Gender differences in knowledge and perceptions of HIV resources among individuals living with HIV in the Southeast.

Objectives: Ancillary services have been associated with beneficial health utilization outcomes among individuals infected with the human immunodeficiency virus (HIV), including greater retention in medical care and greater likelihood of antiretroviral use. Our primary objectives were to examine gender differences in barriers to ancillary services among people living with HIV in the Southeastern United States.

Methods: Survey and chart abstraction data were collected from six tertiary infectious diseases clinics in the Southeast. Using multivariate analyses, we examined the relationship between gender and 1) knowledge of how to access HIV and acquired immunodeficiency syndrome (AIDS) resource information and 2) opinions about the helpfulness of local services for people with HIV/AIDS.

Results: Women were less knowledgeable about HIV/AIDS resources and rated local services less favorably than men. Middle-aged and older African-American women rated local services as less helpful than other survey participants did.

Conclusions: These findings indicate a need for outreach services that are designed to address the specific needs of older African-American women, and women in general.

Key Words: acquired immunodeficiency syndrome, African-American, ancillary services, human immunodeficiency virus, women

**********

People living with human immunodeficiency virus (HIV) often experience multiple psychosocial problems, including economic constraints, inadequate housing options, lack of transportation, and an inability to obtain case management, mental health treatment, and substance abuse treatment. (1-10) The absence of such necessary resources has been associated with poorer health outcomes, including delay of care (1,5) and lower levels of adherence. (11-15) Psychosocial needs have been found to be more prevalent for women with HIV, (16-18) potentially placing them at greater risk for poor medical outcomes. (10,17) A number of studies have identified unequal medical utilization patterns and health outcomes among women with HIV infection when compared with men with HIV infection, including delay in care, (4,19-21) shorter survival times after acquired immunodeficiency syndrome (AIDS) diagnosis, (4,21-23) and lower utilization of antiretroviral medications. (24-27)

Ancillary services, such as case management, transportation assistance, and housing services, can assist individuals in meeting nonclinical needs, thus reducing obstacles to care so that medical attention can become a higher priority. Recent research has shown that receipt of ancillary services is associated with a reduction in unmet needs (3,17) and positive health-related outcomes, including prescription of a protease inhibitor, improved adherence and retention in primary care, (17,28) and higher likelihood of being on HIV medications. (3) However, if people living with HIV are unaware of services, or are unable to access them, they will not be able to benefit from the available ancillary services. Because ancillary services for persons with HIV and AIDS have not traditionally targeted women, and because women often experience higher levels of poverty and greater stigma attached to HIV disease, they may be more likely to experience barriers to ancillary services and to their potential benefits. (10,29)

This study focuses on barriers to receiving HIV/AIDS services--including lack of resource knowledge and lack of confidence in these services--among women living with HIV/AIDS in predominantly rural areas of the South. The southern region of the country continues to experience an increase in the estimated number of new AIDS cases, while other regions experience declines or relatively stable levels. (30) Consistent with this overall trend, AIDS surveillance data indicate that an increasing proportion of AIDS cases among women are occurring in the southern region of the country (41%). (31) In the South, women living in rural areas have been affected by AIDS along with their urban counterparts, as this region of the country has the largest proportion of women with AIDS in rural areas. (32,33) This study describes the characteristics of women with HIV infection who attended one of six Infectious Diseases clinics in the Southeast, and examines whether these women experienced greater barriers than men to accessing ancillary services. In addition, the study evaluates whether race and age mediate the relationship of gender and barriers to service. Because of additional stigma and discrimination, women of nonwhite race and those at an older age may be particularly at risk for barriers to ancillary services. Acquiring a better understanding of the specific barriers to care that women face is critical to developing interventions to improve their access to ancillary services and medical care.

Materials and Methods

Sample

Six academic medical center Infectious Diseases clinics, four in North Carolina and one each in South Carolina and Alabama, were selected for participation in a larger study. (34) The clinics were chosen on the basis of similarity to one another in their provider and patient populations. All the clinics were located in urban areas and provided care for the surrounding rural counties. The clinics provided care for the vast majority of Medicaid-eligible patients in the counties where they were located, as well as the counties surrounding them. Institutional review board approval was received from each site, and a federal certificate of confidentiality was obtained for the study.

