This is an area th at deserves further research. Overall, the participants generally do not have what they feel they need to access health care, principally health insurance and a regular doctor. This corresponds with where they placed their health care access when th ey compared it with other citi zens in Alachua County, at the bottom half, or less than at least half of Alachuas citizens. It corresponds to the results of the Perceived Discrimination Scenario that placed high importance on heal th insurance and the ability to pay cash in order to have an easier time accessing health care. Despite the lack of resources and generally low perceived access to hea lth care, participants reported that they have a lot of control over their healt h, are in good health, and are satisfi ed with life. Many are still utilizing health care to some degree, whether it is through emergency rooms, doctors offices, dental clinics, or some combina tion of these. These experiences with the health care community could be partially reflected in their responses to feeling they could get the care they need under various circumstances. In the qualitative analys is that follows, the participants perceived relative access, and the remaining interview questions, will be discussed. Qualitative Analysis Qualitative analysis was done to answer th e main research question and its two sub questions. The first step of the analysis establ ished codes, which were the actual words of the participants. From these codes thirty categories were created:

PAGE 66

66 Scenario Most important factor Feelings for not having res ource identified in scenario Advice for person in scenario Tell lawmakers Ideal health care system Health and accompanied feelings Control over health Associations with health insurance Relative Access to Health Care Associations with the ladder (perce ived relative access to health care) Feelings accompanying finances Work and health Medical bills Health care costs Thoughts on the emergency room Doctors perceptions Dentists perceptions Money and doctors Medical community Thoughts on the community of Alachua County CHOICES reputation What is wanted out of CHOICES Name of CHOICES Why did I apply? Applying for CHOICES CHOICES services Recruiting for CHOICES In addition to CHOICES Questionnaire Women Men Concepts from the three main theories used in this research, the Human Ecological Model (Bronfenbrenner, 1979), the Theory of Pla nned Behavior (Azjen & Madden, 1986), and the Perceived Unfairness Model (Jackson et al., 2006), were used to synthesize the categories into themes. This was done because it was seen as a clear way to explore the large amount of data through the lenses of each theory. Constant comp arison of the data was done until the reviewers came to consensus.

PAGE 67

67 Each theory lent itself to a different facet of the datas wide extent. Each section below will contain two parts, an introduct ion to the theory and its relevant concepts and a discussion of the data with quotes from the participants. Themes: The Human Ecological Model The Human Ecological Model (B ronfenbrenner, 1979) served as the framework for this research. The participants responses described re lationships at each of the four levels in the model. Beginning with the most immediate, each level will be discussed in terms of its theoretical importance followe d by participants responses. Microsystem The microsystem is the most proximal system according to the Human Ecological Theory (Bronfenbrenner, 1979). Examples of particular ly relevant microsystems are the relationship between a patient and their doctor, the relationshi p between a client and a social worker, and the relationships between individual citizens and their family memb ers, friends, and neighbors. A series of four questions asked whether or not participants felt they could get the care they needed under various health circumstances. These questions were asked as a component of their perceived access to health care. When asked if they felt they could get care if they were slightly sick many responded that they would take care of that themselves, by getting over the counter medication, If I get a cold or flu I dont usually go to the doctor. Some al so indicated that they would go to their mothers, even in a medical em ergency. Sometimes friends or coworkers were mentioned. They were generally described as people with plight similar to that of the participant, but not as resour ces for information or support. Some participants mentioned having a doctor they would normally go to, but are concerned now because that doctor is not listed as partic ipating with CHOICES. Generally, people did not mention having a relationship with a doc tor; although through the CHOICES Health

PAGE 68

68 Questionnaire most indicated that th ey have utilized some aspect of the health care system in the past two years. Responses to questions about possible discrimination in potential encounters with doctors offices can be categorized at the micro level. Many people that indicated that racism may exist saw it as an individual issue not a systemic one. For example, It comes down to whos in the receptionists office, and I th ink, honestly, I think it all depends whos on the other side of the counter. Mesosystem The mesosystem can be described as the links between the microsyste ms (Bronfenbrenner, 1979). The relationship between doctors and staff of the CHOICES program is a critical aspect in the success of the program. Other mesosyst ems include primary care doctors relationships with specialists and lab technicians, and the rela tionship between the social workers and the staff of CHOICES. Participants described a fractured system, one that is not fixing the problems with the affordability of and access to health care. One participant voiced her awareness of the large number of people that have b een without health insurance fo r a long time, and how this has contributed to the progressed state of many of th eir health problems. A couple of participants described seeking information and help only to be told that they have to contact other people, who would in turn, refer them to yet others. As part of the scenario, participants were asked what advice they would have for a person who woul d have a more difficult time getting in to be seen by a doctor; often times th e participant concluded that th is person was having a difficult time because they did not have health insurance or the ability to pay for the visit. Many advised that the person should apply for CHOICES or save to buy insurance. One response in particular from a 48-year old woman captured the se nse of a general lack of options or choices :

PAGE 69

69 I dont know what resources to go toOften times a lot of the in formation is hard to get a hold of, nobody offers it; they act like its a huge secret. So, I wouldnt know what to tell them. I really wouldnt because I have done it all, including owing Shands and ruining my credit. Exosystem In the exosystem policies and organizations are created. It could be likened to a kind of social infrastructure where the values and more s of a society become mo re concrete through the formation and maintenance of such institutio ns (Bronfenbrenner, 1977). For example, CHOICES Health Care Program is part of the ex osystem. It is a program supported and staffed by the Alachua County government and impacts the citizens. Within the disjointed system participants identified in Alachua County, they emphasized the importance of CHOICES. When asked whethe r or not they could ge t care if they were slightly sick, one participant responded that now that they are enrolled in CHOICES, I would say yes. When asked what they would tell th e lawmakers of Alachua County about the health care system here, the CHOICES program was positively recognized. These responses include women and men, African Americans and Caucasians, and span the ra nge of ages in the sample: I went to the doctor for the first time last week because of CHOICES. I hadnt been before. On a roll with getting CHOICES going for the uninsured. Give the CHOICES program a chance. CHOICES is a good start. Right now its doing okay, theres CHOICES. Without CHOICES Id still have this broken tooth in my mouth. Im appreciative. I think its great, cause I wouldnt ha ve any otherwise.

PAGE 70

70 Macrosystem Unlike the other ecological levels, the m acrosystem does not describe a set of relationships. The macrosystem is the body of values and priorities of a given society (Bronfenbrenner, 1979). For exam ple, American society values the structure of the Western medical system. It is generally believed that th is system is open to t hose who would like to use its services. There is a growing sentiment that h ealth care should be more affordable, which can be exemplified by the CHOICES program and others similar to it. Participants were given a brief scenario with gradually more information about two people who wanted to see a doctor about a health concern they had been having. This scenario posed questions about the poten tial for differences in the ease of get ting in to be seen between sexes, between having and not having health insurance, the ability (or lack of it) to pay cash, and among races and ethnicities. Overall, what matte red was having health insurance and the ability to pay, not sex, race, or ethnicity. This theme was summed well with the response of a 51-year old man to the scenario: Money. Green. Thats the one color, green. Insurance, credit card, they see thatonly color America sees is green. Some participants did indicate that discrimination exists at the macro level between sexes, races, and socioeconomic status. A couple of men pointed to WIC and other programs that they perceived as giving women an easier time of getting in to see a doctor. A young African American woman indicated that although there should not be differences between African Americans and Caucasians in the eas e of getting in to be seen by a doctor, she also said that, It shouldnt, but some people might do things differe ntly. She further pointed to the different society younger people were raised in as comp ared to older people. A young Caucasian woman described an interaction,

PAGE 71

71 You know, it depends if, you know, theyre poor. I mean, if theyre a rich white person and a poor black person; if its a white person wi th health insurance or black person, or vice versa, then yeah (race is going to make a difference). Themes: The Theory of Planned Behavior The Theory of Planned Behavior (Azjen & Madden, 1986) served as the theoretical foundation for many of the interview questions. Th e concepts of control and power beliefs were of interest, more specifically, the potential rela tionship between a persons feelings of control over their health and their perc eived access to the communitys h ealth care services. Although it was not specifically asked about, many participan ts shared how perceived access to health care services affects the decision to se ek these services, or intention. Control beliefs Control beliefs are the perceived influences of an environment on a behavior (Glanz et al. 1997). The participants feelings of control over their h ealth do not seem to be positively related to their perceived access to health care. It seems that different factors influence the amount of control one feels over their own health and the amount of control one feels in being able to obtain health services in the community. As seen by their responses to the CHOICES Health Questionnaire, participants are generally utilizing some aspect of the health care system. However, some answers as to how much and unde r what circumstances they are using it, and why they are not using it are found throughout the interviews. Some reasons for not going to the doctor were that they could not afford it, they did not have health insurance, and that they tried to take care of the concern on their ow n. Participants said they went to the doctor when they were re ally sick, if they knew they could work out a payment plan with a doctor, if they knew the doctor would not do unnecessary lab work, or if they would be able to get in-office discounts. Some reasons for not going to the emergency

