UC Berkeley

Bipolar disorder is a serious mental illness that goes untreated more often among blacks than whites in the U.S. In the current study, I identified and tested a proposed mechanism underlying differential treatment in bipolar disorder, namely how treatment providers’ racial biases contribute to the inadequate care of black patients with bipolar disorder. Findings from research in general and mental healthcare show that providers’ clinical decisions may be related to racial biases based on stereotypes that providers hold. Moreover, general healthcare research has demonstrated the presence of implicit racial biases (defined as unconscious preferences for one race over another) among providers that are linked to the differential treatment of blacks and whites. Unfortunately, much less work has been done to show the presence and effects of implicit racial biases among mental healthcare providers. The hypothesized model in the current study proposes that implicit racial biases held by mental healthcare providers bias diagnostic and prognostic assessments as well as treatment recommendations in bipolar disorder and result in the observed treatment disparities between black and white patients. The Implicit Association Test was used to measure general racial biases in a sample of 82 mental healthcare providers. Clinical decision-making was simulated using clinical vignettes about theoretical black and white patients with psychotic and non-psychotic bipolar disorder. Three-way ANOVAs were used to test for racial disparities in decision-making, and structural equation modeling was used to determine whether providers’ implicit racial biases were associated with decision-making outcomes. Results showed that providers held moderate pro-white/anti-black racial biases and neutral explicit attitudes, also known as an “aversive racism” profile. Racial disparity was observed, but confounded by a mismatch in the severity of symptoms of the hypothetical patients. Composite scores for the vignettes were used to overcome this issue in the SEM analyses. These results revealed that there was no link between racial bias and decision-making. Suggestions for future research are given to improve modeling of the relationship between racial bias and patient care. Despite limitations, findings provide some of the first evidence establishing the presence or implicit racial biases in mental healthcare.