CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

HIV- men. Overweight and obesity may be important predictors of mid-life neuropsychological outcomes and later-life cognitive impairments, and should be considered in prevention and intervention planning.

1071 SYNDEMICS AND RETENTION IN CARE AMONG WOMEN LIVING WITH HIV IN RIO DE JANEIRO, BRAZIL Christine M. Zachek 1 , Lara Coelho 2 , Raquel B. De Boni 2 , Jesse L. Clark 3 , Rosa M. Domingues 2 , Paula M. Luz 2 , Ruth K. Friedman 2 , Angela C. Vasconcelos de Andrade 2 , Valdilea Veloso 2 , Jordan E. Lake 4 , Beatriz Grinsztejn 2 1 University of California San Francisco, San Francisco, CA, USA, 2 Institute Nacional de Infectologia Evandro Chagas (INI/Fiocruz), Rio de Janeiro, Brazil, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 University of Texas at Houston, Houston, TX, USA Background: Syndemic psychosocial and reproductive factors impacting women’s engagement in the HIV care cascade remain understudied worldwide. We hypothesized that syndemic conditions would limit retention in care among a cohort of women living with HIV in Rio de Janeiro, Brazil. Methods: We analyzed baseline syndemic prevalence and correlates of non- retention in the INI-Fiocruz women’s cohort from 2000-2015. A syndemic score was created for a lifetime history of: physical/sexual violence, illicit drug use, adolescent pregnancy (<20 years old), or induced abortion. Stepwise backward logistic regression models identified predictors of non-retention, defined as <2 HIV laboratory results within the first year of cohort enrollment. Two separate models analyzed syndemic contributions to non-retention: Model 1 incorporated individual syndemic variables; Model 2 used the syndemic score. Results: Of 915 women, 18%met criteria for non-retention. Prevalence of syndemic factors was: physical/sexual violence 38.3%, illicit drug use 17.2%, adolescent pregnancy 53.2%, and induced abortion 27.3%. Nearly half (41.2%) experienced ≥2 syndemic conditions. Illicit drug use was associated with non- retention in unadjusted analysis (cOR 2.05, 95% CI: 1.37-3.05), but none of the syndemic variables reached statistical significance in adjusted models. In Model 1, <9 vs ≥9 years of education (aOR 1.59, 95% CI: 1.05-2.42), years with HIV (1.06, 1.01-1.11), and seroprevalent syphilis (1.85, 1.11-3.06) increased the odds of non-retention. ART initiation at ≤3 months (0.54, 0.31-0.93) and 4-24 months (0.57, 0.36-0.92) before enrollment (vs >24 months) and cohort enrollment from 2005-2009 (0.18, 0.12-0.32) and 2010-2015 (0.50, 0.31-0.81) vs 2000-2004 decreased the odds of non-retention. In Model 2, syndemic scores of 2 (1.94, 1.10-3.41) and 3 (2.16, 1.10-4.24) were associated with non-retention, and the effect of other covariates remained the same. Conclusion: The syndemic of psychosocial and reproductive factors can limit retention in care for women living with HIV. Syphilis infection independently predicted non-retention, and could be explored as a syndemic factor in future studies. Interventions addressing sex-specific syndemics are needed to optimize HIV care in this vulnerable population.

Poster Abstracts

1070 EFFECT OF OBSTRUCTIVE LUNG DISEASE ON MORTALITY AMONG HIV+ PERSONS WHO INJECT DRUGS Mariah M. Kalmin 1 , M. Brad Drummond 1 , Gregory D. Kirk 2 , Shruti H. Mehta 2 , Daniel Westreich 1 1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 Johns Hopkins University, Baltimore, MD, USA Background: People living with HIV experience increased prevalence of obstructive lung disease (OLD), even after accounting for greater smoking prevalence in this population. Although excessive lung function decline has been shown to lead to increased mortality in HIV-negative individuals, the effect of OLD on mortality among people living with HIV has not been quantified. We investigated whether the effect of incident OLD on mortality differs by HIV status in a cohort of people with a history of injecting drugs. Methods: The ALIVE study is a longitudinal, observational cohort of HIV- positive and negative people with a history of injecting drugs. This analysis included ALIVE participants who had at least one spirometry measure to assess OLD between 2007 and 2016, excluding those who reported never smoking (5%, n=62) or who had OLD at baseline (17%, n=269). Incident OLD was defined as a first measurement of pre-bronchodilator FEV1/FVC<0.70 during follow-up. The effect of incident OLD on mortality among HIV-positive and negative participants was estimated using an inverse-probability-of-treatment weighted marginal structural model controlling for confounders including baseline age, black race, sex, baseline calendar year, HIV, baseline smoking pack-years, time- varying smoking status, and calendar time. Results: Among 1,216 participants, 272 (22.4%) experienced incident OLD and 157 (12.9%) deaths were observed over a median of 5 person-years (IQR=2-8) of follow-up. In the main analysis, OLD did not have a statistically significant effect on mortality (HR=1.22, 95% CI: 0.83-1.79). In the model that assessed effect measure modification by HIV, HIV-positive participants exposed to OLD experienced an increased risk of mortality (HR=1.72, 95% CI 1.06-2.81), while there was no effect of OLD on mortality among HIV-negative participants (HR=0.80, 95% CI: 0.45-1.42). Conclusion: Although OLD did not have a statistically significant effect on mortality after properly accounting for baseline as well as time-varying confounders, there was an apparent effect of OLD among HIV-positive people with a history of injection drug use. These results highlight the need for greater screening and management of OLD among HIV-positive individuals. Further research is needed to determine if there are particular clinical characteristics of HIV-infection that mitigate the risk of death after the occurrence of OLD.

CROI 2019 421