Title

Patterns of substance use and substance use treatment predict viral suppression over time among HIV-positive women in the Women's Inter-agency HIV Study (WIHS): A cross-classified repeated measures latent-class analysis

Presenter

Daniel Feaster

Authors

D. Feaster¹, M. Philbin², C. Parish¹, T.K. Lee¹, E. Golub³, T. Wilson⁴, A. Sharma⁵, S. Kassaye⁶, J. Cohen⁷, A. Adimora⁸, I. Ofotokun⁹, D. Konkle-Parker¹⁰, J. Cook¹¹, R. Cook¹², L. Metsch²

Institutions

¹University of Miami, School of Medicine, Miami, United States, ²Columbia University Mailman School of Public Health, Sociomedical Sciences, New York, United States, ³Johns Hopkins Bloomberg School of Public Health, Epidemiology, Baltimore, United States, ⁴SUNY Downstate Medical Center, Public Health, Brooklyn, United States, ⁵Albert Einstein College of Medicine, Medicine, Bronx, United States, ⁶Georgetown University Medical Center, Medicine, Washington, United States, ⁷University of California San Francisco, San Francisco, United States, ⁸University of North Carolina Chapel Hill, School of Medicine, Chapel Hill, United States, ⁹Emory University Department of Medicine, Infectious Diseases, Atlanta, United States, ¹⁰University of Mississippi Medical Center, Infectious Diseases, Jackson, United States, ¹¹University of Illinois at Chicago, Psychiatry, Chicago, United States, ¹²University of Florida, Epidemiology, Gainesville, United States

Background: HIV viral suppression is an important indicator of effective HIV treatment, and decreases the likelihood of transmission. Substance use (SU) can negatively impact HIV-positive individuals'' care engagement, including a higher likelihood of viremia and transmission. Therefore, it is important to understand the relationship between SU, SU treatment, and viral suppression.
Methods: Data from 2512 women were collected during six semiannual visits (10/2013-09/2016) at nine Women''s Interagency HIV Study (WIHS) sites-WIHS is the largest U.S. cohort study of HIV-positive and HIV-negative women. We compared the use of single substances and SU treatment indicators to viral suppression, and then used latent class analyses of these domains to predict viral suppression.
Results: Substance use patterns were similar across HIV-positive and HIV-negative women. Among HIV-positive women, bivariate models by substance type found that stimulant users had the largest difference in viral suppression between users and non-users (74% versus 82%). In latent class analysis, individuals with consistent use of stimulants, marijuana and binge drinking had the lowest percentage of suppression (62%), with those consistently using all substances with the next lowest percentage (69%). The class that predominately used marijuana (82% virally suppressed) and the class with very low use of any substance (85% virally suppressed) had the highest rates. The interaction of SU and SU treatment classes showed that the rate of virally suppressed individuals in the Low-SU and No-SU-Treatment classes increased from 81% to 88% (baseline to 30 months), whereas the Declining-Use class and all classes including stimulants who were also in the No-SU treatment class had a decreased likelihood of viral suppression from baseline to 30 months. Individuals in Increasing-Professional-Treatment who used stimulants and other substances had the lowest viral suppression at baseline (36%) but also greater increases (13%) than the Low-Use/No-SU-Treatment individuals. Consistent Alcoholics/Narcotics Anonymous (AA/NA) attendance was associated with higher rates of viral suppression across all substance use classes.
Conclusions: Substance using women who attended treatment showed improvement in HIV viral suppression rates relative to those not in treatment. Latent class analysis provides a sharper picture of how specific patterns of SU and SU treatment affect HIV-positive women''s maintenance of viral suppression.

[Substance Use Over Time: 7-Class Solution]

Percent Virally Suppressed	Substance Use Treatment Classes
	Increasing Professional Treatment/ No AA/NA		Decreasing Treatment		Low Professional Treatment/ High AA/NA		No SU Treatment
Substance Use Classes	Baseline	30 Months	Baseline	30 Months	Baseline	30 Months	Baseline	30 Months
Declining Use	55%	69%	70%	82%	68%	96%	73%	69%
Mostly Marijuana + Marijuana & Binge + Binge Drinkers	89%	83%	50%	77%	98%	99%	75%	84%
Stimulants, Marijuana & Binge Drinking + Uses All Substances	36%	49%	70%	68%	24%	88%	68%	64%
Low Use	64%	78%	80%	78%	90%	95%	81%	88%

[Percent viral suppression by substance use classes and substance use treatment classes]

Download the e-Poster