PAGE 72

72 room were that there was a long wait, it is em barrassing, cost, and they had been treated poorly previously because they did not have health insu rance. Despite all of the reasons the emergency room was avoided, it was seen as a place to go where you do not need health insurance or the money to pay for services, a place to if you really needed to be see n, a single place to get a lot of different services, and a place where family and friends go. Power beliefs Power beliefs are the degree to which the envi ronment impedes or facilitates a behavior (Glanz et al., 1997). After partic ipants identified environmental f actors that affected whether or not they go to a doctor or the emergency room, a subsequent question would be to what extent do these factors affect the health care seeking behavior. The follow ing are some responses to the question of whether or not the participant felt they could get care they needed in a medical emergency. Their answers speak particularly to th e seriousness of their inju ry or illness and the cost. Again, these responses span the demographics of the sample. Mm, I mean, I could go to the emergency r oom, but it would be an insane bill, you know, but if I really had to I would go. So, I would, but I woul d go at the very, very, very last second, you know. Yeah, if Im obviously have (sic) broken bones or bleeding, yeah, Id get to (the) emergency room. Definitely not; the emergency room is killing me. I mean, I dont know what to do with these bills. I would have to do whatever I would have to do. I would just go to the hospital, they would take care of me and I would pay for the rest of my life is what would happen. Well, yes and no. Well, it depends on the medical emergencyif I need something that costs $200,000, I doubt that, but I mean, if I broke my leg, of course

PAGE 73

73 Well, I do, but itd take a long time. Last time I went to the emergency room it took almost seven hours. Other information was shared at diff erent times throughout the interviews. When responding to the access chart, one 53-year old participant commented that, Money has been an issue and I actually have resisted going to the docto r, knowing that hes going to prescribe antibiotics that wont be able to afford. So whats the point of going to the doctor? Participants were asked about how the person without insurance mi ght feel, given they are sick and want to see a doctor. A particularly rele vant response was from a 21-year old woman, who said, (They) wouldnt want to go to the doctor, like me. Participants played on the name CHOICES; its lay meaning was used with it and it was an underlying current. For example, one reason a participant did not have the money right now for doctors visit was because they were making car repairs. Another participant used the money they had been saving to repair their car and so they fortunately had it, but unfortunately too, they are struggling with transportation. If you constantly savin up, but you wont be because the moneys gonna go to bills Another particip ant, an African American woman, described the money they would have to pay to see a specialist as the same as a utility bill, so youre making a choice. I mean, either pay my phone bill, my utility bill, or not make a car payment because I need to see a doctor. But $400 is certainly a lot less than going to the emergency room and paying $4000. So there are not many choices that Im learning. We make the decisi ons that are best for us Themes: The Perceived Unfairness Model As previously stated, the Per ceived Unfairness Model (Jackson et al., 2006) is a new model proposed to explain the paths by wh ich perceived unfairness might af fect physical health. To the authors knowledge, this is the fi rst research using this model. There are numerous constructs in this model. The participants responses were an alyzed according to the definitions provided in

PAGE 74

74 the article by Jackson and colleagues (2006) with the understanding that these constructs may be refined over time with research. Perceived unfairness Jackson and colleagues (2006) openly defined perceived unfairness as encompassing a breach of entitlements or of psyc hological contracts related to group membership or to individual experiences (p. 24). Some of the participants responses over lap between this perception, the accompanying emotions, and their identities, but fo r these constructs will be separated to some extent for analysis. Participants perceptions of unfairness are found th roughout the interviews. The responses that follow are pres ented in the order in which the interview questions were asked. When asked if they felt they could get the care they needed if they had a medical emergency, a participant replied that they had contacted a social work er at Shands about emergency room bills and found out that, Im not approved because I fell between the cracks. Because I work, I support myself, barely. Fall through the cracks. I dont fit in any program. In going through the scenario, one of the peopl e presented, John, does not have insurance. The following are replies to how he might feel, gi ven that he has a medical concern and wants to see a doctor about it. In response to Johns not having insurance, one young man replied that, he might do what he can, but still, anybody with health in surance would still, s till be seen first. Another man identified with John as man pe rceiving discrimination between the sexes, Well, its not equal. Its not equal at all. A 39-year old woman identified wi th John not having insurance, Well, if I were John, and Ive experienced both, both ends of that youve just got to get the doctor to see him, regardless. Its just the way the doctors have treated me, versus, you know, its like I go to the docto r with insurance then its li ke, I didnt have insurance and went to a specialist and I paid him mone y, you know, gave him the full price for it, no problem, but he didnt even want to look at me his attitude towards me was that you dont

PAGE 75

75 have insurance, what are you even seeing me for, Im a surgeon? Well, you know, Im in pain, I have a problem, I want to know what it is He didnt give me the time of day. I paid $300 bucks for that, and he said, no you dont have an infection and left. And um, because I didnt have insurance and hes a surgeon and hes a specialist, and I felt that was really negative The next couple of statements were also respons es in the scenario regarding insurance and the ability to pay cash. These responses repr esented the demographics of the sample. Yeah, everybody who has health insurance will get seen by a doc tor quicker than everybody else. I feltI work, I pay my taxes, I do the best th at I can. And theres just nothing available. I have searched and searched for something to help me and nothing. I think, sometimes its like who you know tooit kind of f eels like a scam. Participants indicated that a person needs health insura nce and/or cash to receive thorough, optimal care, care that makes one feel like a citize n, a human being, care that one gets with a real job. Although it was not aske d of them, some participants offered the association of a good job with educ ation, health insurance, and th e ability to get health care. All of the participants were asked what they would like to get out of CHOICES. One 46year old man responded broadly and with some disgus t, to not be able to be seen by a doctor or dentist, to go years at a time without, in this country, of this wealth, with this surplus of medical talent, build ings, facilities When asked about the feelings of control pe ople at the bottom step of the ladder might feel, a 45-year old man responded: They want to find a job, but aint no job available. Some of the problems of the he alth care system in Alachua County that they would share with lawmakers are that: There are a lot of people out there who provide h ealth care, but its not affordable; and, Ive been treated ugly in an emergency room because I didnt have any insurance and I owed Shands. Additional comme nts about the broader sy stem include: I wish

PAGE 76

76 everybody could get health care. It shouldnt be a problem, I dont think; and, I need help. I dont have no (sic) medication. I need some me dication. Well, we cant help you. Its like youre out for dead. Identity In the Perceived Unfairness Model, identity is simply ones concept of self (Jackson et al., 2006, p. 25). Again, the concept of identity can be seen at various points of the interview. The scenario drew out the most here, with par ticipants describing diffe rences between sexes: Men have a tendency to just be able to make more, earn more, and to do more for themselves. Meanwhile, women are runnin around tr ying to take care of their children first. but you know, white male, youre left to forge ahead on your own Being a man, I feel like Im kind of take n advantage, even though Im not still Another woman had seen women sometimes get pr eferential treatment and witnessed African Americans being treated worse than Caucasians, but they had seen everyone being treated the same. Despite these possible diffe rences, she felt that health insurance would make it easier. The following are responses to not having health insurance or cash: If you dont have it, you can forget it. Youre last of the last, bottom line. Theyll feel out of place. Theyll feel like, not being a citi zen, not being human. Yeah that (insurance) could sway it either way. But now, if you re a minority and you dont have insurance, thats eve n, thats the worst s ituation. (In the s cenario the African American had insurance and the Caucasian did not). When asked to describe what level of control the people at the top of the ladder might feel, two responses were: These peopl e are aliens to me; and, We ll, see, you cannot compare the people up here to down there In the additiona l comments below, the juxtaposition between feeling like one is part of the community and yet is not part of it, a second class citizenship,

PAGE 77

77 where people who are working, are not treated like people, with re spect. I do pay a lot of money in taxes. Help me, just please help us. From CHOICES one participant hoped that they would not have to be turned away in the emergency room like an alien. Emotion While perceptions and identity contribute to th e perceptions of unfairn ess, emotions are one of the responses that accompany such perceptions (Jackson et al., 2006). As stated by Jackson and colleagues (2006), this emotional response is not necessarily negative and could be neutral, and the emotional reaction could lead to both negative and positive responses to the perceptions, such as stress and motivation for collective actio n. Participants did not express the motivation for collective action. When asked in the scenario how a person might feel without having health insurance participants ranging in th e demographics of the sample i ndicated that they would feel, Left out. Or, if its like a consistent condition, maybe some sort of depres sion, anxiety, probably, maybe a little frustrationits probably frustration Furthermore, if a person did not have health insurance or cash: I think theyll feel worse. If you aint got it, you cant do nothing. Upset, worried, scared. If hes sick, might feel helpless. Discouraged. I can tell you exactly what its like becau se Ive been going through it for years. Nobody cares. The doctors dont car e, the hospitals dont care. Participants were asked what leve l of control might be attributed to the people at the bottom step of the ladder: Been there, felt completely out of control.

