Yi, Thomas PharmD*; Cocohoba, Jennifer PharmD*; Cohen, Mardge MD†; Anastos, Kathryn MD‡; DeHovitz, Jack A MD, MPH§; Kono, Naoko MPH‖; Hanna, David B MS¶; Hessol, Nancy A MSPH*
The US AIDS Drug Assistance Program (ADAP) was established in 1987 to help HIV-infected patients pay for zidovudine, the first HIV drug treatment. Currently, ADAP is the nation's primary prescription assistance program to help low-income, under- and uninsured HIV-infected patients access life-sustaining medications.1 At the end of 2007, 183,299 people, one third of the US HIV-infected population, were enrolled in ADAP. Ethnic minorities such as blacks and Hispanics represented 60% of enrollees, 72% were uninsured, and 74% had an annual income at or below 200% of the Federal Poverty Level (FPL).1
All ADAP programs are federally funded but most states provide monetary support as well. With the current economic recession, many state ADAP budgets have been reduced. In March 2009, 21 programs experienced budget decreases including eight states that stopped funding altogether. Also, between 2007 and 2008, 40 states experienced an increased number of clients served.1 With rising unemployment, treatment recommendations by the US Department of Health and Human Services (DHHS) for earlier initiation of highly active antiretroviral therapy (HAART),2 and financial deficits for other health insurance programs for low-income populations, the demand for ADAP services is expected to rise.1
In addition to state variation in funding, each state has different ADAP patient eligibility criteria and drug formularies. Eligibility criteria likely reflect the demographic characteristics of each state's population of HIV-infected individuals and the state's budgetary restrictions. In 2008, the eligibility requirement for each state's ADAP ranged from 200% of the FPL in Idaho, Iowa, Louisiana, Nebraska, Oklahoma, Oregon, and Texas to 500% in Arkansas, Delaware, District of Columbia, Maine, Maryland, New Jersey, and Ohio.1 Since July 2007, each state has been required to include at least one medication from each antiretroviral class on its ADAP formulary. However, non-HIV medications covered by each state differed as ADAP formularies in 2008 ranged from 28 drugs in Idaho to 466 drugs in New York.1 For example, medications to treat hypertension, shown to be prevalent among 29% to 34% of HIV-positive patients,3,4 were covered by some states such as New York but not in others such as California or Illinois. In 2008, non-HIV medication accounted for 31% of all prescriptions filled for ADAP nationally.1
Previous studies have identified black ethnicity,5-13 female gender,5,8,9 lower educational attainment,5,10,13 injection drug use,5,7,9,10,12,13 and being uninsured11,13,14 as factors associated with decreased use of HAART. However, there are a limited number of studies that specifically review ADAP and its association with HAART use and few, if any, that explore state-by-state program variability. To our knowledge, no studies have investigated the relationship between ADAP and use of nonantiretroviral medications.
The goal of this study was to investigate the impact of ADAP enrollment on the use of prescriptions for both HIV infection and hypertension. We choose to include hypertension because it is a common condition among people with HIV infection2,3 and enrollment in ADAP may affect both access and use of antihypertensive medications. The aims of the study were to evaluate whether enrollment into ADAP, including state differences in ADAP eligibility criteria and drug formularies, was associated with HAART and antihypertensive medication use by comparing HAART-eligible women enrolled in ADAP with women who relied on other forms of health insurance for medication coverage.
Data from the US Women's Interagency HIV Study (WIHS) were used for this investigation. In brief, recruitment of HIV-infected and uninfected women into the WIHS occurred in 1994 to 1995 and again in 2001 to 2002 for a total of 2791 HIV-infected and 975 HIV-uninfected women. Data for the WIHS were collected from the following six centers: Brooklyn, Bronx, Washington, DC, Chicago, the Los Angeles area, and the San Francisco Bay Area. Recruitment occurred at a variety of venues, including HIV care and testing sites, drug and tuberculosis treatment programs, community-based organizations, and sexually transmitted disease clinic programs. Women were seen for core study visits twice a year and data were collected with a standardized interview-based questionnaire. Detailed information about the WIHS study methodology, quality assurance, and baseline characteristics of enrollees can be found in previously published literature.15,16 Study protocols were reviewed and approved by the Institutional Review Boards, and written informed consent was obtained from the participants.
