A95-09 Impact on Adaptation of Race-Neutral Pulmonary Function Tests on Diagnosis and Treatment of Progressive Pulmonary Fibrosis

Rationale Progressive pulmonary fibrosis (PPF) utilizes physiologic criteria with pulmonary function testing (PFT) to define its classification and inform initiation of anti-fibrotic medications. In PPF, there is a paucity of research examining how the transition to race-neutral equations have affected both classification of PPF and treatment for this group of patients. Methods This was a retrospective cohort analysis of patients seen at the Medical University of South Carolina’s (MUSC) interstitial lung disease (ILD) clinics and diagnosed with non-idiopathic pulmonary fibrosis (IPF) ILD. A secondary prospectively collected dataset from the University of Texas Southwestern (UTSW) was used for validation. Incidence of PPF was determined by a 10% reduction in forced vital capacity (FVC) or diffusion capacity of lungs for carbon monoxide (DLCO) within 1 or 2 years calculated using PFT reference equations for FVC (NHANES, GLI Race-Neutral, and GLI Race-Based) and DLCO (Crapo & Morris and GLI). Cochran’s Q test and McNemar’s test were used to evaluate the differences existing in the proportion of diagnoses between the FVC and DLCO reference equations respectively. All statistical tests were two-tailed with a significance level of α = 0.05 and were performed using R version 4.1.3. Results The cohort consisted of 605 patients — 257 (42.5%) from MUSC and 348 (57.5%) from UTSW with a greater majority of African American patients at MUSC relative to UTSW (33.5% vs 9.5%). Among study outcomes looking at diagnoses met within 2 years, a statistically significant difference (p < 0.05) was observed in the reference equations for the UTSW cohort, with the GLI Race-Based equation showing a significant unadjusted difference compared to NHANES; however, no such difference was observed in the MUSC cohort. Evaluation of the incidence of PPF determined by decreases in DLCO found a greater proportion of diagnoses met using the GLI equation relative to Crapo & Morris in both cohorts. Conclusions A difference in rates of diagnosis of PPF between the reference equations used for PFTs could potentially exist. Specifically, this data demonstrated that diagnostic rates were greater with GLI Race-Neutral reference equations. However, the study was limited by the distribution of observations across the race categories, which may have reduced the ability to detect differences between race-based and race-neutral equations, suggesting that larger or more evenly distributed samples are necessary to better evaluate the diagnostic differences between the equations. This abstract is funded by: None