PAGE 78

78 They dont have any (control). Some people when, in these situations, they fe ll like, some of the pe ople feel like they can have some kind of hope and some just dont. Kind of like, just dont wanna, kind of give up. Nobody cares, you know, always negative it all depends on the person. Coping behavior Like emotions, coping behaviors are other responses to the per ceptions of unfairness (Jackson et al., 2006). Participants expressed the use of avoidance. A 45-year old participant di scussed the emergency room, Most times you go to the emergency room, youre there until 6 oclock, 11 o clock at night. Thats embarrassing, really embarrassing; I dont like to go through that. I do nt want to go. I ju st be hurting and taking some Advil and go to bed. The situation of not having health insurance was particularly salient, people do not want to go to the doctor: As a 24-year old woman said, going to the doctor without having insurance, people dont do it if they dont have insurance. This was concurred with the response of a 26year old woman, He may see it as a onetime deal. I really do think, Im just gonna go ahead and pay this one-time fee and take care of my health and all my i ssues. He may think this is a one-time thing and its not. You have to have health insu rance (with it) you can go anytime. He sees it as a one-time thing; he only going to the hospital when its necessary. This was explained even more by a 25-year old man, He has to pay a lot of money up front. So, like myself, I might pay up front, I dont like to think about going to the doctor. $95 fo r what, you know, going to the doctor and they dont really tell you anything. I could do some thing else with that money, I could pay another bill or something. That could be some times on his mind as far as actually going to the doctor, you know, most people with health insurance, they just go and they just pay their $20 or whatever it is, its nothing really on their mind. I think thatd be one of the main things, so, Im just going to spend a little bit of money no w and you know maybe the bills covered for the rest of it, so I ca n go check just make sure. But somebody who doesnt they wont just go check to make sure, theyre gonna say well, they said it was nothing. Just gotta keep believing.

PAGE 79

79 Despite all of the behaviors that may seem be t ypes of avoidance, all of the participants still heard about CHOICES and applied, wh ich they described as neither an easy nor a quick process. Yet, many participants would still advise others to apply and some are actively telling others to do so. The majority intend on taking advantage of the services that CHOICES offers. An understated coping behavior seen in all of the participants is that each of them are still seeking to change their health care situation. Responses to Utilizing CHOICES Serv ices and Thoughts for Lawmakers Many of the participants descri bed a specific health care service, such as dental work, prescriptions, or exams that they would like to access now that they are enrolled in CHOICES. However, there were a number of people that ex pressed the desire for security and having the choice to get health care. I want security. I want a little; I was so afraid that I was gonna have to have a $20,000, $30,000 operation and it was gonna put me so far out of financial contro l of my life, that this would help so much. They would at le ast pay for some (lab tests). I know they wouldnt pay for the operation, but I would feel like I can have doctors I can talk toMost stuff is really easy to take care of. The security of knowing that I have something, that I could get care. Im trying to take better care of myse lffinding out whats wrong with me. Just to go to a regular doctors appointment fo r an affordable rate; go to get prescriptions. And be able to go to a specia list if I needed to, cause I know they dont offer that now. Health care that makes me feel like the human being that I am. You know what I mean, something to fell back on, you know. Generally, participants wanted to tell lawm akers how appreciative they are of the CHOICES program, and urged them not to cancel it. Help me, just please help us.

PAGE 80

80 Just dont let em cut the program out. I need something, and we need itIm appreciative of what they doSo, if they re willing to help, I think its great. Give the CHOICES program a chance. Yet at the same time, many were dissatisfied at the state of the health care system. Many feel that the program needs to be expanded to in clude more choices more dentists and primary health care providers; add speci alists, vision care, and mental health counselors. Other suggestions were including some emergencies, su ch as broken bones, or perhaps urgent care centers. On a broader level, a number of particip ants desired to see a co ordinated health care infrastructure, one that would include the whole fa mily, once youre in, youre in. Perhaps one that incorporated some of the ideas used to provide affordable housing on a sliding scale to Make it available for everybody. Qualitative Responses to Research Questions RQ1. How do people with low incomes associate feelings of control over their health with having access to health care services? As stated earlier, there seems to be a disconne ct between feelings of control over health and access to health care services. As seen in th e quantitative analysis, generally participants felt they had a lot of control over their health, desp ite half of them having half or less of the resources they felt they needed to gain access. RQ1a. Is perceived access to health care services associated with any form of discrimination? As seen in the scenario, most participants did not feel that sex, race, or ethnicity impacted the ability to access health care services. Howe ver, the majority identified the person without health insurance as disadvantaged and many indicated that lacking the ability to pay for the visit would be a disadvantage. Alt hough the participants did not initiate the use of the word

PAGE 81

81 discrimination, the current of discrimination based on the abili ty to pay for services (through insurance or otherwise) was apparent. RQ1b. Is perceived relative access to health care services associated with feelings of control over health? The result here is similar to that in RQ1; participants did not seem to positively associate their perceived relative access to health care with the feelings of control over their health. Despite the fact that most of the participants placed themselves on the bottom half of the ladder, most felt they had a lot of control over their health. RQ2. Do any of the above concepts, control over health, perceived access to health care, discrimination in health care, and perceived re lative access to health care, show patterns of association with responses from the CHOICES Questionnaire? To answer this question in a qualitative manner would require asking questions that were not included in the current interview schedule. Such questions may need to be very specific, such as, whether or not a person feels that ha ving a medical condition, asthma for example, is related to their feelings of access to the system. This line of questioning may not be particularly illuminating in some circumstances. For example, it may seem intuitive that a person who does not have a chronic medical condition would most likely not rank knowing they can get prescriptions when they need them as highly as a person who does have a chronic condition. What one may find in such research, however, is that knowing may provide a kind of psychological security that a ny person, with or without a ch ronic condition, could want. Summary of Qualitative Analysis As described by the Human Ecological Model (B ronfenbrenner, 2001) it is apparent that each system impacts individuals, as does their inte ractions, or their lack of it, as seen in the mesoand exosystems. Participants identified positive and negative aspects of their health care ecologies, though the negative aspect s hold the weight. From the pe rspective of the Theory of

PAGE 82

82 Planned Behavior health insurance and the abil ity to pay for health care were the dominant factors. Insurance, I think, is a key part to a better health anywa y. And cash, it help (sic) attain insurance. The Perceived Unfairness Model adde d depth and humanity to the data that was described through the other two theo ries. Future research using th ese theories, as well as policy and program implications of this research follow in the next chapter.

PAGE 83

83 Figure 4-1. Distribution of participants by age group Figure 4-2. Distribution if part icipants by race and ethnicity

PAGE 84

84 0102030405060708090100 Regular doctor Trusting doctor Health insurance Knowing cost ahead of time Specialist Lab Work Prescription Payment plan available Getting off work Cash or Credit Transportation Government program Nights or weekends Walk-inAccess ItemsPercent of Participants Figure 4-3. Distribution of access chart f actors that were marked as important

PAGE 85

85 0 5 10 15 20 25 30 35 40 45 50 0-2425-4950-7475-100 The Percent of Top Five Factors a Participant HadThe Percent of Participants in each Category Figure 4-4. Distribution of participants in each quartile of the top five factors they indicated they had which facilitate their access health care

PAGE 86

86 0 5 10 15 20 25 30 35 40 45 50 0-2425-4950-7475-100 The Percent of Top Five Resources a Participant HadThe Percent of Participants in each Category Figure 4-5. Distribution of partic ipants in each quartile of the top five access factors they indicated they had if health insurance is taken away

PAGE 87

87 Figure 4-6: Distribu tion of participants percei ved relative access to health care relative to others in Alachua County

PAGE 88

88 CHAPTER 5 FUTURE RESEARCH AND POLICY IMPLICATIONS The first part of this chapter will review some of the key findings of this research. This will be followed by a more in depth discussion of the qualitative data in the context of the theories. The chapter will conclude with implic ations and recommendations for future research, the CHOICES program, and health care policy. Review of the Results As was seen in the quantitative data, responses did not significantly differ between sexes, races, or age groups. The qualitative data did not reveal substantial differences in these regards either. Some participants did perceive sexe s and races being treated differently and these perceptions should not be taken li ghtly, especially since they can be accompanied to feelings of being left out for dead. The access chart reveal ed that a majority (95.7 %) of the participants thought that having a regular doctor and trusting the doctor were important to being seen. Participants were then asked to rank the factors they felt were important; A majority of participants agreed on only two factors, having a regular doctor (60.7%) and health insurance (65.1%). This brings us to the salience of all of the factors surr ounding not having health insurance and/or the ability to pay for health car e, from exams and prescriptions to emergencies and surgeries. While it cannot be assumed that either or both of these were the predominant factors influencing the participants perceptions of their relative acc ess to health care, which step they placed themselves on the ladder, the data is corroborative. Furthermore, the relative access instrument was administered foll owing the series of questions of whether or not they had the resources indicated on the chart. A significant a ssociation between having cash or credit to pay for services and the ladder was found.

