Secondary Logo

Journal Logo

Clinical Science

Determinants of Complementary Therapy Use in HIV-Infected Individuals Receiving Antiretroviral or Anti-opportunistic Agents

Ostrow, Michelle J.*; Cornelisse, Peter G.A.*; Heath,, Katherine V.*†; Craib, Kevin J.P.*; Schechter, Martin T.*†; O'Shaughnessy, Michael*‡; Montaner, Julio S.G.; Hogg, Robert S.*†

Author Information
Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology: June 1, 1997 - Volume 15 - Issue 2 - p 115-120
  • Free


In the absence of a definitive cure, many persons with human immunodeficiency virus (HIV) infection have chosen to supplement their treatment programs with complementary therapies—treatments that are not commonly provided by medical physicians but may be taken in conjunction with medical interventions. Extolled as a more natural and thereby gentler approach to treating symptoms, therapies such as vitamins, herbal extracts, and massage have provided individuals with a nontraditional approach to dealing with the debilitating effects of HIV(1).

Although the roots of certain therapeutic practices such as Chinese herbal medicine predate modern Western interventions(2), the investigation of the use of complementary therapies as a viable addition to the treatment of HIV-related infections arose in the mid-1980s(3). Since that time, these therapies have increasingly been used as adjuncts to medical treatment regimens for HIV-related illnesses(1,4,5). Three studies demonstrate that between 35% and 40% of HIV-positive individuals use some form of complementary therapy in conjunction with traditional Western medical regimens(1,5,6). Although the rationale supporting motivations for the use of nonprescribed therapies by HIV-infected persons is speculative, mounting pharmacologic costs and the need to assuage symptoms are assumed to have fostered some of the increasing interest in indigenous medicinal herbs and traditional healing methods(7,8).

To further the understanding of the patterns of complementary therapy use among HIV-positive individuals, we conducted the present study to identify sociodemographic and clinical characteristics of persons using complementary therapies in a province-wide HIV/AIDS drug treatment program.


The British Columbia Centre for Excellence in HIV/AIDS Drug Treatment Program provides antiretroviral, anti-opportunistic, and prophylactic agents free of charge to eligible HIV-infected individuals throughout the province. The criteria for eligibility in the treatment program is based on documented HIV infection and a CD4 cell count consistently below 500 cells/mm3. Consenting participants meeting these criteria are enrolled in the drug treatment program by their general practitioner on submission of a prescription for antiretroviral and approved antimicrobial medications(9).

Study participants were individuals who were enrolled in the program and completed an annual participant survey between September 1995 and June 1996. At initial enrollment and annually thereafter, participants enrolled in the drug program were asked to complete a self-administered participant survey. The survey elicited information about the participant's socioeconomic status, social support, clinical status, and current and past use of HIV-related medications and complementary therapies. A list of common complementary therapies were provided, from which survey participants could report use. An open-ended section was included for individuals to report additional therapies not included in the survey form. Participants were also asked to indicate, from a list of reasons, their motivations for taking these therapies. An open-ended answer choice was also provided. In addition, the survey contained the Centre for Epidemiologic Studies Depression Scale (CES-D), which is used to measure the respondent's current level of depressive symptomatology(10).

For the purposes of this study, complementary therapies were defined as treatments that are not commonly provided by medical physicians but may be used in conjunction with medical treatments. To simplify analysis, we divided the therapies into four major qualitatively distinctive groups: dietary supplements, herbal and other medicinal therapies, tactile therapies, and relaxation techniques. Dietary therapies included vitamins, minerals, and special diets. Herbal and other medicinal therapies included agents such as certain herbs and N-acetylcysteine(NAC). Dinitrochlorobenzene (DNCB), a topical agent, was deemed to be a medicinal treatment and was therefore included in this category. Tactile therapies included massage and acupuncture, whereas relaxation techniques included stress-reduction therapies and exercise. In the absence of a standardized strategy for grouping complementary therapies, any such divisions are intrinsically arbitrary to some extent.

Statistical analyses were carried out using both parametric and nonparametric methodologies to identify associations between participant characteristics and use of complementary therapies. Variables considered for inclusion were sociodemographic characteristics such as age, gender, ethnicity, place of residence, income, education, and risk group. Clinical variables including CD4 cell count, clinical stage, and use of antiretrovirals were also considered, as were variables related to quality of life and social support. Comparison of categorical variables were conducted using χ2 tests, and median values were compared using the Wilcoxon rank sum test. Fisher's exact test was used for 2 × 2 contingency tables in which any expected cell frequency was <5. Stepwise logistic regression was used to determine any joint contributions to difference. Variables included in the regression model were those observed to be statistically significant (p ≤ .05) in the univariate analysis. All reported p values are two-sided.


