Date of Conferral

10-27-2026

Date of Award

October 2026

Degree

Ph.D.

School

Health Sciences

Advisor

Peter Anderson

Abstract

HIV disparities may reflect interconnected housing instability, behavioral health conditions, and healthcare access barriers, yet their relationships with HIV-related outcomes remain understudied across New Jersey counties. This quantitative ecological study examined associations between county-level homelessness burden and HIV prevalence, Stage 3 (AIDS) at diagnosis, and the PrEP-to-Need Ratio (PNR) during 2018–2023, with frequent mental distress and binge drinking as potential moderators. The social determinants of health framework guided the study. County-level data from New Jersey Point-in-Time reports, CDC AtlasPlus, NJSHAD, AIDSVu, CDC PLACES, the American Community Survey, and U.S. Census Bureau population estimates were analyzed using generalized estimating equations (GEE). Analytic samples included 105 county-year observations for HIV prevalence (N = 105), 68 for Stage 3 diagnosis (N = 68), and 88 for PNR (N = 88). Negative-binomial, Poisson, and normal-distribution GEE models accounted for repeated observations and contextual characteristics. Homelessness burden was associated with HIV prevalence (r = .672, p < .001), Stage 3 diagnosis (r = .682, p < .001), and PNR (r = −.387, p < .001) in bivariate analyses. Adjusted associations were nonsignificant for HIV prevalence (p = .091), Stage 3 diagnosis (p = .328), and log-transformed PNR (p = .315), suggesting that the bivariate associations were not sustained after adjustment. Moderation effects were also nonsignificant, providing insufficient evidence that behavioral health factors modified these associations. Integrated surveillance of housing, behavioral health, healthcare access, and HIV-related indicators may inform public health planning.

Download

Included in

Epidemiology Commons

Share

 
COinS