Mikhail, Isis S MD, MPH, DRPH*†; DiClemente, Ralph PHD*; Person, Sharina PHD‡; Davies, Susan PHD§; Elliott, Elizabeth MSN, PHD†; Wingood, Gina SCD, MPH*; Jolly, Pauline E PHD, MPH†
In 2002, it was estimated that women accounted for 19.2 million (∼50%) of the 38.6 million adults who were living with HIV infection or the AIDS. Women also accounted for 1.2 million of the 3.1 million deaths attributable to AIDS worldwide.1 In the United States, the fastest growing population at risk for HIV/AIDS is minority women, particularly African-American and Hispanic women.2 During the past decade, the proportion of AIDS cases in adult and adolescent females aged 13 to 49 years has tripled, increasing from 7% to 25%. Once diagnosed, a critical component in prolonging the life of HIV/AIDS patients is appropriate treatment. The typical treatment, highly active antiretroviral therapy (HAART), involves combination antiretroviral therapies that include protease and reverse transcriptase inhibitors.3 In addition to conventional medical treatment, there are a considerable number of individuals who use complementary and alternative medicines (CAM) such as herbs, vitamins, and dietary supplements.4,5
There is wide use of CAM among people with various chronic illnesses and for preventive purposes in the United States,4,6 and CAM has gained increasing interest from patients, clinicians, and researchers.7–10 CAM use among people with HIV has been estimated to range from 29%11 to 76%.12 Many HIV patients who use CAM believe that these medicines boost their immunity and help them to cope with the side effects of their medications.12,13
Review of published studies on CAM use among HIV-infected persons suggests that CAM use is disproportionately high among whites, males, men who have sex with men (MSM), persons with a high school (HS) education or higher, and those with higher incomes.11,14–19 Few surveys have investigated CAM use among women, and we could find no report on CAM use among African-American women in particular. Meneilly et al20 reported that HIV-positive white women were more likely to use a wide variety of herbal remedies compared with a group of men with similar demographic characteristics. A study by Smith et al21 found that African Americans were less likely to use nonprescription drugs and herbs compared with non-Hispanic whites. The sample was predominantly male, however, and the data were not analyzed according to gender. We determined the extent of CAM use by a cohort of predominantly African-American HIV-positive women and investigated associations between the use of CAM, both as an aggregate and individually, and indicators of HIV clinical disease status such as CD4+ T-cell count and viral load.22,23 We assessed additional indicators of clinical disease status such as Centers for Disease Control and Prevention (CDC) categorization, Karnofsky performance score, and number of opportunistic infections.
Data were collected as part of the “Women Involved in Life Learning from Other Women” (WiLLOW) Study. The WiLLOW Study is a National Institutes of Health (NIH)–funded longitudinal behavior intervention conducted in Alabama (The University of Alabama at Birmingham [UAB] School of Public Health) and Georgia (Emory University, Rollins School of Public Health). It was designed to investigate the impact of an educational intervention program on reducing stress, depression, and high-risk sexual behavior among HIV-positive women. The major goal was to reduce the risk of transmission of HIV and other sexually transmitted diseases (STDs). As a part of the WiLLOW protocol, data were collected at enrollment regarding the use of CAM as part of the baseline assessment. Baseline data collected included socioeconomic and demographic variables (ie, age, race, marital status, education level, household income, health insurance status, number of children), disease data (disease duration and HIV medications), and types of CAM used. The clinical HIV disease markers included CD4+ T-cell count, viral load, CDC categories, Karnofsky score, and number of infections. These data were abstracted from the patients’ medical records.
Study participants were recruited from sites in Alabama and Georgia. The Alabama participants were recruited from clinics in 3 cities: Birmingham, Montgomery, and Anniston. The Birmingham participants were women seeking health care at the 1917 AIDS Clinic at the UAB, St. George’s Clinic at the Cooper Green Hospital, and the Family Clinic at Children’s Hospital. The Montgomery participants included women who sought treatment at the Montgomery AIDS Outreach Clinic (MAO) and the Family Clinic at the UAB Montgomery Clinic. Some participants were also recruited from women visiting the AIDS Services Center in Anniston. All participants from Georgia were recruited from 2 locations in Atlanta, namely, the Ponce DeLeon Clinic affiliated with Grady Hospital and the Fulton County Health Department HIV Clinic.