The inclusion criterion for both analyses was enrollment in five of the six WIHS sites (N = 2393) from California, Illinois, and New York during Visit 28 (April through September 2008). Participants from the Washington, DC, area were excluded (N = 419), because this site has participants from more than one jurisdiction (the District of Columbia, Maryland, and Virginia). For the use of HAART analyses, only women who were clinically eligible for HAART, based on the 2008 DHHS HIV treatment guidelines, were included (N = 1139).17 Using longitudinal WIHS data, we identified women who were HAART-eligible based on a history of or current 1) CD4+ count of less than 350/mm3; 2) AIDS-defining illness18; and/or 3) HAART use. Women who currently or have had a history of HAART use were included because HAART should not be discontinued once initiated to prevent resistance. Pregnant women at Visit 28 (N = 45) were excluded, because HIV and hypertensive management differ in pregnant and nonpregnant women.
For the use of antihypertensive medication analyses, we used a subset of women from the first study group. This subset only included women who were clinically defined as hypertensive based on the latest guidelines by the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VII)19 (N = 395). Using longitudinal WIHS data, we identified women with hypertension based on having: 1) a history of diastolic blood pressure 90 mm Hg or greater over two subsequent visits (approximately a 6-month interval); 2) a history of systolic blood pressure 140 mm Hg or greater over two subsequent visits; and/or 3) report of current use of one or more antihypertensive medications.19 Women with a history of antihypertensive use were not included because lifelong use of hypertensive medications is not indicated for all patients with hypertension.
For both analyses, we included women who currently reported taking the medications of interest, HAART, or antihypertensives. This was done because women may have had clinical indications for use of these medications that were diagnosed outside of the WIHS and therefore their disease was controlled during the time of data collection.
Dependant and Independent Variables
The primary dependent variables for this analysis were self-reported current use of HAART (yes or no) and self-reported use of antihypertensive medication (yes or no). We categorized reported HIV medications as HAART using guidelines published by the DHHS at the time of the study visit.17 Combination antiretroviral therapy that was not defined as HAART by the DHHS, monotherapy, or no therapy were all categorized as non-HAART use. Indications for antihypertensive therapy, described previously, were based on guidelines by the JNC VII.19
Our primary independent variables were self-reported enrollment into ADAP (yes or no) and state of residence (California, Illinois, or New York). Type of health insurance was categorized in four mutually exclusive groups: private, Medicare, Medicaid (Medi-Cal in California), and uninsured. Those with private or student health insurance as well as those who reported insurance that could not be classified were categorized in the private category. Because ADAP enrollment may or may not be related to type of health insurance, ADAP was maintained as a separate variable from health insurance.
Sociodemographic factors included in the analysis were race/ethnicity (white non-Hispanics, white Hispanics, Hispanic and non-Hispanic blacks, and other), age (per decade), educational attainment (categorized as completion of some high school, a high school diploma, and some college or more), employment, living situation, and household income (categorized as a yearly income of $12,000 or less, $12,000-$36,000, or greater than $36,000).
Health factors that were assessed include depressive symptoms (a binary indicator of a score of 16 or higher on the Center for Epidemiological Studies Depression Scale)20; indicators of alcohol use (nonuse, light drinker [less than three drinks per week], moderate [three to 13 drinks per week], or heavy [greater than 13 drinks per week] in the past 6 months); individual indicators of crack, cocaine, methadone, or heroin use (yes or no within the past 6 months); and concurrent CD4+ cell count and HIV RNA viral load.
