Race-neutral lung function equations identify severe COPD in Black smokers
Click Here to Manage Email Alerts
Key takeaways:
- Some Black smokers had worse COPD classification after use of race-neutral vs. race-specific lung function equations.
- Abnormal chest CT phenotype and dyspnea prediction was not better with specific equations.
Use of race-neutral/race-free equations to find FEV1 revealed that some Black smokers had more severe COPD than previously found with race-specific equations, according to results published in CHEST.
Further, prediction of abnormal chest CT phenotypes and dyspnea was not improved with use of race-specific equations, according to researchers.
“With rising awareness of structural racism and misconceptions about race in medicine, this is a critical moment for pulmonary clinicians to reconsider the value of continuing to use race when interpreting spirometry measures,” Amy L. Non, PhD, MPH, professor in the department of anthropology at University of California, San Diego, and colleagues wrote.
In a cross-sectional analysis, Non and colleagues assessed two cohorts of current and former smokers (National Health and Nutrition Examination Survey [NHANES] and COPDGene study) to determine if race-specific vs. race-neutral and race-free equations for calculating percent-predicted FEV1 are better at predicting quantitative chest CT abnormalities, dyspnea or Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages.
Before conducting this analysis, researchers generated these equations through cohorts of healthy adults who reported never smoking (NHANES, n = 3,700; 38% white; 21% Black; 18% Mexican-American; 13% Hispanic; COPDGene study, n = 419; 82% white; 18% Black), and then compared them with equations based on Global Lung Initiative (GLI) criteria and Hankinson’s 1999 criteria.
During the comparison analysis of predicted values, researchers found “high correlation” between their generated race-specific equation and those of GLI and Hankinson. Within the healthy cohorts, use of race-neutral vs. race-specific equations resulted in Black individuals having a greater predicted FEV1 and white patients showing no major changes in the predicted value.
Within the COPDGene cohort of smokers, researchers further noted similar percent-predicted FEV1 values generated from their race-specific equation and that of GLI.
Between the two equations, use of the newly developed race-specific equation changed the average GOLD reclassification rate by 1% to 3%, according to researchers.
Notably, use of race-neutral/race-free vs. race-specific equations to find FEV1 led to more Black smokers in the more severely diseased COPDGene cohort to be regrouped under a more severe GOLD stage (19%).
Researchers observed a similar finding in the NHANES cohort of smokers (n = 785), with reclassification of 17.3% of Black smokers to a worse GOLD stage with use of race-neutral/race-free vs. race-specific equations.
Between race-specific and race-neutral equations in models assessing abnormal chest CT scan findings, researchers found comparable AUC values, as well as sensitivity/specificity. This outcome was also seen in models evaluating dyspnea.
Compared with race-neutral and race-free equations, the race-specific equation was not more advantageous in predicting abnormal chest CT phenotypes or dyspnea based on classification error rates.
Lastly, model fit for dyspnea might be improved with use of equations neutral or free of race vs. race-specific equations, according to researchers.
“The effect of adding race as a covariate only marginally improves the fit for some models with the risk of introducing bias driven by environment and social factors,” Non and colleagues wrote. “In light of these concerns, along with the large amount of unexplained variability and the dynamic nature of self-identified race, we maintain that continued use of race-specific equations is not justified.”
Even though the impact of race in pulmonary function testing has become more prevalent over the years, there are still areas that need further research, according to an accompanying editorial by Magnus Ekstrӧm, MD, PhD, associate professor, senior lecturer and supervisor in the department of respiratory medicine, allergology and palliative medicine at Lund University, and David Mannino, MD, PhD, professor at the University of Kentucky College of Medicine and chief medical officer of the COPD Foundation.
“Pulmonary function is used in different settings: clinical (to determine the absence or presence of impairment/disease), administrative (to determine employability or disability) and research (to determine change over time, linkage to other processes, mortality rates),” Ekstrm and Mannino wrote. “How race/ethnicity affects each of these domains when lung function is being interpreted remains an important question that this work and other articles continue to address.”