The study population for this work involved 391 HIV-positive women ranging in age from 18 to 50 years old. They were recruited and enrolled in the study between September 1997 and December 2000. The study protocol was approved by the Institutional Review Boards at both the UAB and Emory University in Atlanta.
Complementary and Alternative Medicine Data Assessment
Assessment of the use of CAM was performed using data from the CAM section of the WiLLOW Study baseline questionnaire. The CAM questions asked whether or not the women ever used (1) herbal and natural medicines, (2) vitamins, (3) dietary and nutritional regimens, (4) religious healing, (5) body work such as massage and acupuncture, and (6) psychic healing. In this article, CAM refers to both complementary and alternative medicines and therapies.
Measures of Indicators of HIV Disease Status
The major clinical outcome measures used to assess HIV disease status in study participants were the CD4+ T-cell count, viral load counts, CDC categories, Karnofsky scores, and number of infections experienced at baseline or up to 3 months before enrollment into the study.
CD4+ T-cell data for this work were abstracted from the medical records and used as a measure of clinical disease among study participants. The CD4+ T-cells counts were categorized as follows: category 1: ≥500 cells/mm3, category 2: 200 to 499 cells/mm3, and category 3: <200 cells/mm3.
HIV viral load is an important predictor of HIV progression.24 The viral load data for participants in this study were abstracted from their clinic medical records. These were divided into 3 categories: <1000 copies/mL, 1000 to 9999 copies/mL, and ≥10,000 copies/mL.25 An “undetectable” level (<1000 copies/mL) of the virus in the blood may occur in some patients, particularly those on protease inhibitors or HAART. The Roche HIV test (Roche Diagnostics, Branchburg, NJ) was conducted by all the laboratories used by all the clinics in the study, except for an “in-house” polymerase chain reaction (PCR) similar to the Roche test that was used by a single laboratory.
The revised CDC classification system that was implemented in January 1993 and replaced the previous 1986 classification was used in the study. This newer classification is closer to a staging system but is still not defined as such.26 It categorizes adolescents and adults as asymptomatic (A), symptomatic with conditions attributable to HIV (B), and true AIDS (C). CDC categories were obtained by abstraction from patient medical records.
The Karnofsky Performance Scale (KPS) is a tool used to describe how well a patient can function in his/her daily activities.27,28 It ranges from 100%, which is the best score and describes a normal healthy adult, to 0%, which indicates death.27 In this study, the KPS score was determined from the clinical records. Some physicians do not include the KPS score in their patient assessment. Therefore, the analysis of this indicator was conducted in a subset of patients for whom these data were available.
The number of HIV-related opportunistic infections has been reported to be a more accurate predictor of HIV disease than CD4+ T-cell count.29 For this reason, data on all infections and other clinical conditions experienced by study participants were retrieved from the clinical database and from patients’ medical records. Infections were classified according to the organ system affected (www.aegis.com/topics/oi).
Descriptive statistical analyses were performed to summarize the data, including types of CAM used, clinical disease variables (ie, CD4 count, viral load, CDC categories, Karnofsky score, number of infections), sociodemographic variables (age, race, marital status, income, health insurance, and number of children), and disease-related variables (duration since disease onset and antiretroviral therapies).
To determine the relation between CAM use and the aforementioned clinical disease as well as sociodemographic and disease-related variables, we conducted a series of bivariate analyses using χ2 test statistics. The prevalence ratio (PR) was used to examine the strength of the bivariate relations between a predictor (CAM use) and the correlates at a 95% confidence interval (CI), and P was set at 0.05. To further examine the relation between CAM use and previously mentioned variables, we conducted a stepwise multivariable logistic regression with the significance level to enter and stay set at 0.10. Analyses were conducted using SAS version 8.0.
Sociodemographic Characteristics of the Study Population
Three hundred ninety-one women aged 18 to 50 years old and living with HIV/AIDS participated in this study. Eighty-four percent of these women are African American (n = 327), 15% are white (n = 58), and 1% are of other races (Hispanic and mixed races; n = 6) (Table 1). Most study participants (both users and nonusers of CAM) were single (86%). Approximately 70% of women had a household income of less than $10,000 per year, had completed a HS education or less, had health insurance coverage (predominantly Medicaid), and had 2 or fewer children (see Table 1).