Contingency table analyses were performed to compare the distribution of participant characteristics by state and P values were based on chi-square or Fisher exact tests. Unadjusted logistic regression models assessed the association between the independent variables and the two dependent variables, HAART nonuse and antihypertensive nonuse. We then performed adjusted backward stepwise elimination logistic regression models to eliminate nonsignificant (P > 0.05) independent variables. Our primary independent variables, ADAP enrollment and state of residence, were retained in the stepwise models regardless of their level of significance. Statistical analyses were performed using SAS software Version 9.2 (SAS Institute Inc, Cary, NC).21
The total number of HAART-eligible women who met our inclusion criteria was 1094 (Table 1). Because two WIHS centers are located in California and two in New York, 40% of our population came from California study sites, 45% from New York, and only 15% from Illinois. Across all states, the majority of women was older than 40 years old and categorized themselves as black. Approximately two thirds of the women had less than a college education, were not married and/or living alone, and reported being unemployed. Half earned less than $12,000 per year and Medicaid was the dominant form of health insurance (50%) in our population. Approximately 24% of the women were enrolled in ADAP. For health-associated factors, 37% reported depressive symptoms, 9% used illicit drugs, 36% reported moderate to high alcohol use, 39% smoked, and 13% used marijuana. Furthermore, 46% had CD4+ cell counts above 500/mm3 and 62% had HIV viral loads that were undetectable (less than 80 copies). Overall 74% of the women who were clinically eligible for HAART reported using HAART regimens, 2% were on non-HAART antiretroviral therapy, and 24% were not on any form of antiretroviral therapy. Between states, there were statistically significant differences in the following: race/ethnicity, living situation, yearly household income, health insurance type, ADAP enrollment, depressive symptoms, illicit drug and alcohol use, cigarette smoking, and marijuana use.
The total number of HAART- and antihypertensive-eligible women who met our inclusion criteria was 395 (Table 2). Overall, baseline population characteristics were similar to the initial HAART analysis population. Among this population, 77% of those who had indications for antihypertensive therapy were on at least one medication for hypertension, whereas 23% were not.
Between states, distributions of race/ethnicity, annual household income, type of health insurance, ADAP enrollment, depressive symptoms, and marijuana use significantly varied.
Relationship of AIDS Drug Assistance Program Enrollment, Study Site, Age, Race/Ethnicity, Income, and Alcohol Abuse on Highly Active Antiretroviral Therapy Use
The unadjusted and adjusted odds ratios for nonuse of HAART are shown in Table 3. After adjustment, ADAP enrollment, age, race/ethnicity, income, and alcohol use were found to be statistically significant (P ≤ 0.05) with HAART nonuse. Women without ADAP were more than two times more likely not to be on a HAART regimen (odds ratio [OR], 2.35; confidence interval [CI], 1.49-3.71), whereas the state of study site showed no association.
Older women were less likely not to use HAART than younger women per decade of age (OR, 0.74; CI, 0.61-0.89). Compared with blacks, Hispanic whites (OR, 0.51; CI, 0.29-0.92) and non-Hispanic whites (OR, 0.21; CI, 0.10-0.44) were less likely not to report taking HAART. Compared with those with an annual income less than $12,000, women with annual incomes from $12,000 to $36,000 were 30% less likely not to be HAART users (OR, 0.70; CI, 0.50-0.99). Women who reported moderate alcohol use were two times more likely not to use HAART (OR, 2.08; CI, 1.19-3.65) and those who reported heavy use were almost four times more likely not to be on HAART (OR, 3.75; CI, 1.59-8.84) compared with alcohol abstainers.
Relationship of AIDS Drug Assistance Program, Study Site, Race/Ethnicity, Income, Smoking, and Health Insurance on Antihypertensive Use
The unadjusted and adjusted regression results for nonuse of antihypertensive medication are shown in Table 4. After adjustment, race/ethnicity, income, smoking, and type of health insurance were shown to have a statistically significant association (P ≤ 0.05) with antihypertensive nonuse. Although ADAP coverage and study site state did not show statistically significant relationships with antihypertensive use, ADAP nonenrollment approached significance (P = 0.07) with a trend toward a greater likelihood of nonuse (OR, 2.37; CI, 0.93-6.03).