PAGE 89

89 Research Question 1 RQ1. How do people with low incomes associate feelings of control over their health with having access to health care services? Neither the quantitative nor qual itative data found that people associate control over their health with having access to health care services. RQ1a. Is perceived access to health care services associated with any form of discrimination? Although statistical significance between perceived access to heat h care and perceptions of discrimination were not found, the qualitative data revealed otherwise. The perceived differences between having health insurance and not having it, and having the ability to pay for care, and not having that ability were prominent. Additionally, pa rticipants spoke of differences between sexes and races. While these discrimina tions were not supported by a majority, they are nevertheless present and s ubstantively significant. RQ1b. Is perceived relative access to health care services associated with feelings of control over health? Neither the quantitative nor qual itative data found that people associate control over their health with their perceived relativ e access to health care services. Research Question 2 Research question two was only answered qua ntitatively in this research and a brief summary of these results follows. RQ2a. Are feelings of control associated wi th selected questions from the CHOICES questionnaire? Statistical significance was f ound between the amount of cont rol a person felt over their health and the number of emergency room visits in the past two years. More participants that had not been to the emergency room in the past two years felt they eith er had a lot of control over their health (88.9%) than those who had been to the emergency room at least once (50.0%).

PAGE 90

90 RQ2b. Is perceived access associated with selected questions from the CHOICES questionnaire? Statistical analysis did not reveal any signifi cance between perceived access to health care and any of the selected questions from the CHOICES questionnaire. RQ2c. Is discrimination associated with selected questions from the CHOICES questionnaire? Of the selected questions from the CHOICE S questionnaire, only one was statistically associated with one type of di scrimination, health status and se x. Statistical significance was found for African Americans and Caucasians and for women as well as men. The participants that did not perceive differenc es between the sexes reported very good or good health status, while those that perceived sexual discriminati on ranged over good, fair, and poor health status. RQ2d. Is perceived relative access associated with selected questions from the CHOICES questionnaire? Statistical significance was not found among a ny of the questions addressing perceived relative access and any of the selected ques tions from the CHOICES questionnaire. Data in the Context of Theories The Human Ecological Model The Human Ecological Model was an invalu able resource because it allowed the participants responses to be view ed from the different levels of the environment. Of particular interest in this study was the rela tionship of the partic ipants with the health care system. The microsystems that the participan ts described with individual un its of the medical system were both positive and negative. The participants had good experiences with doctors and other medical staff, but also poor experiences. Inte restingly, these experiences were characterized by how doctors and staff behaved on the issues of th e participant not having health insurance, not having the ability to pay for all of the serv ices immediately, or needing to not have

PAGE 91

91 unnecessary services to reduce the overall cost. Participants generally did not mention other aspects of health care. The exceptions were of a couple of participan ts mentioning a specific doctor, they felt comfortable talking with about their health concerns (one womans experience was a doctor she had about twenty years ago), one participant mentioning that they called all of the CHOICES providers and none of them could see him for at least a month (although he had a case of shingles), and a few participants menti oning the length of waiting time in the emergency room.This can be summarized by stating that th e participants lack stable and predictable microsystems in the health care system. As Bronfenbrenner (1992) postulates, The degree of stability, consistency, and predicta bility over time in any element of any level of the systemsis critical for the effective operation of the system in question. Extremes either of disorganization or rigidity in structure of f unction represent danger signs for potential psychological growth (p. 163). Unfortunately, many of the participants have a history of the health care system costing them resources, both internally and externally. Th is kind of consistency is not congruent with an optimal condition for human development (p. 163). It is possible that the participants share a macrosystem, in the sense of a subculture (Bronfenbrenner, 1979), which place s greater emphasis on accessible health care for all than the macrosystem of the larger American culture. This hypothesis springs from the despair and disgust when describing the current mesoand e xosystems of health care, and the earnestness with which they wish for a defragmented infrastruc ture and more inclusive health care policies. The participants may be part of a large culture that share such values, but in a society where other values, such as individualism and profitmaking, have shaped the structure the mesoand exosystems of health care.

PAGE 92

92 This lack of connection with significant others and cognitive dissonance could contribute to the feelings of unfairness, which will be discussed in more depth below. Furthermore, the negative experiences and perceptions seem to impact the participants health care seeking behavior, specifically by avoiding the system. Such behavior c ould then impact the system by having entire groups of people th at postpone or neglect addressi ng their health concerns and do not trust that they will recei ve thorough care once a medical professional sees them. The consequences of such behavior increase the ri sks of poor health outcomes and stress not only on the individual, but also for the system that at tempts to only help those once they are drowning downstream1 (Egan & Cowan, 1979) or have proven th emselves to be deficient enough to warrant assistance2 (Bronfenbrenner, 1988b). The Theory of Planned Behavior Although the constructs represen ting those from the Theory of Planned Behavior, control over health and perceived access to health care, did not have a relationship that could be encapsulated by perceived behavioral control, the theo ry still lent itself to analysis of the data. Participants did not express broad discomfort with their ability to access health care, for example, making appointments, being able to secure transportation, or having a lack of knowledge about the services that are available. However, some participants expressed that they were not sure how to find a doctor, even with the list of CHOICES providers that they must choose from, they felt that they did not have enough information about the doctors to make a decision. Such thoughts could be categorized as a persons internal prohib iting power beliefs. Prominent power beliefs seen in the data were external barriers. The degree to which these 1 From a fable that tells of a person who went upstream to find out who was pushing all of the drowning people in the water in order to stop it so that they would not be continuously just pulling people out of the water. 2 Bronfenbrenners Deficit Model de scribes Americas social policy.

PAGE 93

93 barriers, lack of health insurance and the ability to pay for services, canno t be quantified with the current data. The qualitative data reveal that thes e barriers are certainly in place (control belief) and that they strongly influen ce their ability to access health care (power beliefs). The qualitative data additionally communicated that such thoughts affect inte ntion to access health care, and with less surety, the actual behavior of accessing hea lth care when the participant perceives it is needed, and the motivation to act on advice from others if it involves seeking care. Using the lens of the Theory of Planned Behavi or, judgements can be made that are similar to those made using the Human Ecological Model. The consequences for having rigid, external barriers to health care may be far-reaching. The continued us e of the Theory of Planned Behavior with the Human Ecological Model could be very useful as it may be able to account for factors at each ecological level that contribute to a specific behavior. Th e rewards for learning of such factors are then solid foundations with whic h to develop health car e policy that increases peoples perceived behavioral control and inte ntion to perform health-improving behaviors. The Perceived Unfairness Model It was stated earlier, but the fact that th is model provides a theoretical framework for conceptualizing peoples emotions and thoughts, whic h can be used in turn to generate systems that help those same people, cannot be understate d. This model provides a way to capture both the profound feelings of despair at not being tr eated like a human and the resilience to keep trying. It could be used in c onjunction with the Human Ecological Model. When participants were asked about sex and racial discrimination many participants reported that it just depends who is on the other side of the counter. This fi nding may be important in two ways: First, it could signal that there is a d ecrease of such discrimination, a nd second, this possible constant alert could contribute to an in crease in allostatic load, which is a construct in the Perceived Unfairness Model that could have impli cations for poorer health outcomes.

PAGE 94

94 Future Research There are many routes for future research, especially since this was an exploratory investigation. Some methodologi cal considerations will be addressed first, followed by theoretically based avenues. Methodological Considerations Building on the same theoretical basis that was used in the present study, there are many options for future research. These include havi ng a larger sample and utilizing a longitudinal design. It includes having a control group, which may vary depending the angle of the research. Potential control groups include pe ople that have been enrolled in CHOICES for a given amount of time, people that have been denied from CHOICES, people that have not applied for CHOICES with possible sub-group s for those who are aware the program exists and those that do not, people that are privately insured and have been for a given amount of time, people that are enrolled in other governmen t programs, such as Medicaid, people that cycle between being insured and not being insured, wh ether it is privately or thro ugh the government, and finally, people that have never been insured. Human Ecological Model It is important to keep in mi nd that the perspective represente d here is the individual. As stated before, the individual is a ffected by each system and the intera ctions of the systems. What is not represented in the data is a significan t component of the Human Ecological Theory: the effects individuals have on each system and th eir interactions (Bronf enbrenner, 1979). To capture this information, questions would need to be directed toward members of the participants mesoand exosystems, such as doc tors, social workers, and county lawmakers to see the impact individual citizens and populations of people that are uninsured and/or have low income. Another level of the theory that was not included in this research was the chronosystem.