Use of CAM was prevalent among our study population. About 60% of women reported using at least 1 CAM (Table 2). About 16% reported using herbs, 22% used dietary supplements, 27% practiced religious healing, 10% used bodywork (eg, message, yoga), and 1% practiced some type of psychic healing. Vitamins were the most commonly used CAM and were used by 36% of women in the study (see Table 2).
Sociodemographic Characteristics of Complementary and Alternative Medicine Users
Women reporting CAM use tended to be older (>35 years, CI: 0.98–1.52; P = 0.059), were better educated (>HS, CI: 1.12–1.55; P = 0.002), and had no insurance coverage (CI: 0.76–0.99; P = 0.04) compared with those who did not report CAM use (Table 3). On assessing disease duration, women who were living with HIV for a longer period (≥4 years, CI: 1.10–1.68) used significantly more CAM than those with shorter disease duration (<4 years; P = 0.002).
Complementary and Alternative Medicine Use and Clinical Disease Indicators
When the association between CAM use and clinical disease indicators was investigated, a higher number of infections (3 or more) were found to be associated with CAM use (CI: 1.02–1.49; P = 0.028). No significant association was found with other clinical indicators such as CD4+ T cells, viral load, CDC categories, and Karnofsky score, however (Table 4).
Sociodemographic Characteristics and Clinical Disease Indicators of Vitamin Users
Vitamins were identified as the most commonly used CAM among this population (∼36%). Similar to CAM users, vitamin users were women who had significantly higher education levels (CI: 1.09–2.09; P = 0.012), higher income (CI: 1.18–2.16; P = 0.002), and longer disease duration (CI: 1.00–1.46; P = 0.044) (Table 5). A trend toward more vitamin use was noted among white women compared with African-American women (CI: 0.83–1.01; P = 0.09). When examining the association between vitamin use and clinical disease indicators, women who use vitamins had lower viral load levels (P = 0.011) and a higher number of infections (P = 0.003) than those who did not use vitamins.
Multivariate Logistic Regression of Complementary and Alternative Medicine Use
We conducted multivariate logistic regression analysis to control for any potential confounders. Odds ratios (OR) at 95% CI for correlates of CAM use are presented in Table 6. None of the significant associations with CAM users were eliminated during the regression analysis. Women who used CAM were found to be more educated (OR = 2.4; P = 0.0008) than those who did not use CAM. Regarding health insurance coverage, CAM users were about twice as unlikely to have insurance coverage than non-CAM users (OR = 0.49; P = 0.0055). Women who chose to use CAM were found to be more than twice as likely to have longer disease duration (≥4 years) than non-CAM users (OR = 2.21; P = 0.0006). Regarding clinical disease indicators, women who used CAM were approximately twice as likely to have a higher number of infections than non-CAM users (OR = 0.58; P = 0.0173).
Multivariate Logistic Regression of Vitamin Use
ORs at 95% CI for correlates of vitamin use are presented in Table 7. Vitamin users were approximately 2.5 times more likely to have a higher education level (>HS) than those who did not use vitamins (OR = 2.34; P = 0.0055). White women were almost twice as likely to use vitamins than African-American women. HIV-positive women with longer disease duration (≥4 years) were almost twice as likely to use vitamin supplements than women positive for HIV for <4 years (OR = 1.87; P = 0.027).
Most studies conducted on CAM to date have not focused on CAM use among women living with HIV/AIDS, particularly African-American women. We have been successful in including a substantial number of African-American women (n = 327 [84%]) in our study in addition to 15% white women (n = 58). Because Hispanic women and women of mixed racial background comprised only 1% (n = 6) of our study participants, this group was excluded from the analyses. The sociodemographic characteristics of our study population are an important consideration as we discuss our results. This is particularly so because the socioeconomic status of the women (ie, education, income, availability of insurance) is likely to influence choices regarding the use of CAM.4 Most WiLLOW Study participants were single (86%). Most of them (70%) had a household income of less than $10,000 per year, had a HS education or less, had health insurance coverage (primarily Medicaid), and had 2 children or fewer.