Compared with blacks, women categorized as other ethnicities were nearly five times more likely not to use antihypertensive medication (OR, 4.76; CI, 2.48-9.14). Women with an annual income greater than $36,000 were 80% less likely not to use antihypertensive drugs in relation to those who made less than $12,000 per year (OR, 0.20; CI, 0.06-0.71). Smokers were 80% more likely not to report taking antihypertensive therapy (OR, 1.80; CI, 1.03-3.16) than nonsmokers. In comparison to those with private insurance, women on Medicare (OR, 0.22; CI, 0.08-0.62) or Medicaid (OR, 0.23; CI, 0.09-0.60) were less likely not to use antihypertensive medications.
The results of our study show that ADAP enrollment, regardless of state of residence, was associated with increased HAART use among clinically eligible HIV-infected women. Also, we found that HAART- and antihypertensive-indicated women enrolled in ADAP had increased but nonsignificant odds of antihypertensive medication use compared with women not enrolled in ADAP. To our knowledge, this is the first study to suggest that ADAP enrollment may increase the use of nonantiretroviral prescription drugs. Additionally, the breadth of data in this study presents a unique opportunity to assess state differences in ADAP enrollment criteria and formularies and we found that state of residence may have a lesser role in use and nonuse of HAART than other factors.
Although New York had the most lenient income eligibility criteria for ADAP (below 423% FPL) compared with California and Illinois (below 400% FPL),1 women in New York did not have significantly different HAART use rates compared with the other states. California had the largest percentage of women enrolled (42.8%), whereas Illinois and New York had much lower rates at 13.7% and 10.9%, respectively. This likely reflects each state's policy toward Medicaid and ADAP coenrollment. California allows Medi-Cal (the state's form of Medicaid) and ADAP coinsurance, whereas New York and Illinois mostly allow only patients without health insurance to enroll into ADAP. Therefore, although New York had the broadest financial eligibility for ADAP enrollment, differences in coinsurance criteria complicated our analysis.
Other factors including age, ethnicity, income, and alcohol use were found to be associated with HAART use. Non-Hispanic and Hispanic white women had higher likelihood of HAART use when compared with blacks. These findings are similar to many studies,5-9 including those done on the WIHS population,10-13 which consistently have shown that health disparities by ethnicity still exist. Other studies support our finding that increased alcohol consumption is associated with decreased likelihood of antiretroviral use11; however, it is unclear whether alcohol use is related to patient factors such as noncompliance or providers' unwillingness to prescribe therapy. In our study, there was a dose-response relationship; higher consumption was more strongly associated with non-use of HAART use, which leads us to believe that this is a strong predictor. Women with higher income were more likely to report HAART use. This may reflect increased access to care and ability to pay for medication.
Although ADAP enrollment increased the likelihood of using HAART, its relationship with antihypertensive medication use was of borderline statistical significance (P = 0.07), attributable in part to the smaller sample size for this subgroup analyses. State variability was not associated with antihypertensive drug use although New York's ADAP included hypertension medications on its formulary.1 California's ADAP does not cover hypertension medication and as a result, enrollees may access these drugs using Medi-Cal. However, New York enrollees cannot have any other forms of insurance and as a result may use ADAP to access antihypertensive medication when they otherwise could not. Like the HAART analysis, ADAP eligibility policies regarding coinsurance complicated our study.
Although the association between ADAP enrollment and antihypertensive use was of borderline significance, women with Medicaid and Medicare insurance were more likely to use hypertension medication than those with private insurance. A review of data within the Women's Health Initiative found that women insured by Medicaid had higher treatment rates and better outcomes compared with those with private insurance and only Medicare.22 The author hypothesized that differences in medication coverage and/or age distributions could explain this disparity. Therefore, other government-sponsored health and prescription insurance programs may play a larger role than ADAP in terms of antihypertensive use, even in HIV-infected women.
Similar to HAART use, women with higher incomes were more likely to use antihypertensive medication. In addition, cigarette smokers were less likely to use hypertension medications. This finding is particularly troublesome because smokers who are hypertensive are at significantly higher risk of developing cardiovascular complications.23 Studies have shown that smokers may have decreased adherence to medications,24 which may explain this population's decreased use of antihypertensive medication.