Study recruitment was initiated in the summer of 1997 and involved asking consecutive HIV-positive, Medicaid-eligible patients at the clinics if they would participate in a telephone survey and have their medical records examined. Of the 882 patients approached to participate in the study, 833 (94%) consented. From among those who consented to participate and were living at the time of data collection (n = 779), 542 patient surveys were conducted (70% response rate) between August 1997 and July 1998 by the survey firm Louis Harris and Associates. Research nurses conducted medical record abstractions for the year before the date of each patient's survey. The majority of subjects who were not interviewed could not be located (28%). Two percent of the sample refused the interview. Several subjects were identified as attending more than one clinic so their interviews were eliminated, leaving a sample size of 526. No statistically significant differences in race, gender, age, viral load, CD4 count or stage of disease progression were detected between survey respondents and nonrespondents. (34)

Data

The Southeast HIV Patient Survey included questions about each respondent's HIV/AIDS history and symptoms, (35,36) mental health, (37) detailed demographic information, use of illicit substances, and access and barriers to medical care and ancillary services, including knowledge of how to obtain resource information and perception of local services for people with HIV/AIDS. Medical chart abstractions were performed using a standardized form that included clinical history, recent health status, and health services use.

Statistical analysis

Descriptive statistics, including means, medians, and proportions, were used to provide information about the sample. Bivariate statistics, including Pearson [chi square] and t tests, were calculated to provide information regarding gender differences in demographic and social characteristics, perception of local services, and HIV/AIDS resource knowledge.

Multivariate regression analysis was used to examine whether women had different knowledge and perceptions of HIV/AIDS resources. Knowledge of how to access resource information was assessed by the question, "If you have a question about where or how to get services or resources for people with HIV/AIDS, do you know where you can get this information, or not?" Response categories were "know where to get information" and "do not know where to get information." Logistic regression analysis was used to evaluate the relationship between gender and knowledge of how to access resource information. Control variables were selected for the final models if there was a theoretical reason for their inclusion. These variables were level of education, having a case manager, time since HIV diagnosis, mental health status (SF-36 index), (37) having a child under 18 years of age in the home, distance (miles) to receive care at the infectious diseases clinic, and health status based on the Centers for Disease Control (CDC) categories "asymptomatic," "minor symptoms," and "AIDS." (38) Odds ratios were calculated to describe the association of gender and knowledge of how to access HIV/AIDS resources.

To determine whether women differed from men in their perception of how helpful local HIV/AIDS services were, additional multivariate analyses were performed. Perception of local services was ascertained by the survey question, "Overall, how helpful would you say local agencies and organizations are in helping people with HIV/AIDS?" Response categories included "very helpful," "helpful," "neither helpful nor unhelpful," "unhelpful," and "very unhelpful." Respondents could also select a response of "not applicable" for this question, indicating that they had no experience with local HIV/AIDS services. Six percent of survey respondents (n = 34) selected this response. Selection of "not applicable" was not significantly associated with race, gender, or age using Pearson [chi square] tests.

Since the outcome of interest had more than two categories that could be ranked in a specific order, ordered logit models were used to discern the relationship between gender and opinion of local HIV/AIDS services. (39) This method was used instead of logistic regression because collapsing the categories from five to two could result in the loss of more specific information about respondents' opinions. Ordered logit is preferable to ordinary least-squares (OLS) regression because OLS forces the arbitrary labeling of categories, such as "very helpful" = 1 and "helpful" = 2. The problem with this method is that with different labelings, such as 10 for "very helpful," different estimates would result. (40) With ordered logit, the values given to a category do not influence the analysis results.

