Differences in Patient–Provider Communication for Hispanic Compared to Non-Hispanic White Patients in HIV Care

Abstract

Hispanic Americans with HIV/AIDS experience lower quality care and worse outcomes than non-Hispanic whites. While deficits in patient-provider communication may contribute to these disparities, no studies to date have used audio recordings to examine the communication patterns of Hispanic vs. non-Hispanic white patients with their health care providers. To explore differences in patient-provider communication for English-speaking, HIV-infected Hispanic and non-Hispanic white patients. Cross-sectional analysis. Two HIV care sites in the United States (New York and Portland) participating in the Enhancing Communication and HIV Outcomes (ECHO) study. Nineteen HIV providers and 113 of their patients. Patient interviews, provider questionnaires, and audio-recorded, routine, patient-provider encounters coded with the Roter Interaction Analysis System (RIAS). Providers were mostly non-Hispanic white (68%) and female (63%). Patients were Hispanic (51%), and non-Hispanic white (49%); 20% were female. Visits with Hispanic patients were less patient-centered (0.75 vs. 0.90, p = 0.009), with less psychosocial talk (80 vs. 118 statements, p < 0.001). This pattern was consistent among Hispanics who spoke English very well and those with less English proficiency. There was no association between patient race/ethnicity and visit length, patients' or providers' emotional tone, or the total number of patient or provider statements categorized as socioemotional, question-asking, information-giving, or patient activating. Hispanic patients gave higher ratings than whites (AOR 3.05 Hispanic vs. white highest rating of providers' interpersonal style, 95% CI 1.20-7.74). In this exploratory study, we found less psychosocial talk in patient-provider encounters with Hispanic compared to white patients. The fact that Hispanic patients rated their visits more positively than whites raises the possibility that these differences in patient-provider interactions may reflect differences in patient preferences and communication style rather than "deficits" in communication. If these findings are replicated in future studies, efforts should be undertaken to understand the reasons underlying them and their impact on the quality and equity of care.