A new study suggests that Black patients with acute COVID-19 infection are less likely to be diagnosed as having long COVID than White patients, despite no significant difference in long-COVID symptom burden between the two groups. 

For the study, published this month in Health Affairs, researchers led by a team at Stanford University and the University of California, Berkeley analyzed electronic health records for roughly 2.4 million US patients diagnosed with acute COVID from January 1, 2022, to March 31, 2023. The researchers identified six race and ethnicity categories and 222 symptoms within one year after infection.

They found that, when at least one long-COVID symptom was present among patients who visited a healthcare provider at least bimonthly, Black patients were less likely to be diagnosed as having long COVID than were White patients.

Few long-COVID patients receive a formal diagnosis

According to the findings, Black patients were diagnosed with long COVID at a 6% lower rate than White patients (controlled direct effect relative risk, 0.938; 95% confidence interval [CI], 0.887 to 0.992), even when they had similar long-COVID symptom profiles. The researchers did not find any statistically significant differences in diagnosis rates between White patients and patients who were Hispanic or Latino, Asian American or Pacific Islander, or American Indian or Alaska Native.

The researchers also looked at which symptoms were most closely associated with receiving a long-COVID diagnosis in different racial and ethnic groups. Across nearly every racial and ethnic group, pulmonary symptoms, such as cough or shortness of breath, showed the strongest correlation with a long-COVID diagnosis, although the correlations were weak. Among Asian American and Pacific Islander patients, loss of taste or smell had the strongest association with receiving a diagnosis.

In addition, the researchers highlight a broader problem in diagnosing long COVID. Previous research suggests that more than 10% of patients with acute COVID infection may go on to develop long COVID, but electronic health records tend to indicate that less than 1% of patients receive a long-COVID diagnosis. 

“This disparity between true incidence and observed diagnostic proportions raises concerns regarding diagnostic bias,” the researchers write. 

More objective diagnostic criteria needed

More objective diagnostic criteria could help reduce disparities, the researchers argue. “However, objective measures are unlikely to remove diagnostic inequity entirely, as bias may be present in providers’ decisions regarding whether to screen patients, diagnostic effectiveness, and interpretation of diagnostic results,” they add.

Bias may be present in providers’ decisions regarding whether to screen patients, diagnostic effectiveness, and interpretation of diagnostic results.

The findings align with previous research suggesting that racial and ethnic bias among healthcare providers often influences patient care, and it highlights the need for improved equity in healthcare settings.

The researchers conclude that provider education on implicit bias, targeted outreach to underserved populations, and improved diagnostic tools could help improve diagnostic equity and ensure all long-COVID patients are properly identified and treated.