Control variables included level of education, having a case manager, distance to the Infectious Diseases clinic, mental health status, having a child under 18 years of age in the home, and health status based on the CDC categories. (38) Because the use of odds ratios for ordered logit is impractical, predicted probabilities were calculated to describe the strength of the relationship between gender and opinion of local services. (40)

To assess whether age and race modify the relationship of gender to knowledge of how to access services and opinion about local services, the logistic regression and ordered logit models were re-estimated including interaction terms for race and age with gender. Race was defined as African-American or not, and age was split at the median (and mean) age for women, which was 36 (range 18-64). Likelihood ratio tests (41) were performed to assess whether the interaction terms significantly contributed to the models. The likelihood ratio test compares the likelihood ratios between the model with interaction terms and the model without the interaction terms to assess the contribution of the added variables.

Results

Description of women and men in the patient survey sample

Demographic and behavioral characteristics. The majority of survey respondents (65%) were African-American, 29% were white, and 6% were Native American or other race. Just over one-third of respondents (36%) were female. In addition, 26% of respondents did not graduate from high school and over half (52%) reported incomes less than the poverty level. (42) Bivariate analysis revealed a number of statistically significant differences between male and female participants. Participant characteristics and gender comparisons are shown in Table 1. Women were more likely to be African-American (P < 0.0001), to have less than a high school education (P = 0.008), to be heterosexual (P < 0.0001), and to have children under 18 years of age in their home (P < 0.0001). Men reported higher levels of disability (P = 0.004), and were more likely to have ever used illicit substances (P < 0.001) and to have used illicit substances in the previous six months (P < 0.001). Most women, 82%, resided in rural areas in comparison to 72% of men (P = 0.011).

Barriers to medical and social services. The HIV-infected women reported a number of barriers to accessing ancillary services and medical care. A substantial minority of women (38%) were not aware of how to access resource information if they had a question and 22% of the women did not find local services to be helpful. Furthermore, just over one-third of the women did not have an HIV case manager, 19% reported difficulty obtaining transportation to medical appointments in the last year, and 16% reported having no medical insurance. The vast majority (92%) reported having a regular medical provider.

When compared with men, women reported less knowledge of how to access resource information (P = 0.02) and rated local services to be less helpful (P = 0.03). There were no gender differences in having an HIV case manager, experiencing difficulty obtaining transportation, or having medical insurance.

Multivariate regression analyses

HIV/AIDS resource knowledge. Women were significantly less likely to know how to access information about HIV/AIDS services and resources compared with men. Results of the logistic regression are found in Table 2. The odds ratio for being female was 0.42 (P = 0.002), indicating that women were over two times less likely to know how to access resource information. Longer time since HIV diagnosis, having an HIV case manager, and reporting some college education were also significantly associated with knowing how to access HIV/AIDS resource information.

Perception of local services. In the ordered logit analysis, women gave significantly lower ratings for the helpfulness of local services (P = 0.042). The ordered logit results are summarized in Table 3. The predicted probabilities further describe the strength of the relationship between gender and opinion of local services. For example, an African-American woman over 35 with the characteristics of a high school education, no children under 18 years of age in the home, no HIV case manager, median distance to medical care and median mental health score, and CDC category of "AIDS," had a predicted probability of not reporting services to be helpful of 0.21, whereas a male with similar characteristics had a predicted probability of not reporting services to be helpful of 0.13. Having a case manager was associated with having a more favorable perception of local services, whereas poorer mental health status and living at a greater distance to the Infectious Disease clinic where they received care were associated with a less favorable perception of local services.

Modifying effects of age and race on the associations of gender with resource information and perception of local services. In the logistic regression model containing interaction terms for age and race with gender, these interaction terms were not significantly associated with knowledge of how to access resource information. In addition, the likelihood ratio test that compared the model with interaction terms to the model without interaction terms found no significant difference between the models (P = 0.55), indicating no differences in resource knowledge for women by age and race. However, in the model containing interaction terms for age and race with gender and estimating their association with opinion of local services, age and race had a modifying effect on the association between being female and opinion of services (See Table 3 for results from the ordered logit model containing interaction terms for race and age with gender). Specifically, African-American women over 35 (n = 72) reported services to be less helpful than other respondents (P = 0.012) indicating that race and age play a significant role in women's perception of local HIV/AIDS services. The likelihood ratio test detected a significant contribution of the interaction terms to the model (P = 0.011).