In total, 711 (40%) participants completed the annual 1995/96 participant survey during the study period. In a comparison with nonresponders, responders were more likely to be male (94% versus 89%; p < .001), to have ever taken antiretrovirals (95% versus 92%; p = .01), to be living in a city with a population >400,000 (62% versus 57%; p < .02), and to be older (39 versus 38; p=.04). The median CD4 cell count of respondents and nonrespondents at the time of the survey was the same(220 cells/mm3).

Of the 711 participants who completed the annual participant survey, 657 (619 men and 38 women) completed the questions on complementary therapy. Of these participants, 256 (39%) had ever used complementary therapies. At the time of the study, participants had a median age of 39 years and median CD4 count of 220 cells/mm3. The majority of respondents reported having a gross annual income ≤$10,000 Canadian (≤$7,300 U.S.) (66%), had never completed a post-secondary degree or diploma(72%), and were taking antiretroviral therapy (80%).

Table 1 compares the demographic and clinical characteristics of program participants who used complementary therapies in addition to medical interventions to those who used only medical treatments. Univariate comparisons indicated that users of complementary therapies were younger (median, 38.5 versus 40.0 years; p = 0.020), were more likely to have an annual income >$7,300 U.S.(73% versus 65%; p = .032), and were more likely to have completed a university degree (31% versus 21%; p= .008). Users of complementary therapies were also more likely to have reported moderate or severe pain(35% versus 25%; p = .007), to have reported spending half or less than half of their day out of bed (18% versus 12%; p = .029), to have reported that their health was somewhat worse or much worse compared with the previous year (43% versus 33%; p = .015), and to have reported more depressive symptoms (median CES-D score, 42 versus 39; p = .008). Clinical factors, including antiretroviral use, World Health Organization(WHO) clinical stage, CD4 count, and AIDS diagnosis, were not significantly associated with the use of complementary therapies.

Demographic and clinical characteristics of 657 persons in the provincial drug treatment program by self-reported use of complementary therapy

Patterns of use of specific dietary and medicinal or tactile and relaxation therapies were investigated and are summarized in Table 2. The number of participants using dietary and medicinal therapies was 229 (35%). Of these persons, 195 (30%) used dietary supplements, and 141(22%) used herbal and other medicinal therapies. The most frequently chosen therapy was the ingestion of vitamins, which were used by 192 persons (29%). This surpassed those using minerals (100; 15%), a wide variety of Chinese herbs (89; 14%), or NAC (45; 7%). Additional medicinal therapies included Compound Q, a purified extract from the root tuber of a Chinese cucumber(Trichosanthes kirilowii), DNCB, homeopathy, and specific diets. The prevalence of use for each of these therapies was <6% of the total number of participants using complementary therapies. The number of persons using tactile and relaxation therapies was 190 (29%). Of these persons, 145 (22%) used tactile therapies, and 128(20%) used relaxation techniques. The most frequéntly chosen therapies were massage (135; 21%), stressreduction techniques (127; 19%), incorporating meditation and various psychosocial therapies, and acupuncture(49; 8%).

Self-reported complementary therapy use for 657 persons in the provincial drug treatment program

The use of complementary therapies by respondents was not necessarily limited to a single therapy or category of therapy. Among those using complementary therapies, 44(18%) had used therapies from all four categories, 78 (31%) had used therapies from three categories, 70 (28%) had used therapies from two categories, and 58 (23%) had used therapies from one category. Six participants using complementary therapy did not report what specific type of therapy they used. Respondents used a median of three therapies overall (range, 1-11).

Motivations for choosing to supplement treatment programs were predominantly driven by a desire to enhance the participant's immune response (30%), followed by the desires to supplement dietary intake (24%) and to prevent infection (14%).

Table 3 presents the final multivariate model of correlates of complementary therapy use by program participants. A total of 65 (10%) study participants were eliminated from this analysis because of incomplete data. Users of complementary therapies were likely to be younger (p = .003), to have an annual income of>$7,300 U.S. (p = .014), and to have received a university degree (p = .002). In addition, persons choosing to integrate complementary therapies into their HIV/AIDS treatment regimens were more likely to spend half or less than half of their day out of bed (p = .051) and to report suffering from a higher level of pain (p = .003). The log likelihood ratio χ2 statistic was used to assess model fit. We reject the null hypothesis that our parameters are equal to zero (p < .001), indicating that the variables increase the utility of the model beyond the intercept-only model of the null. A goodness of fit test to compare users with nonusers was conducted using the Hosmer and Lemeshow χ2 statistic with g-2 degrees of freedom; we stratified the predicted probabilities into five groups. The associated p value, p = .204 (3 degrees of freedom), indicates agreement between the observed and predicted values, suggesting good explanatory power of the model.

