Examining homeless experienced Veterans’ perceived unsafety within integrated VA primary care settings: A structural equation modeling approach

Abstract: Background: Core challenges to primary care delivery for homeless-experienced patients include stigma, prolonged physiological and emotional stress, and a generalized sense of unsafety. The Generalized Unsafety Theory of Stress (GUTS) draws on prolonged stress and stigma literature and posits these factors contribute to generalized ‘perceived unsafety’ for marginalized populations. To date, no research has quantitatively tested if these factors operate as one measurable construct or as separate dimensions of primary care quality. Elucidating the construct of perceived unsafety may allow for improved interventions and policy approaches for primary care. This study offers the first effort to apply GUTS and the construct of perceived unsafety in homeless-focused primary care using a large national survey of homeless-experienced veterans (HEV). Methods: This quantitative study uses national survey data of 5,766 HEV collected in 2018 across 26 VA clinical sites. GUTS theoretically guided selection from a validated 33-item Primary Care Quality-Homeless (PCQ-H) experience survey. Confirmatory factor analysis (CFA) and structural equation modeling (SEM) formed the perceived unsafety latent construct. Results: The sample of 5,766 HEV were empaneled VA primary care users. Mean age was 57 (SD 14.86), with 89.5% male, 8.8% female (1.8% other), 43.2% White, and 42% Black. Significant medical and mental health comorbidities were common in this sample. The Perceived Unsafety construct resulted in following acceptable model fit parameters χ 2 (75) = 1796.939, p ≤ 0.000, RMSEA ≤ 0.063 (90% CI 0.061 – 0.066), SRMR ≤ 0.053, CFI ≤ 0.926, TLI ≤ 0.910. Social support, physical health, mental health, and substance use diagnoses emerged as significant predictors of perceived (un)safety. A significant moderation of clinic type was found between mental and physical health and perceived unsafety. No moderation of perceived unsafety was found for age and onset of homelessness. Conclusions: GUTS and the construct of perceived unsafety was theoretically and statistically found to be an acceptable measurable construct of primary care perception for HEV. Further analyses found that perceived unsafety was significantly related to the many social, mental, and physical health concerns often identified in the HEV population. These findings offer safety as novel lens into HEVs perception of integrated primary care.