When an African-American woman over 35 with the characteristics described in the example above is compared with a non-African-American woman with the same characteristics, there are substantial differences in how they view local services. The older African-American woman has a predicted probability of 0.23 that she will find services to be very helpful compared with 0.62 for the older non-African-American woman (see Table 4 for predicted probabilities). Furthermore, the older African-American woman has a predicted probability of 0.26 for not reporting services as helpful, in contrast to 0.06 for the older non-African-American woman. There are also differences in the predicted probability of not reporting services to be helpful between older African-American women, 0.26, and other respondents of different ages and races such as older African-American men, 0.14, and younger non-African-American men, 0.19.

Discussion

Study results revealed that the women in this sample of HIV-positive persons living in the Southeastern United States were predominantly residing in rural areas and had less education and greater barriers to ancillary services than their male counterparts. Multivariate analysis indicated that the women were less likely to know how to access information about HIV/AIDS resources, and rated services in their community as less helpful. In addition, African-American women over 35 were significantly more likely than other respondents to rate local services for people living with HIV/AIDS as being less than helpful. These findings raise the concern that barriers to services will ultimately result in poorer outcomes for women. Previous research has identified women to be at greater risk than men for delay in medical care or for not utilizing antiretroviral medications. (4,19-21,24-26) Greater difficulty accessing ancillary services may contribute to these negative outcomes. If women living with HIV/AIDS do not know how to obtain services or are unwilling to utilize services, they may be unable to resolve the psychosocial problems that can interfere with access and adherence to medical care.

HIV/AIDS services are often not targeted toward the specific needs and concerns of women; (4,10,18) thus, women may often be unaware of HIV/AIDS services or may feel uncomfortable accessing services that are not structured for them or are not relevant to their concerns. Because the majority of persons who have HIV/AIDS in rural areas have been men, the needs of women who have HIV/AIDS may have been viewed as less critical, resulting in a lack of services that account for the often complex relationships and responsibilities unique to women. (43)

In addition to a lack of specifically tailored resources, the stigma often associated with HIV disease may contribute to women's perception of services or willingness to seek out resources. Society often tolerates and even promotes male sexuality, while discouraging similar actions in women, creating a climate where HIV-positive women may perceive an additional social stigma associated with their disease. (44) Many women also fear that disclosing their HIV status will cause them to lose custody of their children. (44) The consequences of these concerns regarding stigma and disclosure may be that women choose not to seek resource information or may distrust local services. For those who do seek care, a negative experience such as a breach in confidentiality may result in reluctance to obtain further care.

The finding that African-American women over 35 rated services as less helpful than other respondents may be influenced by a greater distrust of the medical community and government-sponsored agencies because of the social history of discrimination that these respondents have faced. (45) The injustices of the past, including the Tuskegee syphilis experiment, (46) as well as the belief among some African-Americans that HIV is a weapon of government genocide, and the growing body of empirical evidence that African-Americans do not receive the same quality of care as their white counter-parts, (47-49) may lead many individuals to expect discrimination or dishonesty from medical researchers and professionals. (4,45,50,51) Older African-American women may be particularly wary of medical and social services due to greater exposure to discrimination. These women were born during or before 1960 and thus may have been more personally affected by life experiences before and during the civil rights movement. Additional research is needed to determine the specific reasons why older African-American women may hold a lower opinion of services, so that attempts can be made to address this disparity in perception.

Although the exact etiology of the differences between women and men in resource knowledge and regard for services could not be determined in this study, there are some logical steps that can be taken to address this issue. For example, the methods of disseminating information about HIV/AIDS services need to be examined to make sure they effectively advertise and conduct outreach to the female population in need. Furthermore, since case manager contact was associated with greater knowledge of how to access HIV/AIDS services and more favorable perception of services, efforts are needed to ensure that women in need are linked with HIV case management services.

