Christopoulos, Katerina

Objectives To understand the trade-offs between different statistical modeling approaches, using real world data with small sub-populations, with rare exposures and increasingly rare outcomes. In particular, to compare adjusted regression, inverse probability weighting, and matching. Methods Data for these analyses came from the RADAR (N=1,134) and combined CNICS/JHHCC (N=14,434) cohorts. We estimated prevalence ratios (PRs) for self-reported use of specific substances comparing subpopulations (SP), SP-1 vs SP-3 and SP-2 vs SP-3, where SP-1 was the largest proportion (92% in RADAR, 81% in CNICS/JHHCC), SP2 was moderate proportion (18% in CNICS/JHHCC) and SP-3 was the smallest proportion (8% in RADAR, 1% in CNICS/JHHCC) of the population. We calculated PRs using 1) unadjusted relative risk regression (RR); and adjusted estimates controlling for age, race/ethnicity, study site, and year of interview using: 2) standard adjustment in RR; 3) stabilized inverse probability of treatment weighting (IPTW); and 4) matching with up to 3 matches from SP-1 or SP-2 per SP-3 participant. Results For most substances, all methods yielded consistent estimates. There were large weights in some of the IPTW analyses and in three cases these resulted in substantially divergent estimates. For the comparison between SP-1 and SP-3, the estimate for smoking was 1.3-fold greater in the matched analysis (PR=1.33, 95% CI: 1.02-1.75) than in IPTW (PR=1.03, 95% CI: 0.78-1.37 ATE and PR=1.05, 95% CI: 0.88-1.26 ATT). Even more extreme divergence in estimates was observed for differences between SP-2 and SP-3 with respect to methamphetamine/amphetamine, (IPTW-ATE: PR=1.51, 95% CI: 0.79-2.89; IPTW-ATT: PR=2.36, 95% CI: 1.36-4.12); vs Matching: PR=2.71, 95%CI: 1.47-4.99) and cocaine (IPTW-ATE: PR=1.15, 95% CI: 0.55-2.38; IPTW-ATT: PR=1.81, 95% CI: 1.03-3.18); vs Matching: PR=1.38, 95%CI: 0.76-2.53). Conclusion The combination of a rare exposure and a rare outcome can produce challenges for commonly used confounding adjustment strategies, and it is often best to compare different modeling approaches to gain greater insight.

Background: Internalized HIV stigma (IHS) is associated with reductions in antiretroviral therapy (ART) adherence and HIV viremia mediated through depression. However, there is still a need to quantify the direct impact of IHS on ART adherence to inform interventions to improve medication adherence. Methods: The Center for AIDS Research Network of Integrated Clinical Systems is a longitudinal, US-based, multisite cohort of people with HIV who complete patient-reported outcome assessments as part of HIV-care visits. Patient-reported outcome data include IHS items, ART adherence, depression assessments, substance use, and other outcomes. We examined associations between IHS and ART adherence using generalized linear latent and mixed models with a nonparametric random effects intercept to accommodate repeated measures. Results were compared with analyses using generalized estimating equations and marginal structural models. Results: Among 13,119 people with HIV, the mean age was 47.4 years, 17.6% were women, and 59.3% were non-White. Across 33,139 observations, controlled for repeated measures on individuals, HIV medication adherence was reduced by 1.07% for every point increase in IHS after controlling for age, sex, ethnicity, geographic location, and substance use, and 0.58% when additionally adjusted for depression score. Marginal structural models and generalized estimating equations provided similar results. Conclusions: Our study demonstrates that IHS has a direct association with reductions in ART adherence, which could affect other comorbid health outcomes that are influenced by unsuppressed viremia over time. Developing a thorough understanding of the mechanisms through which IHS affects ART adherence is critical for developing interventions to mitigate IHS and improve health outcomes across the lifespan.

Background: Current understanding of associations between alcohol use and falls among people with HIV (PWH) is limited due to few and inconsistent studies. Given the prevalence of heavy episodic alcohol use reported among PWH, this remains an important question. Methods: Alcohol use was measured using the Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) and self-reported alcohol use treatment and/or diagnosis of alcohol use disorder (collectively AUD) among PWH at eight CFAR Network of Integrated Clinical Systems (CNICS) sites from 2021 to 2024. Falls were self-reported using the components of the Falls Risk for Older People in the Community (FROP-Com) questionnaire. Cross-sectional associations between alcohol use and falls were examined using relative risk models adjusted for age, sex, race/ethnicity, and geographic location. Results: Among 5729 PWH, mean age was 55 years; 42% reported White and 41% reported Black/African American race/ethnicity; and 81% were male. Mean AUDIT-C score was 1.9 (range: 0–12); 27% reported heavy episodic drinking; 22% had AUD; and 22% reported falling in the past year. Reporting at least one fall in the past year was not associated with any measure of current alcohol use. AUD was associated with reporting at least one fall [prevalence ratio = 1.54; 95% confidence interval (CI) 1.39–1.70] and number of falls [one fall: prevalence ratio = 1.27 (95% CI: 1.08–1.50); at least two falls: prevalence ratio = 1.84 (95% CI: 1.58–2.15)]. Conclusion: Among PWH, 22% reported falling, which was associated with AUD, but not current alcohol use, suggesting that a mechanism other than acute intoxication could be responsible. Further longitudinal research is warranted to explore mechanisms for why AUD, but not current alcohol use, is associated with falls.