During Deborah Cohen’s first job, as a case manager for adults with a serious mental illness, an older, experienced psychiatrist told her that she would be working with many African American clients, because schizophrenia was more prevalent in this population.
“This was presented to me as a fact, with no explanation needed,” Cohen said. “These myths, false beliefs, and biases can become a sort of self-fulfilling prophecy.”
Cohen went on to obtain her doctorate, and is now a research assistant professor at the Steve Hicks School of Social Work and Dell Medical School Department of Psychiatry focusing on community mental health services and policy. For one of her projects, Cohen is working with a community mental health provider to fine-tune the clinical intake process in ways that promise to reduce excessive rates of schizophrenia diagnosis in African American psychiatric patients.
Historically, rates of clinical diagnoses of schizophrenia in African American individuals, particularly male, are elevated compared with rates for other ethnic groups in U.S. clinical settings. This contradicts expectations from epidemiological surveys, which find similar prevalence rates across different ethnic groups.
There is consensus among researchers that much of the disparity can be attributed to clinicians’ bias when diagnosing. There are multiple hypotheses for these biases such as the disconnect between clinician and patient around the language used to describe religious beliefs, a misunderstanding of a person’s past trauma, and unconscious bias. [For a historical perspective on diagnostic bias, see this article].
Schizophrenia is a complex condition that presents with an extensive range of symptoms and signs that also occur in other serious mood disorders such as major depression. Distinguishing schizophrenia from these other conditions can be difficult, although by definition these conditions must be ruled out before diagnosing schizophrenia.
Studies have shown that clinicians may be taking factors that are ambiguous (such as idioms of distress, cultural differences in expressing illness, and healthy paranoia resulting from previous discriminatory or trauma experiences) as evidence of psychosis during their evaluation of African American clients, and thus minimizing the possibility of mood-disorder diagnoses.
A 2012 study across 6 U.S. university medical centers found that African Americans and white patients did not differ significantly in blinded expert ratings of affective symptoms, although African Americans had been given higher rates of un-blinded schizophrenia diagnosis.
“In diagnosis, we are always balancing the art and the science of it. What this study shows is that, when trying to reduce disparities, we have to give more weigh to the science. Clinicians bring all sort of beliefs into the room when they are making the diagnosis, which may lead to misinterpretations,” Cohen said.
In the 2012 study, clinicians used long, standardized and semi-standardized rating scales to collect information to determine the diagnosis. This solution to address the racial disparity in schizophrenia diagnosis, Cohen said, may be hard to implement in community mental health settings.
“Academic medical centers have access to the latest technologies, to clinicians with higher levels training, and to resources such as regular training that make standardization easier. This is not the case for community mental health clinics, where clinicians may have different types and levels of training, and resources are usually scarce.”
A POSSIBLE SOLUTION: POP-UP WINDOWS!
Taking advantage of the widespread use of electronic health records, Cohen’s team is adjusting the diagnosis workflow that clinicians currently use in community mental health clinics.
“We developed an online form that pops up on the clinician’s screen when they give a diagnosis of schizophrenia,” Cohen explained. “The form asks them to confirm that they have checked for and ruled out several other conditions with similar symptoms.”
Implementation of the online form does not bring major changers to the clinics’ current intake process and neither does require special training of staff members or large investment of resources.
“Essentially, the form slow clinicians down and asks them to reconsider before they give a diagnosis of schizophrenia,” Cohen said.
Cohen and her team have analyzed racial and diagnosis distribution data for the past three years and found that adult African Americans are 20% of the population in the clinic but approximately 40% of the clients diagnosed with schizophrenia.
The hope is that the implementation of the online form will make a dent in this disparity.
“We are only six months in, so it is too early to tell if this is working,” Cohen said. “We will keep going for a year and evaluate. If nothing else, we’ll find that the online form is not enough, and that we have to find a feasible way of bringing more consistency and standardization to the overall diagnosis process in community-mental-health settings.”