In addition to having knowledge about support services, women must also perceive that the available services are nonthreatening and relevant to them. A local needs assessment can be beneficial in determining the needs of women in a specific community and in identifying how best to address these concerns. To minimize issues of stigma, confidentiality breach, and discrimination, support service providers must be trained to be sensitive to these issues. This is particularly important for services that are not tailored specifically for people with HIV/AIDS, as staff at these agencies may have less experience with HIV and related issues.

The results of this study must be taken in the context of its limitations. Only those patients who attended the infectious diseases clinics and could be reached by phone for survey were able to participate in this study; thus, the sample may be biased toward those with more stable lives and greater participation in medical treatment. Furthermore, the women in this sample were drawn from several states in the Southeast; therefore, the results may not be as relevant to other areas of the country where cultural and social climates differ. Finally, study results indicate that women had less knowledge of and lower opinions of services; however, it is not feasible to assess whether this resulted in less use of services.

Conclusion

This study reveals important information about barriers to support services that HIV-positive older African-American women, as well as HIV-positive women in general, are experiencing in the rural Southeast. Since unmet psychosocial needs can result in failure to adhere to medical appointments and treatment, addressing these barriers may be crucial to improving women's medical outcomes. Engaging women in case management and ancillary services designed to meet their needs has the potential to contribute significantly to the duration and quality of their lives.

It is not enough to have a good mind; the main thing is to use it well.
--Rene Descartes

Table 1. Characteristics of the survey respondents

                                      Women      Men       Full sample
Characteristic                      (n = 188)   (n = 328)   (n = 526)

Demographic
  Average age (yr)                   36 (a)     39         37
  African-American                    0.77 (a)   0.58       0.65
  Less than high school education     0.34 (a)   0.23       0.27
  Disabled                            0.50 (a)   0.63       0.58
  Unemployed                          0.08       0.09       0.09
  Child under 18 in the home          0.69 (a)   0.14       0.34
  Multiple housing situations in      0.43       0.51       0.48
    the last year
  Income under the poverty level      0.52       0.52       0.52
  Heterosexual                        0.87 (a)   0.37       0.55
  Distance to Infectious Diseases    47         43         42
    clinic (miles)
  Live in a rural area                0.82 (b)   0.72       0.74
  Have no medical insurance           0.15       0.17       0.16
Behavioral
  History of illicit substance use    0.53 (a)   0.73       0.66
  Illicit substance use in last       0.18 (a)   0.34       0.28
    six months
  Trouble with law in past year       0.053 (b)  0.12       0.10
Medical and Support Services
  Knowledge of how to access          0.62 (a)   0.80       0.64
    resource information
  Do not rate local services to be    0.22 (a)   0.13       0.20
    helpful
  Have an HIV case manager            0.64       0.69       0.63
  Have a regular medical care         0.90       0.93       0.90
    provider
  Experienced difficulty obtaining    0.20       0.18       0.19
    transportation
Health Status
  Time since diagnosis (years)        4.29 (b)   5.30       4.80
  Asymptomatic                        0.25       0.20       0.23
  Minor symptoms                      0.33 (a)   0.18       0.37
  AIDS                                0.34 (a)   0.58       0.34

(a) Significantly different from men at P < 0.01 on [chi square] or t
tests.
(b) Significantly different from men at P < 0.05 on [chi square] or t
tests.

Table 2. Logistic regression analysis of respondents' knowledge of how
to access HIV/AIDS resources

Independent Variable        [beta]        OR

Female                      -0.87 (a)   0.42
African-American            -0.29       0.75
Over 35 years of age        -0.0066     0.99
No high school education     0.059      1.06
Some college education       0.87 (a)   2.38
College graduate (b)         0.58       1.79
Time since diagnosis         0.074 (c)  1.08
Child under 18 in the home   0.035      1.04
Distance to Infectious      -0.001      1.00
  Diseases clinic (miles)
HIV case manager             0.74 (a)   2.10
Asymptomatic                -0.40       0.67
Minor symptoms (d)          -0.035      0.97
Mental health--SF 36 score   0.0083     1.01

(a) P < 0.01.
(b) Omitted category is high school graduate.
(c) P < 0.05.
(d) Omitted category is AIDS.