Results of logistic regression: association of participant characteristics and complementary therapy use (versus non-use)


In agreement with a number of studies(4-6,11), our results indicate that a significant proportion of HIV-positive individuals use complementary therapies. Those participants using such therapies in combination with medical treatments tended to be younger, to have a higher level of education, to have an annual income level>$7,300 U.S., to report living with a greater severity of pain, and to report spending half or less than half of their day out of bed.

Among the 39% of participants in the drug treatment program using complementary therapies, vitamin supplements were determined to be the most popular form. For the past 20 years, the 1970-72 Nutrition Canada survey has provided the only comprehensive data on the nutritional status of Canadians; however, information about vitamin supplements is not included in the survey. A 1990 survey by Hoffmann-LaRoche drug company found that 28% of Canadian adults take vitamins regularly, with an additional 5% using supplements when they are sick(12).

Our results showed that users of complementary therapies tended to be younger that those who adhered only to traditional therapies. The absolute difference in median age between the groups, however, was slight (38.5 versus 40 years, respectively). Age has not been found to be a predictor in some studies(5,13), whereas others have concluded that users of complementary therapy tend to be older(11).

The relation found between the use of complementary therapies and a higher level of education was consistent with previous studies in this field(4,14). Greenblatt et al.(4) have suggested that a higher level of education may provide the skills necessary to seek and administer therapies not prescribed by medical practitioners. The speculative nature of many complementary therapies often restricts their evaluation in peer-reviewed journals. The onus of investigation of "unproven" therapies has therefore fallen on the patient. Those with a more extensive educational background may conduct such unassisted research with greater facility than those who have a lower level of education. Cassileth et al(15) have also described users of complementary therapies as having a higher level of education. Although the focus of the latter study was restricted to cancer patients, many AIDS patients face both a similar prognosis and comparable complicated therapeutic regimens with associated frequent side effects.

We found users of complementary therapies generally had higher annual incomes than those not using these treatments. This may suggest that persons living below a certain income threshold may be less able or willing to purchase these compounds or treatments. Rowlands and Powderly(13) reported that the majority of persons using such therapies spent between $100 and $2000 U.S. per year on average. Although persons included in this study were receiving antiretroviral or prophylactic medications free of charge, the costs of all complementary therapies were incurred by the participants.

The association between increased pain and the use of complementary therapies presents a potential motivation for seeking supplementary therapeutic interventions. Our results are in agreement with those by Langewitz et al.(16), who have shown that persons with HIV who use complementary therapy are more likely to report a greater degree of suffering. These results suggest that persons suffering from associated effects of HIV may seek complementary therapies for their palliative role rather than curative effects. This may also be reflected in our finding that those using complementary therapies were more likely to spend half or less than half of their day out of bed. This finding, however, is based on a small proportion of study participants, with an absolute difference of just 6% between groups. Users of complementary therapies may also find the self-medicating component of these therapies provides a heightened sense of personal empowerment(2,3).

There are a number of limitations inherent in this type of epidemiologic study. Most importantly, the study sample is a source of potential bias, because the background characteristics and responses of those who participated may not be representative of the total HIV-positive population in British Columbia. Current back calculation estimates suggest that approximately 9000 persons have ever been infected with HIV in British Columbia(17). It is likely that the study sample is not representative of this large, diverse population of HIV-positive men and women. Important subgroups of HIV-infected individuals may not be adequately represented, including street youths, intravenous drug users, indigenous peoples, and recent immigrants. In addition, the study sample may not be representative of other HIV-positive populations across Canada. Nevertheless the majority of study respondents were middle-aged men living in a large urban center, which is consistent with the current epidemiology of HIV/AIDS in British Columbia(18).

We also recognize that inaccurate information may have been reported if the participant did not answer questions truthfully or failed to provide a complete history of complementary therapy use. Although the annual participant survey is anonymous, participants may have felt some reluctance to report therapies deemed unacceptable by medical standards, either for fear of losing access to present HIV/AIDS drug treatment or as a function of a social desirability bias. This unwillingness to reveal use of complementary therapies has been reported in other studies(5,11,13). It is important to note that subjects in this study group were participants in a drug treatment program that provides traditional medications free of charge. Persons in this group therefore may not feel a strong need to seek external treatment options. Had the survey not been restricted to participants of the treatment programs, the findings may have been somewhat different. As always, there is a possibility that other, unidentified variables may play important roles, and the underlying motivations driving complementary therapy use are likely to be multifaceted and remain speculative.

