Clinical and Structural Correlates of COVID-19 Hardship in Latino Adults

The COVID-19 pandemic illuminated longstanding structural inequities in health access and outcomes across the United States, with Latino communities consistently among the most affected.

To get a more comprehensive understanding of pandemic-related hardships within this population, a recent Maryland-based study—published in June 2025 in the Journal of Racial and Ethnic Health Disparities and part of the NIH RADx-UP initiative—used latent class analysis (LCA) to identify varying levels of risk for specific subgroups. The analysis, drawn from 720 participants, reveals four distinct profiles based on self-reported challenges across six key areas: healthcare, housing, food, water, transportation, and medication access. These data offer a more differentiated view of pandemic impact and clarify how structural conditions cluster across distinct segments of the population.

Here's a brief look at the study and what it found.

Study Design and Analytical Framework

Participants, recruited between August 2022 and August 2023, completed a structured survey covering six core indicators of hardship, mental health screening (Generalized Anxiety Disorder 2-item and Patient Health Questionnaire-2), self-rated health, vaccination history, and demographic variables. Each hardship was rated as a major challenge, minor challenge, or not a challenge. Additional items captured employment status, English proficiency, internet access, health insurance, and years spent in the US.

The study applied LCA to identify unobserved subgroups based on patterns in how participants experienced these challenges. Fit indices including the Akaike information criterion (AIC), Bayesian information criterion (BIC), the likelihood ratio chi-square, and entropy supported a four-class model as the best representation of the data. Researchers then conducted multivariate logistic regressions to assess how class membership related to sociodemographic traits and health outcomes, using the least-burdened class as a reference.

Four Distinct Hardship Profiles

The LCA revealed four experience-based subgroups. About one-third of participants (31.7 percent) reported minimal hardship across all indicators. Another third (32.6 percent) experienced targeted difficulties, especially with access to food, medicine, and healthcare. A smaller group (13.4 percent) reported minor challenges across all areas, and 22.3 percent of respondents faced major challenges across the board.

Sociodemographic variables closely aligned with class membership. Participants in the highest-burden group were more likely to lack health insurance, be unemployed, have limited English proficiency, and report fewer years in the US. They also tended to live in unstable housing and were less likely to have internet access. These markers—particularly the intersection of insurance, language, and duration of US residency—outlined a structurally excluded segment of the population.

While immigration status was not directly measured, these characteristics suggest informal exclusion from federal relief efforts, even when emergency policies technically guaranteed access to testing and care. Participants in high-burden classes were also less likely to report receiving COVID-19 testing, vaccination, or flu shots, despite being among those most at risk.

Mental Health and Health Status Differentiation

The mental health results did not follow a simple gradient. Participants in the group with minor but widespread challenges were most likely to screen positive for both anxiety and depression and reported the poorest overall health. By contrast, those in the highest-burden class had significantly elevated odds of depression but not anxiety, nor significantly worse self-rated health.

These findings suggest that persistent moderate strain may be more psychologically taxing—or at least more openly reported—than more severe hardship, which may prompt different coping mechanisms or social adaptation. The study did not measure protective factors like social support, which could moderate these relationships, but it raises the possibility that accumulated stress does not always correspond to obvious clinical flags.

Implications for Access and Outreach

Access to services followed predictable structural lines. Participants with insurance, internet access, and English proficiency were significantly more likely to report receiving COVID-19 tests and vaccines. These patterns indicate that health behavior is not merely a matter of personal choice or belief—it is frequently governed by resource infrastructure.

Outreach strategies dependent on digital scheduling, English-language materials, or proof of eligibility may have missed entire subgroups. While broad access policies were in place during the pandemic’s peak, the operational design of many programs assumed a level of system fluency that did not match the realities of many immigrant households.

Planning for Post-Pandemic Response

These findings point toward the importance of subgroup differentiation in public health planning. Demographic labels such as “Latino” can obscure important internal variation when not coupled with measures of structural access. As emergency policies phase out, returning to traditional eligibility frameworks risks further marginalizing groups who were least likely to benefit even during peak relief periods.

Maintaining expanded language access, flexible service delivery, and insurance-independent entry points may be necessary not only for equity but for baseline public health performance. If those in greatest need cannot access or trust core services, system resilience is weakened in future crises.

Latent Class Analysis as a Public Health Tool

Beyond its immediate findings, the study demonstrates the utility of LCA in segmenting population-level risk. This method identifies patterns of lived experience that offer more targeted avenues for intervention. The resulting classifications may help health systems allocate limited resources more efficiently, design outreach based on actual need profiles, and avoid the bluntness of one-size-fits-all frameworks.

Reference:
Ramirez, G., Ou, Y., Saxton, R. et al. Patterns of social, economic, and health challenges due to the COVID-19 pandemic experienced by Latino communities: a latent class analysis. J Racial Ethn Health Disparities. Published online June 27, 2025. doi:10.1007/s40615-025-02514-6