Table 3. Ordered logit analysis of respondents' perceptions of local
services for persons with HIV/AIDS

                       Model with interaction  Model without interaction
                       terms for race and age  terms for race and age
Independent Variable     with gender ([beta])     with gender ([beta])

Female                        0.72                     0.48 (a)
African-American             -0.40                    -0.13
Over 35 years of age         -0.0035                  -0.14
African-American (a)         -0.026                   -
  age over 35
African-American (a)         -0.0080                  -
  female
Female (c) age over          -2.01 (a)                -
  35
African-American (a)          2.15 (a)                -
  female (a) age over
  35
No high school               -0.12                    -0.14
  education
Some college                  0.19                     0.13
  education
College graduate (b)          0.39                     0.44
Distance to
  Infectious Diseases         0.0069 (c)              -0.0061 (c)
  clinic (miles)
Child under 18 in the         0.15                     0.17
  home
HIV case manager             -0.55 (c)                -0.50 (a)
Asymptomatic                  0.036                    0.097
Minor symptoms (d)           -0.29                    -0.22
Mental health-SF 36          -0.013 (a)               -0.012 (c)
  score

(a) P < 0.05.
(b) Omitted category is high school graduate.
(c) P < 0.01.
(d) Omitted category is AIDS.

Table 4. Predicted probabilities for perception of local HIV/AIDS
services (a)

                                        Neither
                       Very             helpful nor    Not      Very
                      helpful  Helpful  unhelpful    helpful  unhelpful

African-American       0.23      0.51      0.09        0.08     0.09
  woman over 35
Non-African-American   0.62      0.32      0.02        0.02     0.02
  woman over 35
African-American man   0.41      0.46      0.05        0.04     0.04
  over 35
African-American       0.25      0.51      0.09        0.07     0.08
  woman under 35
Non-African-American   0.31      0.50      0.07        0.06     0.06
  male under 35

(a) Predicted probabilities are for participants with the following
characteristics: high school graduate, CDC definition of AIDS diagnosis,
no children under 18 in the home, no case manager, and median mental
health score and distance to medical care.

Acknowledgments

This policy analysis was supported by the Health Resources and Services Administration (HRSA) Grant # H76HA00614 of the US Department of Health and Human Services. The conclusions reached in this report are those of the authors and may not reflect the views of the funding agency. We thank Traci Dreher, Jeanette Johnson-Licon, Melinda Steele, and Jai Uberoi for collecting patient consents, and Dr. James Raper for his assistance with patient recruitment and institutional review board approval. We also thank the clinicians and staff at each of the participating infectious diseases clinics for their assistance in obtaining institutional review board approval and enrolling patients, and for their general support of this research.

Accepted November 16, 2003.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9704-0342

References

1. Cunningham WE, Anderson RM, Katz MH, et al. The impact of competing subsistence needs and barriers on access to medical care for persons with Human Immunodeficiency Virus receiving care in the United States. Med Care 1999;37:1270-1281.

2. Marx R, Katz M, Park MS, et al. Meeting the service needs of HIV-infected persons: Is the Ryan White CARE Act succeeding? J Acquir Immune Defic Syndr 1997;14:44-55.

3. Katz MH, Cunningham WE, Fleishman JA, et al. Effect of case management on unmet needs and utilization of medical care and medications among HIV-infected persons. Ann Intern Med 2001;135(8 Pt 1):557-565.

4. Raevis V, Siegel K, Gorey E. Factors associated with HIV-infected women's delay in seeking medical care. AIDS Care 1998;10:549-562.

5. Conover C, Whetten K. The impact of ancillary services on primary care use and outcomes for HIV/AIDS patients with public insurance coverage. AIDS Care 2002;14(suppl 1):S59-71.

6. McKinney M. Service needs and networks of rural women with HIV/AIDS. AIDS Patient Care 1998;12:471-480.

7. Crystal S, Jackson MM. Psychosocial adaptation and economic circumstances of persons with AIDS and ARC. Fam Community Health 1989;12:77-88.

8. Piette JD, Fleishman JA, Stein MD, et al. Perceived needs and unmet needs for formal services among people with HIV disease. J Community Health 1993;18:11-23.