Fifty-nine percent of study participants used at least 1 CAM (n = 229). CAM use among HIV/AIDS patients had been found to range between 27% and 100%.12,30–34 These studies included either men only or men and women who were predominantly white.4 An important distinction regarding our study is that it focused solely on women, most of whom are African American. Our results are consistent with those of some of these studies. The lower frequency of CAM use in our study may be a result of the higher number of African Americans in the study. CAM use among African Americans has been shown to be lower than among whites.21 Another reason could be that a large number of women (∼72%) in our study population have a lower level of education and CAM use has been found to be more common among individuals with a college level education.7,35 Conversely, the frequency of CAM use among our population is higher than that in some other reported studies. This may be explained by the fact that our population consists solely of women. Use of CAM was found to be higher among women than among men.4,36
In this study, women reporting CAM use tended to be older (≥35 years), were more highly educated (>HS), and had no insurance coverage (P = 0.04) compared with those who did not report CAM use. These results are consistent with those of a study by Astin,7 who found that older individuals (aged 35–49 years) use more CAM than other age groups.
When we examined the association between CAM use and time since disease onset, we found that women with longer disease duration (≥4 years) use more CAM than women with shorter duration of disease (P = 0.0022). This may be a result of the fact that the longer patients live with HIV/AIDS, the greater is the number of health problems that arise; this may lead patients to seek alternatives such as CAM. Such an association needs to be further investigated, however. The finding that a higher proportion of women with a higher number of infections used CAM compared with women with fewer infections could also be a result of the fact that women with more infections seek CAM to resolve the problems and cope with HIV/AIDS.
The types of CAM that have been shown to have an effect on CD4 count and viral load levels are predominantly vitamins, dietary supplements, and herbs.22,37,38 Contrary to such studies, no significant association was found between CAM use and CD4+ T cells or viral load in this study. Similarly, other studies have not shown any significant change in viral load or immune measures using dietary supplements in HIV-infected individuals.39 The negative finding in our study could be explained by the fact that the combined CAM consisted of different types of therapies, including psychic healing and bodywork. Psychic healing has not been shown to have an effect on CD4 count and viral load levels and may have obscured our results. A few studies that have been conducted on bodywork have shown beneficial results, however. Scafidi and Field40 conducted a study on the effect of massage therapy on 28 HIV-positive babies and found higher daily weight gain and Brazelton performance scores among the therapy group. Another study by Standish et al41 has shown massage therapy to be the most commonly used CAM among a large population of HIV-positive individuals from the Bastyr University AIDS Research Center of Alternative Medicine Care Outcomes in AIDS (AMCOA).
We noted that when vitamin use was examined independently of other types of CAM in our study population, and despite the smaller sample size compared with combined CAM, a significant association was found between vitamin use and decreased viral load (P = 0.01). Of all the different types of CAM, vitamin use has been shown in many studies to be the CAM of choice by HIV-infected patients.41,42 Vitamin intake represents a self-promoting behavior that contributes to general good health and well-being. Studies have shown that individuals who practice self-promoting behaviors take more vitamins in addition to eating a healthy diet, getting adequate amounts of sleep, and engaging in regular exercise.43 Vitamins were the CAM most commonly chosen by our participants: 36% of participants reported using vitamins. Similar findings were reported by Nokes et al42; they surveyed 145 HIV-positive patients and found that among 55 different CAMs used, vitamins were most commonly used. Likewise, Fairfield et al12 conducted a survey of 289 HIV-positive gay men and found that vitamins and dietary supplements were most commonly used.
Vitamin use has also been shown to be associated with socioeconomic factors. Smith et al21 have shown that African Americans were less likely to use vitamins than whites but that HIV patients from both racial groups with a college education were more likely to use vitamins. Similarly, in our study population, vitamin users were women who had significantly higher education (P = 0.01) and income levels (P = 0.004).
