Race-adjusted Lung Function Increases Inequities in Diagnosis and Prognosis and Should Be Abandoned

Evidence before this study We searched MEDLINE and Embase using search terms including “race”, “ethnicity”, “pulmonary function”, “spirometry”, and “prediction equations” from database inception and January 10, 2022, for papers published in English. A total 33 papers related to lung function and race were identified. Race-adjusted lung function reference values were recommended by major guidelines for use internationally. Race-specific references assume a 10-15% lower lung function, such as the forced expired volume in one second (FEV₁) and forced vital capacity (FVC), in black people and 4-6% lower in Asian people compared with in whites. Compared to not adjusting for race, race-adjusted lung function values have recently been questioned as they have been found to not improve prediction of outcomes in population-based studies or in people at risk of obstructive pulmonary disease. Concerns have been raised that, contrary to the intent, race-adjusted reference values may contribute to under diagnosis of disease in disadvantage minorities, with the largest differences reported in black (Afro-American) people, and may worsen race-related health inequalities. Data on the impact of race-adjusted lung function values across the ethnically diverse population are limited and data on how to decrease racial bias in lung function assessment are needed.

Added value of this study We analyzed the impact of using different race-specific (GLI-2012) reference equations for FEV₁ and FVC across the US population in the National Health and Nutrition Examination Survey (NHANES) 2007-2012. Outcomes were prevalence of lung function impairment (value < lower limit of normal), breathlessness on exertion, and mortality up to December 31, 2015. Compared to using references for whites, black reference values were less likely to identify lung function impairment across all races but especially in blacks (9.3% vs. 36.9%); and those identified had lower lung function, more breathlessness, and worse prognosis. Black people with lung function normal by black standards but impaired by white standards had increased prevalence of breathlessness and mortality, compared to those normal also by white standards. Thus, race-adjusted reference values labeled black people as normal despite worse outcomes. White normal values identified people with similarly good lung function, and low rates of breathlessness and mortality across races groups.

Implications of all the available evidence The findings from this study support that race-adjusted reference values markedly under diagnose lung function impairment, and related breathlessness, and mortality in underprivileged groups across the US population. Normal values for whites were most sensitive to identify lung function impairment related to worsening outcomes and people classified as having normal lung function with similar good outcomes irrespective of race group. These findings suggest that lung function should not be adjusted for race. When applied across the population, white reference values were most sensitive to identify smaller or earlier impairment and most specific to identify people with normal lung function with similarly good outcomes across race groups. Given the large impact shown, abandoning the use of race-adjusted lung function values is likely to contribute to improved health equity.