9. Butz AM, Hutton N, Joyner M, et al. HIV-infected women and infants: social and health factors impeding utilization of health care. J Nurse Midwifery 1993;38:103-109.

10. Weissman G, Melchior L, Huba G, et al. Women living with substance abuse and HIV disease: medical care access issues. J Am Med Womens Assoc 1995;50:115-120.

11. Bartlett J: Addressing the challenges of adherence. J Acquir Immune Defic Syndr 2002;29(suppl 1):S2-S10.

12. Singh N, Squier C. Determinants of compliance with antiretroviral therapy in patients with Human Immunodeficiency Virus: Prospective assessment with implications for enhancing compliance. AIDS Care 1996;8:261-269.

13. Angelino AF, Treisman GJ. Management of psychiatric disorders in patients infected with Human Immunodeficiency Virus. Clin Infect Dis 2001;33:847-856.

14. Shelton D, Marconi K, Pounds MB, et al. Medical adherence among prenatal, HIV seropositive African-American women: Family issues. Fam Syst Med 1993;11:343-356.

15. Sankar A, Luborksy M, Schuman P, et al. Adherence discourse among African-American women taking HAART. AIDS Care 2002;14:203-218.

16. Sherer R, Stieglitz K, Narra J et al. HIV multidisciplinary teams work: support services improve access to and retention in HIV primary care. AIDS Care 14(suppl 1):S31-44, 2002.

17. Magnus M, Schmidt N, Kirkhart K, et al. Association between ancillary services and clinical and behavioral outcomes among HIV-infected women. AIDS Patient Care STDs 2001;15:137-145.

18. Meredith K, Delaney J. A survey of women with HIV about their expectations for care. AIDS Care 1997;9:513-522.

19. Siegel K, Karus D. Testing and treatment behaviour of HIV-infected women: White, African-American, Puerto Rican comparisons. AIDS Care 1997;9:297-309.

20. Cashetta MB. Clinical manifestations of HIV infection in women. Treatment Issues 1990;5:3.

21. Brettle R, Leen C. The natural history of HIV and AIDS in women. AIDS 1991;5:1283-1292.

22. Ickovics J, Rodin J. Women and AIDS in the United States. Health Psychol 1992;11:1-16.

23. Frank KO, Blundo R, Brabant S. HIV-infected women in Louisiana. AIDS Patient Care 1995;9:111-120.

24. Shapiro MF, Morton SC, McCaffrey DF, et al. Variations in the care of HIV-Infected adults in the United States. JAMA 1999;281:2305-2315.

25. Smith SR, Kirking DM. Access and use of medications in HIV disease. Health Serv Res 1999;34:123-144.

26. Hellinger F. The use of health services by women with HIV Infection. Health Serv Res 1993;28:543-562.

27. Anderson KH, Mitchell JM. Differential access in the receipt of antiretroviral drugs for the treatment of AIDS and its implications for survival. Arch Intern Med 2000;160:3114-3120.

28. Ashman JJ, Conviser R, Pounds MB. Associations between HIV-positive individuals' receipt of ancillary services and medical care receipt and retention. AIDS Care 2002;14(suppl 1):S109-S118.

29. Seals B, Sowell RL, Demi AS. Falling through the cracks: Social service concerns of women infected with HIV. Qual Health Res 1995;5:496-515.

30. Kaiser Family Foundation: HIV/AIDS and other sexually transmitted diseases (STDs) in the southern region of the country: Epidemiological overview 2002; Available at http://www.kff.org/hivaids/upload/14362_1.pdf. Accessed February 23, 2004.

31. Hader SL, Smith DK, Moore JS, et al. HIV infection in women in the United States: Status at the Millenium. JAMA 2001;285:1186-1192.

32. McKinney MM. Variations in rural AIDS epidemiology and service delivery models in the United States. J Rural Health 2002;18:455-466.

33. HIV/AIDS Surveillance Supplemental Report. Atlanta, GA: Centers for Disease Control and Prevention;2000;6:(No. 2):1-16.

