Extreme heat is an increasing health threat in the United States, contributing to an estimated 12,000 deaths annually. Heat-related illness (HRI), although largely preventable, is a multifactorial condition whose risks are unevenly distributed across populations. Place-based factors are known to shape heat vulnerability, yet the mechanisms that moderate individual HRI risk remain unclear. This study examines how tract-level contextual factors interact with individual characteristics to shape the severity of HRI.

Using HRI case data from the University of Alabama at Birmingham Hospitals, multilevel logistic models were estimated for patients nested within residential Census tracts. Contextual predictors included housing stock age, redlining history, and neighborhood education, while individual characteristics included race, sex, age, insurance status, and heat-sensitive preexisting conditions (hypertension, obesity, and autoimmune disease).

Results revealed patterned cross-level heterogeneity in HRI severity. Sensitivity to built-environment effects was concentrated among White males, who were more than five times as likely to experience Severe HRI than Black females (OR = 5.12, 90% CI = 3.02–8.67). Higher neighborhood education was selectively protective for Black females, adults aged 30–49, insured patients, and non-obese individuals. Patients without preexisting conditions were more responsive to contextual tract-level effects than those who had them, suggesting saturation of individual vulnerability.

These findings highlight the importance of modeling cross-level interactions to identify mechanisms through which structural context shapes climate-related health risks. Population health interventions focused solely on individual risk factors may overlook structural neighborhood conditions that differentially shape vulnerability across population subgroups.