Despite the small sample size of white women, a tendency toward greater use of vitamins was found for white women compared with African-American women (P = 0.09). These results are also consistent with the findings of other studies discussed earlier.4,21
One limitation of this study is that it is a cross-sectional design, which impedes the determination of causality between CAM use and clinical disease status. Follow-up longitudinal studies and randomized clinical trials are needed to confirm these results.44 Despite this limitation, this project provides significant information toward our understanding of patterns of CAM use among women, especially African-American women living with HIV/AIDS. Trends of CAM use among women in general and African-American women in particular had not been sufficiently investigated previously.
These data have important implications for improvement in the care of HIV-positive women with involvement of their health care providers. Physicians and health care professionals need to develop greater awareness of the prevalence of CAM use among their patients and to learn more about these therapies, their effectiveness, and potential side effects. Patients should be encouraged to inform their physicians of their use of CAM so that the physicians can monitor and regulate CAM use for the ultimate well-being of the patients.
The authors extend special thanks to study interviewers and health educators and all WiLLOW study staff who went beyond the call of duty to recruit and interview this special population and accomplish this distinctive work. Special gratitude goes to all the women living with HIV/AIDS for their participation in this study and for sharing their unique and personal experiences.
1. UNAIDS. Report on the global HIV/AIDS epidemic. UNAIDS. 2002.
2. Centers for Disease Control and Prevention (CDC) Report. HIV Prevention. Strategic Plan Through 2005. CDC. 2001.
3. Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir on persons with human immunodeficiency virus infection and CD4 T cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. N Engl J Med. 1997;337:725–733.
4. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA. 1998;280:1569–1575.
5. Kaptchuck TJ, Eisenberg DM. The persuasive appeal of alternative medicine. Ann Intern Med. 1998;129:1061–1065.
6. Eisenberg DM, Kessler RC, Foster C, et al. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N Engl J Med. 1993;328:246–252.
7. Astin JA. Why patients use alternative medicine: results of a national study. JAMA. 1998;279:1548–1553.
8. Jonas WB. Alternative medicine and the conventional practitioner. JAMA. 1998;280:1616–1618.
9. Roufogalis BD. The growth in complementary and herbal medicine. Pharm News. 1999;6:1–9.
10. Goldbeck-Wood S, Dorozynski A, Lie LG. Complementary medicine is booming worldwide. BMJ. 1996;313:131–133.
11. Greenblatt RM, Hollander H, McMaster JR, et al. Polypharmacy among patients attending an AIDS clinic: utilization of prescribed, unorthodox and investigational treatments. J Acquir Immune Defic Syndr. 1991;4:136–143.
12. Fairfield KM, Eisenberg DM, Davis RB, et al. Patterns of use, expenditures, and perceived efficacy of complementary and alternative therapies in HIV-infected patients. Arch Intern Med. 1998;158:2257–2264.
13. David E. HAART attack: metabolic disorders during long-term antiretroviral therapy. BETA. 1999;12:10–14.
14. Anderson W, O’Connor BB, MacGregor RR, et al. Patient use and assessment of conventional and alternative therapies for HIV infection and AIDS. AIDS. 1993;5:561–566.
15. Bates BR, Kissinger P, Bessiner RE. Complementary therapy use among HIV-infected patients. AIDS Patient Care STDS. 1996;10:32–36.
16. Carwein VL, Sabo CE. The use of alternative therapies in HIV infection: implications for patients care. AIDS Patient Care STDS. 1997;11:79–85.
17. Duggan J, Peterson WS, Schultz M, et al. Use of complementary and alternative therapies in HIV-infected patients. AIDS Patient Care STDS. 2001;15:159–167.
18. Paterson MS, Carwein VL, Sabo CE. The effect of income on the utilization of alternative therapies by HIV positive individuals [abstract Mo.B.1172]. Presented at the 11th International Conference on AIDS, Vancouver, July 1996.
19. Singh N, Squier C, Sivek C, et al. Determinants of nontraditional therapy use in patients with HIV-infection: a prospective study. Arch Intern Med. 1996;156:197–201.
20. Meneilly GP, Carr R, Brown L. Alternative therapy use in HIV-positive women [abstract MOB-301]. Presented at the 11th International Conference on AIDS, Vancouver, July 1996.
21. Smith SR, Boyd EL, Kirking DM. Nonprescription and alternative medication use by individuals with HIV disease. Ann Pharmacother. 1999;33:294–300.