PAGE 95

95 The chronosystem captures the aspect of time, implied and necessary when discussing human development (Bronfenbrenner, 1988a). Future res earch should be conducted at the individual level to construct a more complete picture of th e perceived environments of others in similar circumstances. The Theory of Planned Behavior As stated previously, research on health care that utilizes the Theory of Planned Behavior should incorporate the Human Ecologi cal Model. More specific areas of future research include exploring internal barriers and f acilitators of the intention to a ccess health care and the behavior of accessing care. Additionally, research should be done to explore the factors contributing to a persons sense of control in the health care system Finally, the other cons tructs of the Theory should be explored as they pe rtain to the effects they coul d have on a persons perceived behavioral control, intention, and behavior in acce ssing health care. The Perceived Unfairness Model There are many constructs in the Perceived Unfairness Model that were beyond the scope of the present research which need to be expl ored, such as stress res ponses, allostatic load, morbidity, and mortality, particular ly their potential relationships w ith the constructs used in the present research. Not only will such research explain more about how the environment affects physical health, it could shed light on how it a ffects mental health, and have implications for designing programs and policies that develop pe oples coping behaviors and a sense of empowerment. Future research using this model should c onsider using it in conj unction with the Human Ecological Model since some of the basic precept s of the Human Ecological Theory fit well with it. The Human Ecological Model is a theory, in which the perceived is viewed as more important than the actual, the unrea l as more valid than the real, where the motivation that steers

PAGE 96

96 the persons behavior inheres in external objects, activities, persons, and groups, andwhere the content of all these complicated structures re mains unspecified (Bronfenbrenner, 1977, p. 43). Other Areas of Future Research Research examining various asp ects of control could explore numerous directions. This research may include identifying locus of control over ones hea lth, the amount of control one feels in a health care setting, a nd how much control a person feel s in their community. In turn, research could focus on how each of these might influence help-seeking behavior within health care and broader social services, and how each of these might interact and influence physical and mental health. In the present research, participants were asked to indicate wher e they stood in the community in terms of access to health care us ing a ten-rung ladder. Statistical analysis indicated that where they placed themselves on th e ladder was not associated with whether or not they had a regular doctor or hea lth insurance, but it was associated with having the ability to pay cash or credit for services. The lack of signi ficance with health insura nce and a regular doctor could be a result of the small sample size, but this is an area for future research. One might delve more deeply into what factors contribute to where a person places themselves, and how these might be related to factors that affect physical an d mental health. For exam ple, if research finds that the ability to pay for health care services directly and significantl y affects where a person places themselves in their community with respec t to access to health care, what implications might this have in terms of a persons perceptions of attachment to and acceptance within the community, and what implication might this ha ve for public policy how can the heath of a nation be drastically improved when the gaps in incomes continue to widen?

PAGE 97

97 Implications and Recommendations for Alac hua Countys CHOICES Health Services Though this investigation did not examine the CHOICES progr am directly, there are a number of recommendations that could be made from the results. A few are mentioned here. First, implement multi-layered, ecologically based evaluations of the program that include the perspectives of all of those involved in the pr ogram, particularly the enrollees. The evaluation plan should include preand posttest measures, where possible, to account for what changes and what does not, for example, given enrollment in the program or use of specific services. One may start thinking about an inte rnal evaluation of factors aff ecting enrollment by examining the model presented at the end of this chapter (Figure 5-1 on p. 98 ). Similar models could be developed to inform other evaluations and potential consequences of program decisions. Although research is just beginn ing to explore the relationships between social status and health, between emotions and health, and the path ways by which these relationships exist, it is important that policies within the program are awar e that these relationships exist. This could generate policies within the pr ogram that are empowerment base d and flexible; that recognize the enrollees as knowledgeable a bout their health and circumst ances, and can work with the enrollees to identify health concer ns and solutions. This approach is in opposition to the goals of simply educating ignorant enrollees about their health and how their be haviors influence it. Given the results of this study, the program s hould expand its services and providers. The participants felt that they do not actually have many choices and this was associated with feelings of discomfort. Along these lines, outreach needs to be expanded. The participants were not clear what CHOICES was, even after hearing advertisements on the ra dio and television. It was suggested that reasons should be given as to why people should apply and not simply rely on the emergency room. One particip ant succinctly stated that th e reasons for applying need to make sense to those at the bottom (referring to the ladder). Talk about helping the community

PAGE 98

98 and government do not mean anything because the people at the bottom are surviving from day to day. Supplementing with other responses, it could be that the people on the bottom are disenchanted and do not feel sympathy or want to participate in what CHOICES refers to as the community since they feel excluded from it. Implications and Recommendations for Health Care Policy The Human Ecological Model expl ains how other exosystems and other ecological layers affect this part of the exosystem, and it in tur n, it exerts itself on them. However, in the end, health care is focused on taking ca re of individuals. By looking at the health and functioning of individuals in their microsystems, one can see the success of health care policy. As was recommended for CHOICE S, health care policy could foster empowerment-based approaches for medical professionals to work with patients. People should be included in more of the decision-making processes of health care. As one of the participants stated, We make the decisions that are best for us. These are not ig norant people simply ne glecting their health or arbitrarily avoiding medical care. Although the few participants that used the te rms national or universal health care were aware of the unpopularity of su ch terms and used them with hesitation, the basic ideas were the same: health care should be affordable for everyone and the health care infrastructure should be less fragmented; the running around associated with seeking help for medical care or medical bills, and the numerous rejections s hould be curtailed, for example, by having one application that for both adults and children, so the family would be covered. Such policy changes are not impossible to implement, though they may require shif ts in existing power structures and changes in the prof itability of certain ec onomic markets. As seen from the data, it is possible that the current heal th care system discourages peopl e who do not have the ability to pay for care from seeking it. Taking this one step further, this behavior could be reinforced by

PAGE 99

99 avoiding the stress associated with incurring medical bills that they will never be able to pay off. Forestalling such changes may become more cos tly to the economy and hazardous to the health of people, particularly those that pe rceive the injustice of the system. Conclusion In conclusion, there needs to be more resear ch examining the factors that affect health care-seeking behavior, more research examin ing the relationships among the environment, perceptions, and health outcomes. There also ne eds to be evaluations of health care programs and policies that are informed by talking with people and using their knowledge and perceptions, and in turn using these to further research, pol icy making, and program planning as well. All of the factors are interwoven; all of them can learn from and build onto the others. Drastic changes in our health care system call fo r changes in our macrosystem. As a society, it must be recognized that it not onl y takes a village to raise child, but it takes a village to foster development at every age in the lifespan. It must be recognized that things occurring in the various ecological levels, and their interactions, have consequences for individuals and families. It is too idealistic to conceive of a world where no one is disc riminated against. It is not unrealistic to conceive of a world where people have voices in their families and communities. Voices from which they can build resilience a nd strong identities to buffer or counteract the negative consequences that could include persis tent stress and feelings of helplessness and isolation, which could translate to poor physical and mental health. When we understand, we can act. If we know how to ma ke human beings human, we are obligated to do our best to do so (Hamilton & Ceci, 2005, p. 284).

PAGE 100

100 Figure 5-1. Ecological view of fact ors affecting CHOICES enrollment

PAGE 101

101APPENDIX A CONSTRUCT/THEORY/PURPOSE CHART Table A-1. Construct /theory/purpose chart Concept Question Literature/Theory Construct Purpose/Analyze Access 1 1) Aday: potential and realized 2) Health theories: is perceived access generally assumed? (N/A) 1) Purpose of research: Look at perceived access among workers with low-income 2) Get participants definitions of access 3) Get participants access by own definition (2 mo ago and current) Relative access 6 1) Relative status 2) Relative income 3) Discrimination 4) Effects on physical and mental health (N/A) Purpose of research: Does relative access affect perception of control over health? Perceived access to community resources 7 Building on perceived access and relative access concepts (N/A) Should correspond with Questions 1 Control beliefs 11, 13, 15, 17 Theory of Planned Behavior Azjen et al. Perceived behavioral control Identify perceptions of discrimination factors that impede behavior

PAGE 102

102Table A-1. Continued Concept Question Literature/Theory Construct Purpose/Analyze Power beliefs 12, 14, 16, 18, 22 Theory of Planned Behavior Azjen et al. Perceived behavioral control Identify impact of perceived discrimination Perceived behavioral control 19 Theory of Planned Behavior Azjen et al. Perceived behavioral control Should correspond with Questions 112 if discrimination has impact Relative access and control 20 (N/A) (N/A) If people who have more resources have more control resources are a significant factor; May provide evidence that other forms of discrimination are not prominent at this time, although it does not mean they do not play an important role Health and emotions 218 Building on ideas of emotions affecting health (N/A) (N/A) Possible Alternate explanation 29 (N/A) (N/A) To be used in data analysis Is it appropriate to think that because someone enrolls in a health care program they intend to perform additional healthpromoting behaviors?