In contrast to HAART use, blacks were more likely to use antihypertensive medication compared with non-Hispanic whites and other ethnicities. Prior studies have reinforced our finding but also report that although blacks are more likely to use antihypertensive medication than non-Hispanic whites, rates of controlled hypertension are lower.25-27
The juxtaposition between ethnicity and HAART and antihypertensive drug use warrants further investigation into the reasons why these disparities exist. A review of HIV care in minority populations by Cargill and Stone theorized that patient satisfaction with their care, prescriber bias, and patient-physician racial discordance could be possible provider-related reasons for HIV treatment disparities.28 They also found that minorities needed more information and time to make HIV-related treatment decisions. Stigma related to HIV, which can lead to delayed seeking of treatment and even denial of infection, has also been thought to influence antiretroviral treatment use and adherence.28 These reasons highlight the need for better outreach and education to minority populations about the benefits of HAART and to destigmatize HIV as a disease.
One limitation of this study is the cross-sectional design. Thus, the results represent only one point in time and we are unable to determine a temporal relationship between enrollment into ADAP and use of HAART or antihypertensive medication. However, longitudinal data exist and could be the subject of a future investigation. Although longitudinal data were used to determine our study population and eligibility criteria, health outcomes, although collected by WIHS, were not included within this analysis. Therefore, we did not determine whether ADAP's influence on HAART and antihypertensive medication use are also predictive of improved health results such as elevated CD4+ cell counts and blood pressure control. Another limitation is that other unmeasured factors that may influence medication use such as prescriber patterns and frequency of clinic visits were not assessed. In addition, these results are based on HIV-infected women who were recruited from large urban centers in three states in the United States and may not be representative of HIV-infected men or women residing in nonurban areas. Finally, our study was done with mostly self-reported data and consequently could be subject to participant bias. However, highly trained interviewers using standardized, interview-based questionnaires were used to collect the most accurate data possible.
In summary, in light of recent and proposed funding cuts for ADAP and projected increase in demand of ADAP services, we provided evidence that this program was strongly associated with better HAART medication use. We also found that populations that constitute the majority of ADAP enrollees, those with lower income and of black descent, had decreased HAART use compared with those with higher incomes and of nonblack descent. As a result, state ADAPs should be continued to improve antiretroviral use in these at-risk populations. We found that state of study site was not associated with increased likelihood of HAART or antihypertensive use. However, ADAP enrollment was associated with an increased, but nonsignificant, likelihood of blood pressure medication use, whereas Medicare and Medicaid were strongly associated with increased use. Therefore, government-funded programs that provide prescription drug coverage such as ADAP may play a valuable role in promoting increased access and use of essential medications for chronic diseases for underserved HIV-infected women.
Data in this manuscript were collected by the Women's Interagency HIV Study (WIHS) Collaborative Study Group with centers (Principal Investigators) at New York City/Bronx Consortium (Kathryn Anastos); Brooklyn, NY (Howard Minkoff); Washington, DC, Metropolitan Consortium (Mary Young); The Connie Wofsy Study Consortium of Northern California (Ruth Greenblatt); Los Angeles County/Southern California Consortium (Alexandra Levine); Chicago Consortium (Mardge Cohen); Data Coordinating Center (Stephen Gange).
1. National ADAP Monitoring Project Annual Report
. The Henry J. Kaiser Family Foundation web site. Available at: www.kff.org/hivaids/upload/7861.pdf
>. April 2009. Accessed May 5, 2009.
3. Gazzaruso C, Bruno R, Garzaniti A, et al. Hypertension among HIV patients: prevalence and relationships to insulin resistance and metabolic syndrome. J Hypertens. 2003;21:1377-1382.
4. Jung O, Bickel M, Ditting T, et al. Hypertension in HIV-1-infected patients and its impact on renal and cardiovascular integrity. Nephrol Dial Transplant. 2004;19:2250-2258.
5. Andersen R, Bozzette S, Shapiro M, et al. Access of vulnerable groups to antiretroviral therapy among persons in care for HIV disease in the United States. Health Serv Res. 2000;35:389-416.
6. Baillargeon J, Giordano, TP, Rich JD, et al. Accessing antiretroviral therapy following release from prison. JAMA. 2009;301:848-857.