22. Liang B, Chung S, Araghiniknam M, et al. Vitamins and immunomodulation in AIDS. Nutrition. 1996;12:1–7.
23. Tang AM, Smit E. Selected vitamins in HIV infection: a review. AIDS Patient Care. 1998;12:263–273.
24. Saag MS, Holodniy M, Kuritzkes DR. HIV viral load markers in clinical practice: recommendations of an International AIDS Society-US Expert Panel. Nat Med. 1996;2:625–629.
25. Contopoulos-Ioannidis DG, Ioannidis JPA. Maternal cell-free viremia in the natural history of perinatal HIV-1 transmission. A meta-analysis. J Acquir Immune Defic Syndr. 1998;18:126–135.
26. Centers for Disease Control and Prevention (CDC). 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR. 1992;41:RR-1.
27. Murri R, Scoppettuolo G, Damiano F, et al. Karnofsky performance status and assessment of global health status. J Acquir Immune Defic Syndr. 1996;13:294–295.
28. Zubrod CG, Schneiderman M, Frei E III. Appraisal of methods for the study of chemotherapy of cancer in man. J Chron Dis. 1960;11:7–33.
29. Hays R, Cunningham WE, Sherbourne CD. Health-related quality of life in patients with human immunodeficiency virus infection in the United States: results from the HIV cost and services utilization study. Am J Med. 2000;108:714–722.
30. Ernst E. Complementary AIDS therapies: the good, the bad, and the ugly. Int J STD AIDS. 1997;8:281–285.
31. De Visser R, Ezzy D, Bartos M. Alternative or complementary? Nonallopathic therapies for HIV/AIDS. Altern Ther Health Med. 2000;6:44–52.
32. Langewitz W, Ruttiman S, Laifer G, et al. The integration of alternative treatment modalities in HIV infection—the patient’s perspective. J Psychosom Res. 1994;38:687–693.
33. Sparber A, Wootton JC, Bauer L, et al. Use of complementary medicine by adult patients participating in HIV/AIDS clinical trials. J Altern Complement Med. 2000;6:415–422.
34. Dwyer JT, Salvato-Schille AM, Coulston A, et al. The use of unconventional remedies among HIV-positive men living in California. J Assoc Nurses AIDS Care. 1995;6:17–28.
35. Deleted in proof.
36. Kirksey KM, Goodroad BK, Kemppainen JK, et al. Complementary therapy use in persons with HIV/AIDS. J Holistic Nursing. 2002;20:264–278.
37. Tang AM, Graham NM, Kirby AJ, et al. Dietary micronutrient intake and risk of progression to AIDS in human immunodeficiency virus type 1 (HIV-1)-infected homosexual men. Am J Epidemiol. 1993;138:937–951.
38. Abrams B, Duncan D, Hertz-Picciotto I. A prospective study of dietary intake and acquired immune deficiency syndrome in HIV-seropositive homosexual men. J Acquir Immune Defic Syndr Hum Retrovirol. 1993;6:949–958.
39. Pichard C, Sudre P, Karsegard V, et al. A randomized double-blind controlled study of 6 months of oral nutritional supplementation with arginine and omega-3 fatty acids in HIV-infected patients. Swiss HIV Cohort Study. AIDS. 1998;12:53–63.
40. Scafidi F, Field T. Massage therapy improves behavior in neonates born to HIV-positive mothers. J Pediatr Psychol. 1996;21:889–897.
41. Standish LJ, Green KB, Bain S, et al. Alternative medicine use in HIV-positive men and women: demographics, utilization pattern, and health status. AIDS Care. 2001;13:197–208.
42. Nokes KM, Kendrew J, Longo M. Alternative/complementary therapies used by persons with HIV disease. J Assoc Nurses AIDS Care. 1995;6:19–24.
43. Gielen AC, McDonnell KA, Wu AW, et al. Quality of life among women living with HIV: the importance of violence, social support, and self-care behaviors. Soc Sci Med. 2000;52:315–322.
44. Power R, Gore-Felton C, Vosvick M, et al. HIV: effectiveness of complementary and alternative medicine. Prim Care. 2002;29:361–378.
© 2004 Lippincott Williams & Wilkins, Inc.