• Race Bias in the Diagnosis of Schizophrenia

      Fernandez, Rose M. (2013-01-30)
      Accuracy of diagnoses appears to be influenced by the client’s race. Previous research indicates that African Americans are overdiagnosed with Schizophrenia, whereas findings for Hispanic patients have been inconsistent. However, there is evidence that the diagnostic approach used by the clinician may influence accuracy and that bias may be reduced by using a structured assessment of symptoms prior to assigning a diagnosis. There is also some limited evidence for the role of clinician race in diagnosis, with Caucasian clinicians more likely than non-Caucasian clinicians to diagnose African American patients with Schizophrenia. This study utilized a vignette methodology to examine the effects of race of the client, diagnostic approach, and clinician race on diagnoses. A national sample of psychologists read two cases containing a mixture of psychotic and mood disorder features in which patient race was varied, rated the symptoms in the cases, and assigned diagnoses. Diagnostic approach was manipulated by asking half of the clinicians to rate symptoms before assigning a diagnosis (i.e., simulated DSM-IV approach) and half to assign a diagnosis before rating the symptoms in the case (i.e., simulated prototype approach). It was hypothesized that the African American version of the cases would receive more diagnoses of Schizophrenia, particularly by Caucasian clinicians and when clinicians utilized the prototype approach. The results of the study provided little support for the hypotheses. There were few differences based on patient race, diagnostic approach, or significant interactions between them, and the best predictors of diagnoses were the symptom ratings. There also were no significant interactions between clinician race and patient race, although minority clinicians assigned more mood disorder diagnoses and fewer Schizophrenia diagnoses than Caucasian clinicians, regardless of patient race. This latter finding may be a function of the particular symptoms included in the two cases or the low number of minority clinicians, despite efforts to recruit a diverse sample. Overall, clinicians were reasonably consistent in their assessment of the symptoms and the assigned diagnosis, and were not biased by patient race or diagnostic approach. This is an encouraging finding and suggests that these clinicians may be more aware of concerns about misdiagnosis than in the studies conducted in the past, although the limitations of the study design may have contributed to the lack of significant findings. Despite the mostly negative findings and failure to support the hypotheses, the influence of race and other individual differences on assessment and diagnosis is an important topic and worthy of continued research.