The Impacts of Adverse Provider Conditions on Provider Bias and Health Disparities

Project Summary Healthcare disparities that adversely affect patients from racial and ethnic minoritized groups are often attributed to provider level factors such as implicit bias. However, efforts to reduce disparities have been hindered by an incomplete knowledge of factors that leave providers more vulnerable to bias and their underlying mechanisms.

This proposal will address this critical knowledge gap by identifying situations that increase bias among healthcare providers that can be intervened on at the provider and system level. The proposed research leverages theory from psychology and psychoneuroimmunology to document the impact of three adverse

provider conditions (APCs) that could increase provider bias: (1) workplace demands (e.g., heightened levels of stress or burnout), (2) concerns about one’s own health, and (3) increased inflammation occurring during illness. We hypothesize that when providers experience higher (versus lower) workplace demands, are more (versus

less) concerned about their own health, and experience higher (versus lower) acute inflammation, they will demonstrate more bias against patients from racial and ethnic minoritized groups. We will test our hypothesis in the context of abdominal pain treatment in emergency departments. Participants will be attending physicians,

nurse practitioners, physician’s assistants, and residents in emergency medicine and internal medicine hospitalists who treat abdominal pain for patients who are admitted. We will address three specific aims to document effects of APCs on quality of care disparities. To address Aim 1, we will use a correlational design to

determine the relationship between APCs and standardized assessments of bias (implicit and explicit bias, perception of others’ pain, and social avoidance). To address Aim 2, we will use a randomized controlled intervention design; we will use the influenza vaccine to simulate illness by experimentally increasing

inflammation (indexed by pro-inflammatory cytokine levels). Increases in cytokines are hypothesized to be positively associated with standardized assessments of bias. To address Aim 3, we will use a prospective, longitudinal design to tracks provider illness symptoms, stress, burnout, and local cases of infectious illness over

the course of a year. These provider data will be correlated with patient data from abdominal pain cases treated by the provider on those days. This study has high ecological validity and findings will identify distinct contributions of APCs to lower quality of care, especially for patients from racial and minoritized groups. Overall,

the proposed research will have widespread positive impact by outlining paths for novel interventions at multiple levels and domains of influence within the NIMHD Research Framework, such as by engaging administrator and clinical management stakeholders in conversations about consequences of staffing policies for health disparities.