Ann Pharmacother. 1999 Mar; 33(3): 294-300.

Nonprescription and alternative medication use by individuals with HIV disease.

Smith SR, Boyd EL, Kirking DM.

Division of Pharmaceutical Policy and Evaluative Sciences, School of Pharmacy, University of North Carolina at Chapel Hill 27599, USA.

OBJECTIVE: To examine the strength of the associations between predisposing, enabling, and need-for-care variables and the self-treatment of HIV disease; and to compare sociodemographic and illness-related factors associated with the use of vitamins, nonprescription medications, herbs, and recreational substances among HIV-infected individuals. METHODS: Data were derived from 7887 interviews conducted as part of the AIDS Cost and Services Utilization Survey. The conceptual framework was the Andersen Behavioral Model of Health Services Use. Factors associated with nonprescription and alternative medication use were assessed using logistic regression. Generalized estimating equations were applied to adjust variance estimates for within-person correlations of drug use over time. RESULTS: After adjusting for perceived health status, T cell count, and stage of disease, the results indicated that African-Americans were less likely to use nonprescription drugs (odds ratio [OR] 0.65, 95% CI 0.52 to 0.81), vitamins (OR 0.59, 95% CI 0.48 to 0.73), and herbs (OR 0.41, 95% CI 0.22 to 0.76), compared with non-Hispanic whites. Similarly, Hispanics were less likely to report use of herbs (OR 0.58, 95% CI 0.34 to 0.98) or recreational drugs (OR 0.34, 95% CI 0.15 to 0.76) than were non-Hispanic whites. Oppositely, individuals who had a college education were more likely to use vitamins (OR 1.26, 95% CI 1.05 to 1.50) and herbs (OR 2.47, 95% CI 1.56 to 3.91). Enabling variables such as insurance status and income were generally associated only with use of recreational drugs. Need-for-care variables were generally associated only with use of nonprescription drugs and vitamins. CONCLUSIONS: Predisposing, enabling, and need-for-care variables from the Andersen Behavioral Model were significantly associated with the use of four categories of drugs to self-treat HIV disease. However, there was not a consistent pattern across the drug categories.