34. Whetten K, Nguyen T, Heald A. Characteristics of individuals infected with the Human Immunodeficiency Virus and provider interaction in the predominantly rural Southeast. Southern Med J 2001;94:212-222.

35. Bozzette SA, Hays RD, Berry SH, et al. Derivation and properties of a brief health status assessment for use in HIV disease. J Acquir Immune Defic Syndr Hum Retrovirol 1995;8:253-265.

36. Wu AW, Rubin HR, Mathews WC, et al. A health status questionnaire using 30 items from the medical outcomes study; preliminary validation in person with early HIV infection. Med Care 1991;29:786-798.

37. Ware JE, Snow KK, Kosinski M, et al. SF-36 Health Survey: Manual and interpretation guide. Boston, Health Institute, New England Medical Center, 1993.

38. Centers for Disease Control and Prevention: 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. Morb Mortal Wkly Rep 1993;41(RR-17):1-17.

39. Borooah VK. Logit and Probit: Ordered and multinomial models. Thousand Oaks, Ca, Sage Publications, 2001.

40. Statacorp. Stata reference manual release 6: Volume 2. College Station, TX, Stata Press, 1999.

41. Gujarati DN. Basic Econometrics. New York, McGraw-Hill, 1995.

42. Federal Poverty Guidelines. Federal Register 1997;62:10856-10859.

43. Castaneda D. HIV/AIDS-related services for women and the rural community context. AIDS Care 2000;12:549-565.

44. Bunting S. Sources of stigma associated with women with HIV. Adv Nurs Sci 1996;19:64-73.

45. Shavers V, Lynch C, Burmeister L. Factors that influence African-Americans' willingness to participate in medical research studies. Cancer 2001;91(suppl 1):233-236.

46. Francis C. The medical ethos and social responsibility in medicine. J Natl Med Assoc 2001;93:157-168.

47. Blendon RJ, Scheck AC, Donelan K, et al. How Whites and African Americans view their health and social problems: Different experiences, different expectations. JAMA 1995;273:341-346.

48. Fongwa M. Overview of themes identified from African American discourse on quality of care. J Nurs Care Qual 2002;16:17-38.

49. Shi L, Starfield B. The effect of primary care physician supply and income inequality on mortality among Blacks and Whites in US metropolitan areas. Am J Public Health 2001;91:1246-1250.

50. Thomas SB, Quinn SC. The Tuskegee Syphilis Study, 1932 to 1972: Implications for HIV Education and AIDS risk education programs in the Black community. Am J Public Health 1991;81:1498-1505.

51. Corbie-Smith G, Thomas S, Williams M, et al. Attitudes and beliefs of African Americans toward participation in medical research. J Gen Intern Med 1999;14:537-546.

RELATED ARTICLE: Key Points

* Women with HIV/AIDS have a potentially higher risk for experiencing barriers to ancillary services because many HIV-related services have not been traditionally targeted and structured to address their specific needs.

* If women living with HIV/AIDS do not know how to obtain services or are unwilling to utilize services, they may be unable to resolve the psychosocial problems that can interfere with access and adherence to medical care.

* Study data indicate that women are less likely to know how to access information about HIV/AIDS resources, and that they rated services in their community as less helpful.

Kathryn Whetten, PHD, MPH, Susan Reif, PHD, MSW, Kristin Lowe, BS, Lois Eldred, DRPH, MPH

From the Health Inequalities Program, Duke University Center for Health Policy, Law and Management, Durham, NC, and the Health Resources and Services Administration, Demonstration Project Development and Evaluation Branch, Rockville, MD.

The authors have no commercial or proprietary interest in any institution, service, or product mentioned in the article. This policy analysis was supported by the Health Resources and Services Administration (HRSA) Grant # H76HA00614 of the US Department of Health and Human Services, Rockville, MD. Institutional review board approval was received from each site involved in the study, and a federal certificate of confidentiality was also obtained for the study.

Reprint requests to Kristin Lowe, Duke University Center for Health Policy, Law and Management, Box 90253, Durham, NC 27708. Email: [e-mail]