PAGE 103

103Table A-1. Continued Concept Question Literature/Theory Construct Purpose/Analyze Miscellaneous Questions 30 32 Question 30 Statement to lawmakers Empowerment Question 31 Any additional information offered by participant Question 32 Questions for me (N/A) (N/A) CHOICES Enrollee Health Questionnaire Fourteen Multiple choice/ shortanswer questions (N/A) 1) Perceptions of their health 2) Diagnoses of the following conditions: asthma, diabetes, high blood pressure, congestive heart failure and depression 3) Global satisfaction with their life 4) Frequency of utilization of emergency rooms in the last two years 5) Utilization of hospital inpatient services in the last two years with name of local hospital(s) 6) Frequency of primary health care visits 7) Length of time since respondents last physical and dental exams 8) Primary barrier to utilizing health care services 9) Prior to enrolling in CHOICES, whether of not respondents had a usual source of health care Could be used in data analysis Are: 1) feelings of control over health 2) health emotions 3) perceptions of access 4) perceived behavioral control 5) perceived discrimination related to perceptions of health, diagnoses, global life satisfaction, utilization of health care services, length of time since last physical and dental exams, and/or whether or not they had a usual place of health care? 6) Are the participants responses in agreement with other research on people with low income?

PAGE 104

104APPENDIX B CONCEPT PATTERN Figure B-1. Pattern of potential relationships among concepts

PAGE 105

105 APPENDIX C INTERVIEW SCHEDULE AND KEY 1. Introduce myself 2. Review IRB-Approved Authorized Consent Form 3. Ask if they have any questions before we begin 4. Turn on tape recorder Ive asked you to participate in this research because you are a CHOICES participant. We are conducting this research because it is important to learn what people think and how they feel about issues related to their health and health care. Thats why we are interested in what you think about health care. Remember there aren t any right or wrong an swers; its all your thoughts and ideas. Administer access chart (Appendix D) This is a list of things that ot her people have said are needed to see a doctor or other health care provider. 1a. What else you can think of that is importa nt to getting to see a health care provider? 1b. Could you mark all of the thi ngs that you think are needed? 1c. Of the things you marked, could you rank the t op 5 things, with 1 being the most important, 2 being the next, and so on? (Appendix D) 2. (ATTACH ADDITIONAL CARD TO CHART) (A ppendix D) Keeping all of these things marked in mind, even the ones not ranked, if you th ink back two months ago, which of these did you have and which ones didnt you have? What about now? Administer Percieved Relative Health Care Access Scale (Appendix E) For the next couple of questions you can just answer yes or no. YES NO Do you feel like you could get the care you needed if you were slightly sick, like if you had a cold or the flu? What if you needed if you had a medical emergency? Do you feel like you could get the care you needed if you were depressed? 8. What if you had a problem with drugs or alcohol? Now Im going to describe a scenario and you can ask me whatever you

PAGE 106

106 need to to answer the questions that follow. Marie and John have been having tr ouble breathing lately. Both of them think its time to see a doctor about it. Do you think that one of them will have an easier time getting in to be seen here in Alachua County? (The questions asked could show where the participant places importance in access; may correspond to access chart already completed. Not every question will necessarily be asked and the order of questions cannot be determined either. Questions will be asked based on the participants own follow-up questions and subseque nt responses until information is gathered on each topic.) In a conversation-like way, will also ask about the following: 9a. Do you think theres a difference becau se Marie is a woman and Johns a man? [If no, skip to 9d.] 9b. If there is a difference based on sex, how do you think that makes Marie/John feel? (Will insert whomever fares worse) 9c. What would you have to say to Marie/John ; would you have any advice for her/him? (Will insert whomever fares worse) 10a. What if you learned that Marie has health insurance and John doesnt ? Do you think that would affect who gets to see a doctor first? 10b. [Read if not previously discussed feelings] If there is a difference based on health insurance, how do you think that makes Marie/J ohn feel? (Will insert whomever fares worse) [Read if they have talked about feelings] Would Marie/John would feel differently than what you have already described? 10c. [Read if not previously given advice] What would you have to say to Marie/John; would you have any advice for her/him? (W ill insert whomever fares worse) [Read if already given advice] Would you say something different to Marie/John than what you said before? 11a. [Read only if both sex and health insurance status are factors] Do you think (insert whomever fared worse on sex) being a woman/be ing a man matters more, less or about the same as (insert whomever fared worse on health insurance status)?

PAGE 107

107 12a. So, now we now that Marie has health insu rance and John doesnt. What if we also found out that Marie is an African American and John is white? Do you think their different races/ethnicities would a ffect who gets to see a doctor first here in Alachua County? 12b. Do you think it would be different for any other races or ethnicities? 12c. [Read if not previously discussed feelings] If there is a differen ce based on race/ethnicity, how do you think that makes Marie/John/Pe ople who are of part icipant-identified race(s)/ethnicity(ies) feel? (Will insert whomever fares worse) [Read if they have talked about feelings] Would Marie/John would feel differently than what you have already described? 12d. [Read if not previously given advice] What would you have to say to Marie/John/People who are of participant-id entified race(s)/ethnicit y(ies); would you have an y advice for her/him? (Will insert whomever fares worse) [Read if already given advice] Would you say something different to Marie/John than what you said before? 11b. [Read only if both sex and race/ethnicity are factors] Do you think (insert whomever fared worse on sex) being a woman/being a man ma tters more, less or about the same as (insert whomever fared worse on race/ethnicity)? 11c. [Read only if health insurance status and race/ethnicity are factors] Do you think (insert whomever fared worse on health insurance st atus) being a woman/being a man matters more, less or about the same as (insert whom ever fared worse on race/ethnicity)? 13a. So now we know that Marie has health insura nce and is an African American, but we also learn that her job doesnt pay enough for her to be able to make the co-payments that are required by her insurance, so she would need to find a doctor that would let her make payments for her visit. On the other hand, John, who is white, makes enough money to pay cash for his visit, even though he doesnt have insurance. Do you think their ability to pay will make it easier for one of them to see a doctor? 13b. [Read if not previously discussed feelings] If there is a differen ce based on ability to pay, how do you think that makes Marie/John? (Will insert whomever fares worse) [Read if they have talked about feelings] Would Marie/John would feel differently than what you have already described? 13c. [Read if not previously given advice] What would you have to say to Marie/John would you have any advice for her/him? (W ill insert whomever fares worse) [Read if already given advice] Would you say something different to Marie/John than what you said before?

PAGE 108

108 Ask the following as needed to determine ranking of factors: 11d. [Read only if both sex and income are factors] Do you think (insert whomever fared worse on sex) being a woman/being a man matters more, less or about the same as (insert whomever fared worse on income)? 11e. [Read only if health insurance status and income are factors] Do you think (insert whomever fared worse on health insurance st atus) being a woman/being a man matters more, less or about the same as (insert whomever fared worse on income)? 11f. [Read only if race/ethnicity and income are factors] Do you think (insert whomever fared worse on race/ethnicity) being a woman/being a man matters more, less or about the same as (insert whomever fared worse on income)? 14. Please answer yes or no for whether or not you agree with the following statements: [Stop once their level of control is stated] YES NO I cant do anything to change my health. My health is out of my hands. I have some control over my h ealth, but not a lot of control. I have a lot of control over my health, but not complete control. I have complete control over my health. 15. Looking again at this ladder and imagining absolutely everyone who lives in Alachua County, do you think the people standing on the t op step of the ladder would answer the following statements with yes or no? BRING OUT CUE CARD WITH THE FOLLOWING AND READ THEM OFF (End of Appendix): [Stop once level of control is stated] a. That they cant do anythi ng to change their health. b. That their health is out of their hands. c. Would they say they have some control over their health, but not a lot of control. d. That they have a lot of control over their health, but not complete control. e. Or that they have comp lete control over their health.

PAGE 109

109 16. Repeat Question 15 for people standing on the top step of the ladder un til level of control is stated. Now a couple unrelated questions 17. What do you want to get out of being in CHOICES? [OPEN RESPONSE, BUT I WANT TO BE CLEAR ABOUT THEI R EXPECTATIONS, SO A FOLLOW-UP IS LIKELY.] 18. If you could tell lawmakers one thing about th e health care system in Alachua County what would it be? The last group of questions is about your h ealth and visits to a health care provider. (Administer CHOICES Enrollee Questionnaire) 19. Is there anything else you would like to add before we finish? 20. Do you have any questions for me? Thank you very much for talking with me. I really appreciate it. Have a good day. Cue Card for Questions 15 and 16 a. That they cant do anythi ng to change their health. b. That their health is out of their hands. c. They have some control over thei r health, but not a lot of control. d. That they have a lot of control over their health, but not complete control. e. Or that they have comple te control over their health. INTERVIEW SCHEDULE KEY Question 1: Definition of health care access Question 2: Level of health care access two months ago Question 3: Current level of health care access Question 4: Relative Health Care Access Scale Questions 5: Perception of access to community health care resources Question 9: Control and Powe r Beliefs scenario/discussion

PAGE 110

110 Question 14: Assessing percei ved control over health Question 15: Perception of contro l people on the top step of th e ladder feel over own health Question 16: Perception of contro l people on the bottom step of th e ladder feel over own health Question 17: Expectations about participation in CHOICES Question 18: Statement to lawmakers (Administer CHOICES Enrollee Health Questionnaire) Question 19: Additional information offered by participant Question 20: Answer partic ipants questions, if any

