Identifying healthcare experiences associated with perceptions of racial/ethnic discrimination among veterans with pain: A cross-sectional mixed methods survey

Background Healthcare experiences associated with perceived racial/ethnic discrimination among patients are poorly understood. Objective Identify domains of patient dissatisfaction associated with perceived racial/ethnic discrimination among patients with pain. Design Semi-structured telephone surveys completed in 2013-2015. Participants White, African American, and Latino participants who reported receiving pain management from 25 Veterans Affairs (VA) Medical Centers. Main measures Surveys included open-ended questions about healthcare satisfaction/dissatisfaction and a measure of perceived racial/ethnic-based discrimination while seeking VA healthcare. Binary indicators for ten qualitative domains of dissatisfaction were derived from open-ended questions. We used multilevel models to identify dissatisfaction domains associated with perceived discrimination, adjusting for patient characteristics and site. Within domains associated with discrimination, we identified the most frequent codes and examined whether patients primarily referenced clinical or non-clinical staff in their experiences. Key results Overall, 622 participants (30.4% White, 37.8% African American, 31.8% Latino; 57.4% female; mean age = 53.4) reported a median discrimination score of 1.0 (IQR: 1.0-1.3) on a scale of 1 to 5; 233 (37.5%) perceived any racial/ethnic discrimination in healthcare. Individually, 7 of 10 qualitative domains were significantly associated with perceived discrimination: dissatisfaction with care quality, facilities, continuity of care, interactions with staff, staff demeanor, unresolved pain, and pharmacy services (ps<0.005). In combined models stratified by racial/ethnic group, 3 domains remained statistically significant: poor interactions for Latinos (adjOR = 5.24, 95% CI = 2.28-12.06), negative demeanor for African Americans (adjOR = 2.82, 95% CI = 1.45-5.50), and unresolved pain for Whites (adjOR = 6.23, 95% CI = 2.39-16.28). Clinical staff were referenced more often than non-clinical staff for all domains (interactions: 51% vs. 30%; demeanor: 46% vs. 15%; unresolved pain: 18% vs. 1%, respectively). Conclusion Negative interpersonal experiences and unresolved pain are strong correlates of perceived racial/ethnic discrimination among patients with pain. Future studies should test whether interventions targeting these domains reduce patient perceptions of racial/ethnic discrimination in healthcare.