Sex, Drugs, and Viral Load: Novel Epidemiologic Methods in HIV Research

Abstract

Substance use is highly prevalent among people living with HIV. Although substance use and its association with condom use and HIV viral load have been well-documented in people living with HIV, these associations may be explained by a number of co-occurring conditions including mental health conditions, poor medication adherence, or general health-seeking behaviors. We used two novel approaches to address the substantial confounding expected to affect these relationships. In analyzing the effect of substance use immediately prior to sex on condom use, we focused on young adults living with perinatally-acquired HIV. We also examined sexual reporting agreement and inconsistencies over time to determine reliability of self-report in this population of adolescents and young adults. In Chapter 1, we examined reliability of oral, vaginal, and anal sex reporting in adolescents and young adults living with perinatally-acquired HIV and those perinatally HIV-exposed and uninfected in the Pediatric HIV/AIDS Cohort Study (PHACS) Adolescent Master Protocol (AMP) and AMP Up. We also assessed associations of inconsistent reports with several demographic and health measures. Young adults who reported having sex in adolescence had a much higher proportion of inconsistent responses for anal sex compared to vaginal and oral sex measures. In bivariate analyses, there was no significant difference in inconsistent reporting between young adults living with perinatally-acquired HIV and those perinatally HIV-exposed and uninfected. Living in the South and younger age at first survey were associated with multiple measures of inconsistency, while increased time between survey administrations and male sex were associated with inconsistent report of age at first vaginal sex. In Chapter 2, we employed a case-crossover approach, a case-only design in which each individual is compared with him/herself, to determine if there is an effect of alcohol or drug use by young adults living with perinatally-acquired HIV or their partners on having condomless sex. All sexually active participants in the PHACS AMP Up protocol were asked in annual online surveys to recall the most recent time they had vaginal sex with a condom (control period) and without a condom (case period). Information about substance use as well as situational confounders (i.e., factors that vary between sexual encounters) immediately prior to sex was obtained. Among inconsistent condom users, substance use by a young adult living with perinatally-acquired HIV or their partner increased the odds of having condomless sex versus sex with a condom, and may be a prime target for interventions to increase condom use in this population. In Chapter 3, we estimated individual and joint effects in women living with HIV of hypothetical alcohol and smoking avoidance and ART medication adherence interventions sustained over three years on prevalence of detectable HIV viral load at the end of follow-up. Our estimates can be interpreted as arising from a randomized trial in which participants were randomly assigned and adhered to the specified substance use interventions. Among women living with HIV in the Women’s Interagency HIV Study (WIHS), our estimates do not provide evidence that three-year interventions targeting alcohol use or cigarette smoking alone would decrease prevalence of detectable viral load.