PAGE 111

111 APPENDIX D ACCESS CHART Figure D-1. Factors that contri bute to a participant accessing hea lth care, how the participant ranks these factors, and whet her they have the factors A

PAGE 112

112 Figure D-1. Continued B

PAGE 113

113 Figure D-1. Continued C

PAGE 114

114 APPENDIX E PERCEIVED RELATIVE HEAL TH CARE ACCESS SCALE

PAGE 115

115 APPENDIX F SAMPLE INTRODUCTORY LETTER May 10, 2007 [Enrollees First Name] [Enrollees Last Name] [Street Address] [City], [State] [Zip Code] Dear Ms./Mr. [Enrollees Last Name]: My name is Amanda Holzworth. I am a gradua te student at the Univ ersity of Florida and an intern at CHOICES Health Services Program. I am intere sted in learning about your ideas and experiences with heal th care in Alachua County. As part of a research project I will be doing interviews to find out what people in CHOICES think about health car e in Alachua County. Your an swers would help build an understanding of what people th ink about health care and th eir experiences in Alachua County. Your answers will help look at the impact of CHOICES and also help future program planning. I know that you are busy, but your participation is important to the success of this study. I would like to meet with you to hear about your experiences with heal th care. I will be calling you soon to arrange a time that we could meet. To encourage participation, I will be offering all participants $15. Please keep in mind that although I am working with CHOICES, this money is not from them. Please feel free to contact me with any questions, or for more information, at (352) 271-7305, or via email at anicole8@ufl.edu. Thank you in advance for your time and support. I look forward to meeting with you and learning about your experiences in this community. Sincerely, Amanda N. Holzworth Carolyn S. Wilken, Ph.D., Research Advisor Graduate Student and Associate Professor, FYCS Intern at CHOICES Health Services Program Susan Myers, Ed.S. CHOICES Interim Director

PAGE 116

116 LIST OF REFERENCES Aday, L., Andersen, R. & Fleming, G. V. (1980). Health care in the U.S.: Equitable for whom? London: Sage. Adler, N. E., Epel, E. S., Castellazzo, G. & Ickovi cs, J. R. (2000). Relationship of subjective and objective social status with ps ychological and physiological functioning: Preliminary data in health white women. Health Psychology, 19 (6), 586. Aetna (2001). Living Well Retrieved February 12, 2007, from http://www.aetna.com/ members/health_wellness/liv ing_well/living_well.html Agresti, A. & Finlay, B. (1997). Statistical methods for the social sciences (3rd ed.).Upper Saddle River, NJ: Prentice Hall. American Cancer Society, American Diabetes Association and Ameri can Heart Association. (n.d.). Everydaychoices.org Retrieved February 12, 2007, from http://www.everydaychoices.org American Cancer Society, American Diabetes Association and Ameri can Heart Association. (n.d.). See Your Doctor Retrieved February 12, 2007, from http://www.everydaychoices.org/doctor.html American Diabetes Association. (n. d.). Virtual Grocery Store Retrieved February 12,2007, from http://vgs.diabetes.org/homepage.jsp Azjen, I. & Madden, T. J. (1986). Prediction of goa l-directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experiment al Social Psychology, 22 453. Baicker, K., Chandra, A., Skinner, J. S., & Wennberg, J. E. (2004). Who you are and where you live: How race and geography affect the treatment of Medicare beneficiaries. Health Affairs, 23 33. Bailis, D. S., Segall, A., Mahon, M. J., Chipperf ield, J. G. & Dunn, E. M. (2001). Perceived control in relation to socioeconomic and behavioral resources for health. Social Science and Medicine, 52 1661. Bandura, A. (1986). Social foundations of thought and action Englewood Cliffs, NJ: PrenticeHall, Inc. Belle, D. & Doucet, J. (2003). P overty, inequality, and discrimination as sources of depression among U.S. women. Psychology of Women Quarterly, 27 101. Blue Cross and Blue Shield of Florida, Inc. (2004). Health and Wellness Retrieved February 12, 2007, from http://www.bcbsfl.com/ index.cfm?fuseaction=HealthWellness.home

PAGE 117

117 Borrell, L. N., Dallo, F. J. & White, K. (2006). Education and diabetes in a racially diverse population. American Journal of Public Health, 96 (9), 1637. Bridle, C., Riemsma, R. P., Pattenden, J., Sowden, A. J., Mather, L., Watt, I. S. &Walker, A. (2005). Systematic review of the effectiveness of health behavior interventions based on the transtheoretical model. Psychology and Health, 20 (3), 283. Bronfenbrenner, U. (1977). Lewinian space and ecological substance. In Bronfenbrenner (Ed.) (2005). Making human beings human Thousand Oaks, CA: Sage (p. 41). Bronfenbrenner, U. (1979). The ecology of human development Cambridge, Massachusetts: Harvard University Press. Bronfenbrenner, U. (1988a). Interacting syst ems in human development. In Bronfenbrenner (Ed.) (2005). Making human beings human Thousand Oaks, CA: Sage (p. 67). Bronfenbrenner, U. (1988b). Strengthening fam ily systems. In Bronfenbrenner (Ed.) (2005). Making human beings human Thousand Oaks, CA: Sage (p. 260). Bronfenbrenner, U. (1992). Ecological system s theory. In Bronfenbrenner (Ed.) (2005). Making human beings human Thousand Oaks, CA: Sage (p. 106). Bronfenbrenner, U. (2001). The bioecological th eory of human development. In Bronfenbrenner (Ed.) (2005). Making human beings human Thousand Oaks, CA: Sage (p. 3). Bronfenbrenner, U. (Ed.). (2005). Making human beings human Thousand Oaks, CA: Sage. Brown, E. R., Ojeda, V. D., Wyn, R., & Levan, R. (2000). Racial and ethnic disparities in access to health insurance and health care: A public ation of UCLA Center for Health Policy Research and The Henry J. Kaiser Family Foundation. Brown, K. M. (1999). Theory of reasoned action/Theory of pla nned behavior. Retrieved June 25, 2006, from http://hsc.usf.edu/~kmbrown/TRA_TPB.htm Bruce, B. G., Lennon, M. C. & Dohrenwend, B. P. (1993). Socioeconomic status and depression: The role of occupations involvi ng direction, control and planning. American Journal of Sociology, 98 (6), 1351. Bryman, A. (2004). Social research methods (2nd ed.). Oxford: University Press. Chen, Y., Subramanian, S. V., A cevedo-Garcia, D. & Kawachi, I. (2005). Womens status and depressive symptoms: A multilevel analysis. Social Science and Medicine, 60 49. Chitwood, D. D., Comerford, M., & McCoy, H. V. ( 2002). Satisfaction with access to health care among injection drug-users, other drug users and non-users. Journal of Behavioral Health Services &Research, 29 (2), 189.

PAGE 118

118 Centers for Medicare and Medi caid Services. (December, 2004). Staying Healthy: Medicares Preventive Services (Publication Number CMS-11100) in Prevention Toolkit, Retrieved from February 12, 2007, from http://www.cms.hhs.gov/PrevntionGenInfo/ 01_overview.asp Collins, S. R., Schoen, C., Colasanto, D., & Downey, D. A. (2003). On the edge: Low-wage workers and their health insurance coverage The Commonwealth Fund. Retrieved November 15, 2005, from http://www.cmwf.org/publications/ publications_show.htm?doc_id=221541 Cunningham, P. & Hadley, J. (2004). Expandi ng care versus expanding coverage: How to improve access to care. Health Affairs, 23 (4), 234. de Vaus, D. A. (2001). Research design in social research London: Sage Publications. De Vet, E., De Nooijer, J., De Vries, N. K., & Brug, J. (2005). Stages of change in fruit intake: A longitudinal examination of stability, st age transitions, and transition profiles. Psychology and Health, 20 (4), 415. Egan, G. & Cowan, M. A. (1979). People in systems: A model for development in the humanservice professions and education Monterey, CA: Brooks/C ole Publishing Company. Federman, A. D., Vladeck, B. C., & Siu, A. L. ( 2005). Avoidance of health care services because of cost: Impact of the Medicare savings program. Health Affairs, 24 (1), 263. Fifty-state comparisons: Health insurance coverage for the total population (States2003; U.S. 2004). The Henry J. Kaiser Family Foundation. Retrieved May8, 2006, from http:www.statehealthfacts.org Fishbein, M. and Azjen, I. (1975). Belief, attitude, intention an d behavior: An introduction to theory and research Reading, MA: Addison-Wesley. Frist, W. H. (2005). Overcoming disparities in U.S. health care. Health Affairs, 24 (2), 445. Gallo, L. C., Bogart, L. M., Vranceanu, A. & Ma tthews, K. A. (2005). Socioeconomic status, resources, psychological experiences, and emo tional responses: A test of the reserve capacity model. Journal of Personality and Social Psychology, 88 (2), 386. Garcia, K. & Mann, T. (2003). Fr om I wish to I will: Social-cognitive predictors of behavioral intentions. Journal of Health Psychology, 8 (3), 347. Glanz, K., Lewis, F. M., & Rimer, B. K. (Eds.) (1997). Health behavior and health education: Theory, research, and practice (2nd ed.). San Francisco: Jossey-Bass Publishers. Gulliford, M. & Morgan, M. (Eds.). (2003). Access to health care London: Routledge. Hamilton, S. F. & Ceci, S. J. (2005) Afterword. In Bronfenbrenner (Ed.) Making human beings human : Thousand Oaks, CA: Sage (p. 283).

