# Identifying the vulnerable among the vulnerable: applying quantitative intersectionality methods to assess potential inequities in the HIV continuum of care for people living with schizophrenia in the united States

## TL;DR

This study uses intersectionality methods to identify which groups of people with schizophrenia in the U.S. are most vulnerable to HIV testing and care disparities.

## Contribution

The paper introduces intersectional MAIHDA models to assess how overlapping identities affect HIV care outcomes for people with schizophrenia.

## Key findings

- Older Black men with schizophrenia had lower-than-expected retention in HIV care.
- Intersectional positions explained a small but significant portion of variance in HIV testing and care retention.
- The study highlights the need for targeted interventions to address structural barriers for vulnerable groups.

## Abstract

People living with schizophrenia face disproportionate risk of HIV, yet HIV testing remains low. Differential testing rates and engagement in care may be impacted by compounding social marginalization, partly linked to structural barriers. Grounded in intersectionality, we set out to identify the riskiest intersectional positions for HIV testing and engagement in HIV care in the United States.

We created a retrospective cohort of people living with schizophrenia and matched controls, using 2012 national Medicaid claims data. We coded intersectional positions based on schizophrenia diagnosis, race/ethnicity, sex, and age. We used Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) models to assess intersectional effects for two outcomes, HIV testing and retention in HIV care (RIC) defined as ≥2 CD4 or HIV viral load tests ≥90 days apart.

Of 777,887 patients in the testing cohort, 7.7% tested for HIV; 39% of the 17,913 patients in the RIC cohort were retained in care. In MAIHDA models without fixed effects, intersectional positions explained 12.7% of the variance in HIV testing and 7.4% of the variance in RIC. In final models including fixed and random effects, intersectional positions accounted for 1.4% of the variance in HIV testing and 0.8% of the variance in RIC. Older Black men with schizophrenia had lower-than-expected RIC prevalence in final models.

Intersectional MAIHDA models can identify both vulnerable and resilient intersectional positions. The antagonistic intersectional effects for older Black men with schizophrenia highlight the need for targeted interventions to address structural barriers.

The online version contains supplementary material available at 10.1007/s00127-025-02972-7.