A June 1 study published in JAMA Network Open verifies a medical testing flaw with significant outcomes. The University of Pennsylvania Health System study estimates that between 2010 and 2020, around 40% more Black male patients may have gone undiagnosed with breathing problems.
Doctors have long presumed potential faults in the current diagnostic system, which this study’s results now confirm. The software that tests lung function was programmed with algorithms that include obsolete race-based assumptions. Researchers determined that updated, race-neutral algorithms could have assisted in nearly 400 additional cases of lung impairment in Black patients. The study conversely identified a slight overdiagnosis of lung restriction in white patients.
The computer software is not the only determinant in the diagnostic process; physicians also consider a patient’s symptoms, family history, X-rays and other lab work. However, the pulmonary function testing is an imperative factor, “especially when patients are borderline,” said Albert Rizzo, MD, the chief medical officer at the American Lung Association.
The test, often done with a spirometer, runs through the computer software to generate a report with the patient’s breathing score. This can indicate if a patient is at risk for certain pulmonary conditions, such as asthma or COPD, and requires further testing or treatments.
Initially, the system was built with formulas that included equations for a patient’s race. While other measurements, such as age, height and sex, are biologically based factors, race is a social construct and has no bearing, said a group of Penn LDI (Leonard Davis Institute) fellows. This original assumption that Black people had naturally worse lung function than white people extends to other areas of health, including heart and kidney function, and has since been proven inaccurate and misleading.
Now, the American Thoracic Society recommends replacing the race-based algorithms with modern, unbiased ones. Darshali Vyas, MD, a pulmonary care doctor at Massachusetts General Hospital, acknowledges that this technical adjustment will take time to refine, develop and implement. Some doctors and researchers are concerned with overcorrecting the system or a scenario where hospitals begin using different versions of adjusted software.