PAGE 119

119 Harrell, J. P., Hall, S. & Taliaferro, J. ( 2003). Physiological res ponses to racism and discrimination: An assessment of the evidence. American Journal of Public Health, 93 (2), 243.Indigent Care and Trauma Center Surtax, 212.055.4, (2004). Individal state profiles: Florida: Health in surance coverage of non-eldery uninsured by employment status. (States 2003; U.S ., 2004). The Henry J. Kaiser Family Foundation. Retreived May 8, 2006, from http://www.statehealthfacts.org Jackson, B., Kubansky, L. D. & Wright, R. J. (2006). Linking perceived unfairness to physical health: The Perceived Unfairness model. Review of General Psychology,10 (1), 21. Karisen, S. & Nazroo, J. Y. (2002) Relation between racial disc rimination, social class, and health among ethnic minority groups. American Journal of Public Health, 92 (4), 624 630. Kawachi, I., Daniels, N. & Robinson, D. E. (2005). Health disparities by race and class: Why both matter. Health Affairs, 24 (2), 343. Kawachi, I. & Kennedy, B. P. (1997). The relations hip of income inequality and mortality: Does the choice of indicator matter? Social Science and Medicine, 7 ,1121. Kawachi, I., Kennedy, B. P., Gupta, V. & Prot hrow-Smith, D. (1999). Wo mens status and the health of women and men: A view from the States. Social Science and Medicine, 48 21 32. Kawachi, I., Kennedy, B. P. & Wilkinson, R.G. (Eds.) (1999). The society and population reader, volume 1 income inequality and health New York: The New Press. Kennedy, E. M. (2005). The role of the federal g overnment in eliminating health disparities. Health Affairs, 24 (2), 452. Kennedy, B. P., Kawachi, I., Glass, R. & Pr othrow-Smith, D. (1998) Income distribution, socioeconomic status, and self rated health in the United States: Multilevel analysis. British Medical Journal, 317 (7163), 917. Kennedy, B. P., Kawachi, I. & Prothrow-Smith, D. (1996). Income distri bution and mortality: Cross-sectional ecological study of the Robin Hood index in the United States. British Medical Journal, 312 (7037), 1004. Kessler, R. C., Mickelson, K. D. & Williams, D. R. (1999). The prevalence, distribution, and mental health correlates of perceive d discrimination in the United States. Journal of Health and Social Behavior, 40 208. Kossak, S. N. (2005). Exploring the elements of culturally releva nt service delivery. Families in Society,86 (2), 189.

PAGE 120

120 Lasser, K. E., Himmelstein, D. U. & Woolhandler S. (2006). Access to care, health status, and health disparities in the United States and Canada: Results of a cross-national populationbased survey. American Journal of Public Health, 96 (7), 1300. Lavrakas, P. J. (1993). Telephone survey methods: Sampling, selecting, and supervision. (2nd ed). Newbury Park, CA: Sage. Link, Lennon & Dohewnwend (1993). Socioeconomic status and depression: The role of occupations involving, direct ion, control, and planning. American Journal of Sociology, 98 (6), 1351. Lochner, K., Pamuk, E., Makuc, D., Kennedy, B. P. & Kawachi, I. (2001). State-level income inequality and individual mortality risk: A prospective multi-level study. American Journal of Public Health, 91 385. Luszczynska, A. & Schwarzer, R. (2005). Multidim ensional health locus of control: Comments on the construct and its measurement. Journal of Health Psychology,10 (5), 633. Marshall, C. and Rossman, G. B. (2006). Designing Qualitative Research (4th ed.).Thousand Oaks, CA: Sage. Mechanic, D. (2000). Book review essay: Rediscovering the soci al determinants of health Health Affairs Retrieved January 11, 2006, fr om Lexis Nexis database. Montano, D. E., Kasprzyk, D. & Fishbein, M. (19 96). Application of an integrated behavioral model to predict condom use: A prospective study among high HIV risk groups. In Glanz, Lewis & Rimer (Eds). (1997). Health behavior and health education: Theory, research, and practice (2nd ed.). San Francisco: Josse y-Bass Publishers. (p. 102). Moradi, B. & Hasan, N. T. (2004). Arab Am erican persons reported experiences of discrimination and mental health: The me diating role of personal control. Journal of Counseling Psychology, 51 (4), p. 418. Motl, R. W., Dishman, R. K., Ward, D. S., Saunder s, R. P., Dowda, M., Felton, G. &Pate, R. R. (2005). Comparison of barriers self-efficacy and perceived behavioral control for explaining physical activity across 1 year among adolescent girls. Health Psychology, 24 (1), 106. Naar-King, S., Wright, K., Parsons, J. T., Frey, M., Templin, T. & Ondersma, S. (2006).Transtheoretical model and subs tance abuse use in HIV-positive youth. AIDS Care, 18 (7), 283. Nguyen, M. N., Potvin, L. & Otis, J. (1997). Re gular exercise in 30to 60-year old men: Combining the stages of change model and th e theory of planned behavior to identify determinants for targeting heart health interventions. Journal of Community Health, 22 (4), 233.

PAGE 121

121 Payne, N., Jones, F. & Harris, P. (2002). The imp act of working life on health behavior: The effect of job strain on the cogni tive predictors of exercise. Journal of Occupational Health Psychology, 7 (4), 342. Prochaska, J. O. (1979). Systems of psychotherapy: A transtheoretical analysis Homewood, IL: The Dorsey Press. Rosenstock, I. M. (1966). Why people use health services. Milbank Memorial Fund Quarterly, 44 94. Ruggiero, K. M. & Taylor, D. M. (1997). Why minority group members perceive or do not perceive the discrimination that confronts th em: The role of self-esteem and perceived control. Journal of Personality and Social Psychology, 72 (2), 373 389. Schulz, A. J., Gravlee, C. C., Williams, D. R ., Israel, B. A., Mentz, G. & Rowe, Z.(2006). Discrimination, symptoms of depression, and self-rated health among African American women in Detroit: Results from a longitudinal analysis. American Journal of Public Health, 96 (7), 1265. Skinner, J., Weinstein, J. N., Sporer, S. M. & Wennberg, J. E. (2003) Racial, ethnic, and geographic disparities in rates of knee arthroplasty among Medicare patients. New England Journal of Medicine, 349 1350. Strauss, A. and Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed). Thousand Oaks, CA: Sage. Subramanian, S. V., Kawachi, I. & Kennedy, B. P. (2001). Does the state you live in make a difference? Multilevel analysis of self-rated health in the US. Social Science & Medicine, 53 9. The Uninsured and Their Access to Health Care (2005). The Henry J. Kaiser Family Foundation. Retrieved April 14, 200 6, from http://www.kff.org Terry, D. J., Gallois, C. & McCamish, M. (Eds.) (1993). The theory of reasoned action: Its application to Aids-preventive behavior Oxford: Pergamon Press. Wan, T. T. H. and Soifer, S. J. (1974). Determ inants of physician utilizat ion: A causal analysis. Journal of Health and Social Behavior, 15 100. Williams, D. R. & Collins, C. (2001). Racial resi dential segregation: A fundamental cause of racial disparities in health. Public Health Report, 116 404. Williams, D. R. & Jackson, P. B. (2005). Social sources of racial disparities in health. Health Affairs, 24 (2), 325. Williams, D. R., Neighbors, H. W. & Jackson, J. S. (2003). Racial/ethnic discrimination and health: Findings from community studies. American Journal of Public Health, 93 (2), 200 207.

PAGE 122

122 Wu, Z. and Schimmele, C. M. (2005). Racial/e thnic variation in func tional and self-reported health. American Journal of Public Health, 95 (4), 710. Zizzi, S. J., Keeler, L. A. & Watson II, J. C. (200 6). The interaction of go al orientation and stage of change on exercise beha vior in college students. Journal of Sport Behavior, 29 (1) 96 110.

PAGE 123

123 BIOGRAPHICAL SKETCH Amanda Holzworth received her bachelor of sc ience in psychology from the University of Florida in 2003. She received he r master of science in famil y, youth, and community sciences from the same institution in 2007. After graduati on, she will pursue professional interests that allow for a balance with all other aspects of her life.