Improving Diagnostic Safety for Patients of Color

Each year, roughly 12 million Americans are affected by diagnostic error. Due to inequities in healthcare access and health outcomes, BIPOC patients may experience greater risk through a combination of factors, including reduced access to care, communication barriers, and explicit racial bias upon presenting for care.

Amid an ongoing effort to improve diagnostic safety in U.S. healthcare, TDC Group hosted presentations and a panel discussion featuring three prominent physicians to address how medical professionals and organizations can make healthcare safer by making it more equitable. Strategies to reduce disparities in diagnosis include addressing both explicit and implicit bias, filling gaps in research data, increasing clinician access to peer networks and decision support, and improving access to healthcare for BIPOC communities.

The panelists were Elizabeth A. Howell, MD, MPP, Harrison McCrea Dickson President’s Distinguished Professor and Chair of the Department of Obstetrics and Gynecology, Perelman School of Medicine at the University of Pennsylvania; Lakshmi Krishnan, MD, PhD, Internist and Cultural Historian of medicine at Georgetown University, where she is the Founding Director of the Medical Humanities Initiative; and Ronald Wyatt, MD, MHA, Senior Fellow at the Institute for Healthcare Improvement and Private Healthcare Consultant.

Obstacles Hamper Timely, Accurate Diagnosis for BIPOC Patients

Diagnostic error is the number one cause of serious harm among medical errors that are followed by medical malpractice suits. BIPOC patients may be at greater risk of these harms for many reasons, including:

Bias: Sadly, explicit racial bias is still affecting diagnostic accuracy for BIPOC patients. Dr. Krishnan defines explicit racial bias as “prejudice or discrimination on a conscious level. Examples might be openly racist or pejorative comments on ward rounds.”

More subtle and more pervasive, Dr. Krishnan says, is implicit racial bias, “which we all carry. And it's a form of cognitive bias.” Examples of implicit bias that impact the diagnostic process include “downgrading patient-reported symptoms, spending less time speaking with and gathering information from patients for whom English is a second language, or making linguistic assumptions.”

Missing data: Before medical students complete their training, their ability to accurately diagnose patients is already hampered by “lack of an evidence base with diverse racial, ethnic representation, and lack of medical education showing disease states across the continuum of race and ethnicity,” Dr. Krishnan says. The medical data on which students are trained is still derived overwhelmingly from white men, which increases the risk of misdiagnosis in women and people of color, in whom conditions may present differently.

Dermatology stands out as an example. “I remember when I went to medical school, we did not learn how particular conditions appeared in the skin of people of color,” says Dr. Krishnan, “leading to missed or delayed diagnoses.” Lack of skin tone representation in medical training can directly impact patients’ outcomes and lifespans.

Reduced access to care: Some patients struggle simply to access healthcare, due to negative social determinants of health. For instance, says Dr. Howell, “In Philadelphia, the most recent maternal mortality report suggests that one-fifth of all the pregnancy-related deaths had no prenatal care.” This lack of access to care, like other healthcare disparities, is not equally distributed, and patients of color face significant barriers. These barriers endanger their prospects for timely symptom recognition, with all its benefits to the diagnostic process.

Diagnosis Improves with Standardization, Collaboration, and Increased Access to Care

While misdiagnosis presents safety risks for patients and liability risks for providers, improving diagnosis presents a major opportunity to mitigate those risks from the start of a patient’s healthcare journey. Opportunities to improve diagnostic accuracy engage both healthcare systems and individual medical professionals:

Standardized workflows and decision support: Referencing a new series on diagnostic excellence from JAMA, Dr. Krishnan advised, “To address clinician bias, one area is standardized workflows and decision support that prompt clinicians to consider specific presentations.” Increasing clinician access to these tools may reduce the influence of implicit bias.

Peer networks: The loss of the physicians’ lounge is often cited as a contributor to burnout, but bringing back various types of spaces for medical professionals to congregate and converse helps patients, too. Dr. Krishnan says that diagnostic accuracy improves “when physicians receive treatment recommendations collaboratively through an information exchange and a peer network, versus our standard process of solo, independent reflection.”

Community assets to enhance access: All panelists agreed that partnerships with trusted community organizations, such as places of worship and community centers, were invaluable in improving timely seeking of care, thus promoting timely symptom recognition. In other words, community networks can help with what Dr. Howell calls “engaging patients into care.”

Dr. Wyatt also emphasized the longstanding and valuable role of community health workers: “If you look at the history in the United States, the Indian Health Service, they have used community health workers for decades,” he says. “And that's a way also to build a connection, communication, and trust out to individual practitioners' offices and partner.”

This access-to-care piece of the puzzle is vital, all panelists agreed, as it syncs up to diagnostic-safety-promoting measures in healthcare systems, such as decision support for clinicians and access to peer information exchange. And as Dr. Wyatt puts it, “Now is the time to partner with local physicians and other people in the community like we never have before.”