The results of this paper were based on data prior to the introduction of protease inhibitors. The inclusion of these drugs to treatment regimes may alter individuals' desire to seek out complementary therapies. The impact of these new drugs on the use of complementary therapies will be an interesting addition to research in this area.

We have found that a large proportion of HIV-infected individuals receiving antiretroviral, anti-opportunistic, or prophylactic agents in the province of British Columbia use complementary therapies. Complementary therapy use appears to be most prevalent in young and highly educated individuals and to be associated with the debilitating and chronic nature of HIV-disease. Whether motives to choose these therapeutic options are rooted in the desire to regain control over one's body or simply a belief in the immune-enhancing properties of natural medicines, the supplementation of Western medical treatment programs with complementary therapies is a trend warranting further investigation.

Acknowledgments: The authors are indebted to colleagues in the AIDS Care Group at St. Paul's Hospital and to Dianne Campbell, Bonnie Devlin, Rita Dewletian, Elizabeth Ferris, Nada Gataric, Myrna Reginaldo, and Benita Yip for their research assistance. This work was supported by the National Health Research Development Programme of Health Canada through a National Health Research Scholar Award to Drs. Hogg and Montaner and National Health Scientist Award to Dr. Schechter.


1. Hand R. Alternative therapies used by patients with AIDS. N Engl J Med 1989;12:672-3.
2. Steinberg C. Integrating traditional medicine with other therapies in the treatment of HIV-infected individuals. Md Med J 1990;39:183-8.
3. Abrams D. Alternative therapies in HIV infection. AIDS 1990;4:1179-87.
4. Greenblatt RM, Hollander H, McMaster JR, Henke CJ. Polypharmacy among patients attending an AIDS clinic. J Acquir Immune Defic Syndr 1991;4:136-43.
5. Anderson W, O'Connor BB, MacGregor RR, Sanford Schwartz J. Patient use and assessment of conventional and alternative therapies for HIV infection and AIDS. AIDS 1993;7:561-6.
6. Barton SE, Jadresic DM, Hawkins DA, Gazzard BG. Complementary therapies used by people with HIV infection. AIDS 1994;8(4):561.
7. Tilwe S. Role of alternative medicine in HIV infection [abstract PB0879]. Tenth International Conference on AIDS, Yokohana, Japan, August 1994.
8. Kabatesi D. Collaborating with traditional healers in AIDS research, prevention and care in Uganda [abstract 539B]. Tenth International Conference on AIDS, Yokohana, Japan, August 1994.
9. British Columbia Centre for Excellence in HIV/AIDS. Therapeutic guidelines for the treatment of HIV/AIDS and related conditions. British Columbia: St. Paul's Hospital, 1995.
10. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychological Measurement 1977;1(3):385-401.
11. Singh N, Squier C, Sivek C, Nguyen MH, Wagener M, Lu VL. Determinants of nontraditional therapy use in patients with HIV infection. Arch Intern Med 1996;156:197-201.
12. Market builders: new ideas help vitamin makers spur growth in the sector. Marketing 1991;96(51):1-3.
13. Rowlands C, Powderly WG. The use of alternative therapies by HIV-positive patients attending the St. Louis AIDS Clinical Trials Unit. Mo Med 1991;88:807-10.
14. Cohen CJ. Determinants of nonconventional treatment use among HIV-infected individuals [SD 781]. Sixth International Conference on AIDS, San Francisco, 1990.
15. Cassileth BR, Lusk EJ, Strouse TB, Bodenheimer BJ. Contemporary unorthodox treatments in cancer Medicine. Ann Intern Med 1984;101:105-12.
16. Langewitz W, Ruttiman S, Laifer G, Maurer P, Kiss A. The integration of alternative treatment modalities in HIV infection: the patient's perspective. J Psychosom Res 1994;38(7):687-93.
17. Archibald C, Lior L. Cumulative HIV estimates for Canada. Proceedings of the Fourth Annual HIV Epidemiology Research Meeting, Banff, Alberta, 1995.
18. Strathdee S, Schecter MT, Hogg RS, O'Shaughnessy MV. Current trends in HIV/AIDS epidemiology in British Columbia. BC Med J 1994;36:98-115.

HIV infection; AIDS; Complementary therapy use; Canada

© Lippincott-Raven Publishers.