7. Keruly JC, Conviser R, Moore RD. Association of medical insurance and other factors with receipt of antiretroviral therapy. Am J Public Health. 2002;92:852-857.
8. King WD, Minor P, Ramirez Kitchen C, et al. Racial, gender and geographic disparities of antiretroviral treatment among US Medicaid enrollees in 1998. J Epidemiol Community Health. 2008;62:798-803.
9. Gebo KA, Fleishman JA, Conviser R, et al. Racial and gender disparities in receipt of highly active antiretroviral therapy persist in a multistate sample of HIV patients in 2001. J Acquir Immune Defic Syndr. 2005;38:96-103.
10. Cook JA, Cohen MH, Grey D, et al. Use of highly active antiretroviral therapy in a cohort of HIV-seropositive women. Am J Public Health. 2002;92:82-87.
11. Lillie-Blanton M, Stone VE, Snow Jones A, et al. Association of race, substance abuse, and health insurance coverage with use of highly active antiretroviral therapy among HIV-infected women, 2005. Am J Public Health. 2010;100:1493-1499.
12. Cohen MH, Cook JA, Grey DD, et al. Medically eligible women who do not use HAART: the importance of abuse, drug use, and race. Am J Public Health. 2004;94:1147-1151.
13. Cook JA, Cohen MH, Grey D, et al. Use of highly active antiretroviral therapy in a cohort of HIV-seropositive women. Am J Public Health. 2002;92:82-87.
14. Cunningham WE, Markson LE, Andersen RM, et al. Prevalence and predictors of highly active antiretroviral therapy use in patients with HIV infection in the United States. J Acquir Immune Defic Syndr. 2000;25:115-123.
15. Barkan SE, Melnick SL, Preston-Martin S, et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology. 1998;9:117-125.
16. Bacon MC, von Wyl V, Alden C, et al. The Women's Interagency HIV Study: an observational cohort brings clinical sciences to the bench. Clin Diagn Lab Immunol. 2005;19:1013-1019.
18. Centers for Disease Control and Prevention. 1993 Revised Classification System for HIV Infection and Expanded Surveillance Case Definition for AIDS Among Adolescents and Adults. Morbidity and Mortality Weekly Report. Available at: www.cdc.gov/mmwr/preview/mmwrhtml/00018871.htm
>. December 18, 1992. Accessed April 4, 2010.
19. US Department of Health and Human Services. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. JNC 7 Full Report. Available at: www.nhlbi.nih.gov/guidelines/hypertension/jnc7full.pdf
>. August 2004. Accessed March 8, 2010.
20. Bozzette SA, Hays RD, Berry SH, et al. Derivation and properties of a brief health status assessment instrument for use in HIV disease. J Acquir Immune Defic Syndr. 1995;8:253-265.
21. SASInstitute Inc. SAS/STAT User's Guide, Version 9.2. 2008.
22. Oparil S. Women and hypertension: what did we learn from the Women's Health Initiative? Cardiol Rev. 2006;14:267-275.
23. Ockene IS, Miller NH. Cigarette smoking, cardiovascular disease, and stroke. A statement for healthcare professionals from the American Heart Association. Circulation. 1997;96:3243-3247.
24. Lavigne M, Rocher I, Steensma C, et al. The impact of smoking on adherence to treatment for latent tuberculosis infection. BMC Public Health. 2006;6:66.
25. Giles T, Aranda JM, Suh DC, et al. Ethnic/racial variations in blood pressure awareness, treatment, and control. J Clin Hypertens. 2007:345-354.
26. Gu Q, Burt VL, Paulrose-Ram R, et al. Gender differences in hypertension treatment, drug utilization patterns, and blood pressure control among US adults with hypertension: data from the National Health and Nutrition Examination Survey 1999-2004. Am J Hypertens. 2008;21:789-798.
27. Hertz RP, Unger AN, Cornell JA, et al. Racial disparities in hypertension prevalence, awareness, and management. Arch Intern Med. 2005;165:2098-2104.
28. Cargill VA, Stone VE. HIV/AIDS: a minority health issue. Med Clin North Am. 2005;89:895-912.
© 2011 Lippincott Williams & Wilkins, Inc.