The relationship between substance use and HIV health status

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)



First Committee Member

Gail Ironson - Committee Chair


This study investigated the relationship between past and current substance use and protected health status in people living with HIV. Past substance use variables were based upon diagnostic categories of use, abuse, dependence and severity derived from the SCID DSM-IV and used to generate a dysfunctional drug use factor for each drug class (alcohol, marijuana, cocaine, opioids, hallucinogens, stimulants, sedatives and poly drugs). The percentage of variance explained by these factors across the drug classes ranged from 76% to 92% and the internal reliabilities of the factors ranged from .80 to .87. Current substance use was measured by self-report questionnaire reporting frequency of use in the past month across a range of drug classes. Three central comparisons were made, examining differences in substance use, between an HIV+ group (COMP, n = 201) assumed to have typical disease course and 3 distinct HIV+ protected health groups: (1) long-term survivors of AIDS (LTS, n = 48); (2) healthy low CD4 patients (HLC, n = 37); and (3) non-progressors (NPG, n = 48). The relationships between substance use and disease progression markers, psychosocial variables and health behaviors were also examined. Variables significantly related to both measures of substance use and health status were examined as mediators. Compared to COMP, the LTS group had lower past dysfunctional use of alcohol (beta = -.166; t(170) = -2.19, p < .05) and cocaine (beta = -.152; t(185) = -2.09, p < .05), and lower current marijuana use (beta = .130; t(242) = 2.04, p < .05), and higher current prescription sedative use (beta = .193; t(240) = 3.04, p < .01). Compared to COMP, the HLC group had lower past dysfunctional use of alcohol (beta = -.153; t (167) = -2.03, p < .05) and marijuana (beta = -.188; t(168) = -2.54, p < .05). Compared to COMP, the NPG group had higher past sedative use (beta = .194; t(196) = 3.062, p < .01). Measures of past and current drug use were positively related to drug use coping, behavioral disengagement, perceived stress, anxiety, depression, and medication non-adherence and negatively related to social support, adaptive coping, and safe sex behavior. Neither psychosocial measures nor health behaviors mediated the relationship between past or current substance use and protected health status.


Psychology, Psychobiology

Link to Full Text


Link to Full Text