Industry research has highlighted the potential of racial bias in the interpretation of spirometry and other PFT that can lead to delayed care (or lack of care) in Black patients.

By Bill Pruitt, MBA, RRT, CPFT, FAARC

Industry research has highlighted the potential of racial bias in the interpretation of spirometry and other pulmonary function testing that can lead to delayed care (or lack of care) in Black patients. As a result, the ATS recently recommended use of average reference equations rather than those based on race or ethnicity for the assessment of pulmonary function in clinical and laboratory practice. This article will explore research on these racial biases and the proposed solutions to eliminate them from pulmonary function test (PFT) interpretation and improve health equity.

Historical Background and Implications of Racial Biases

Pulmonary function testing is a useful tool in obtaining objective measurements that help diagnose obstructive and restrictive lung diseases {such as COPD, asthma, cystic fibrosis, and pulmonary fibrosis). PFTs are used in evaluating workers in certain occupations (ie, fire fighters, commercial underwater divers, military service) or when looking for issues with certain occupations that may be harmful to the lungs (ie, exposure to coal dust, asbestos, cotton dust, etc).1-2

PFTs are also used other areas such as performing an evaluation to qualify for life insurance, determining need for noninvasive ventilatory support for neuromuscular weakness, qualifying for lung transplantation for interstitial lung disease, or determining a patient’s fitness for surgical lung cancer resection.2 

In the interpretation of PFT results, the measurements are compared with reference values obtained from testing a sample of healthy individuals. From a historical perspective, the reference tests in the 1960s were obtained from primarily white men who were free of any lung disease (although in early testing, tobacco smokers were included). Predicted values were based on height and age. Over time, the reference population included white women and men but did not include tobacco users.1

As interpretation results were retrospectively evaluated and analyzed, what appeared to be race-based differences had become apparent (ie, the results in Black subjects showed lower lung volumes). Interpretation of the results based on the white reference population led to such issues as delayed, inappropriate, or no treatment, denial of access to jobs, not qualifying for insurance, etc. To attempt to address this, a scaling factor was incorporated to adjust the predicted values for Black subjects. For example, in 1978 the US Occupational Safety and Health Administration (OSHA) began using a factor of 0.85 to adjust the predicted values established from white reference populations when testing Black job applicants.3

The Third National Health and Nutrition Examination Survey (NHANES III)

In the late 1990s the Third National Health and Nutrition Examination Survey (NHANES III) was performed which tested 6,984 individuals, including Black, white, and Mexican American men and women, and the results were used to produce race-specific reference equations.1 The NHANES III reference set for predicted values was recommended in 2005 by the American Thoracic Society (ATS) and European Respiratory Society (ERS). (Some years later, as racial biases continue to be problematic, the ATS/ERS recognized that, “The lack of standards for how to derive and use PFT reference equations has led to considerable confusion in the interpretation of PFT results.”)4

The Global Lung Function Initiative (GLI)

The next major step in developing reference equations took place in 2012 with the Global Lung Function Initiative (GLI), which offered a wider range of ages and produced predicted values for four ethnic categories: Caucasian, African American, Southeast Asian, and Northeast Asian. (The NHANES III provided reference equations covering 8 to 80 years of age while GLI (2012) provided coverage from 3 to 95 years of age).4 The GLI (2012) testing included 74,185 individuals to create a much broader reference population.5 There was data gathered for people of other racial/ethnic groups but the data was not statistically robust enough to create any other categories beyond the four mentioned above. 

One issue with the GLI (2012) reference equations was there was no explanation concerning how race and ethnicity were defined and no reliability statistics for the four categories were reported. Another issue was in testing where the subject did not identify with any of the four ethnic groups or identified as a mix between two or more ethnic groups. To address this, a fifth category was included, called GLI-Other, which used a blend of results from the other ethnic categories. However, this still was problematic due to a heavier weight toward the GLI-Caucasian values in the GLI-Other derivatives.1 

In 2022, the ATS and ERS updated the technical standards for strategies to use in interpreting PFTs and recommended using the GLI (2012) for the new reference equations. However, in this update from 2022, the ATS/ERS recognized that there was an “uncomplete understanding of the role of race/ethnicity on lung function.”4 They also recommended that if the ethnic/racial background was unknown or uncertain, the GLI-Other reference equations should be used and that any PFT reports and research publications include the details as to what reference equation was used. 

Depending on which ethnic category is used when producing an interpretation, the outcome of the interpretation can change. For example, in a test of a 40-year-old man who is six feet tall, assume that the measured FEV1 is 3.50 L. If the GLI Caucasian category is selected as a reference set, the predicted FEV1 will be 4.50 L and the 5th percentile threshold is 3.56 L. Given these numbers, the results show the FEV1 to be 78% predicted and falls below the fifth percentile. In contrast, if the GLI African American category is selected, the predicted FEV1 will be 3.84 L and the fifth percentile threshold is 2.95 L. Given these numbers, the results show the FEV1 to be 91% predicted and is above the fifth percentile.2

Other factors were mentioned in the ATS/ERS 2022 update that can affect the results of PFT testing. These factors included many determinants of health, including prematurity, maternal smoking, early-life exposure to second-hand smoke, poor air quality, lower respiratory tract infections, undernutrition, obesity, and lower socioeconomic status (SES).2 Also, assigning ethnicity is a challenge both in the fact that the individual being tested self-reports their ethnicity and the ethnicity may be unknown or may be a mixture of many ethnicities.4 

The GLI Global (2022)

More recently, a new reference set was created to address issues with testing subjects of uncertain race or ethnicity. This new reference set, called GLI Global (2022), was based on the testing performed in the original GLI (2012). GLI Global gives equal weight to the blend of the four categories from GLI (2012) and is described as a race-neutral approach to the issue of PFT interpretation.5 The term race-neutral points to the prediction equations that do not require a selection of race for the purpose of performing/interpreting the test.2 This newest reference set has been suggested to be “a more valid representation of the composite GLI data.”1 

ATS Statement Establishes New Path

After review and discussion of the published research concerning bias related to race and ethnicity in the interpretation of PFT results (limited at this point to spirometry and not to DLCO and lung volume measurements), the ATS recommended in 2023 that race-neutral reference sets be used for spirometry.2 They have also recommended discontinuing the use of adjustment factors for race and ethnicity in interpreting lung volumes and DLCO. The recommendations went on to say that the GLI Global reference set should be used for interpreting results.2 The ATS statement states that a “single set of reference equations better matches the relationship between pulmonary function and survival and incident chronic lung disease between Black and white persons.”2 

The ATS stated that there are important limitations that will need further research and evaluation to resolve. One is that there are many of the world’s population that have not been included in the weighted average of the data, as these groups have not been tested. The race-neutral equation is a composite of findings from selected racial and ethnic groups and there are many groups that need to be included to develop a more complete composite. Another limitation is related to the results that are at or near the lower limit of normal (LLN). The ATS recommends that in these cases the interpretation should be done cautiously and include consideration of the subject’s history, and other test results, imaging findings, or diagnostic results that are appropriate.2 

The ATS statement recognized and discussed the impact of changing to the race-neutral GLI Global reference equation. There is a possibility that this change may bring about an increase in persons of color being found unfit for occupations such as firefighting or commercial diving and may also cause an increase in premiums for life and health insurance.2 Another concern relates to people of color who are found to be near the lower PFT thresholds under the new reference equations in qualifying for surgical resection of lung cancer. Finally, the ATS notes that changing to the new reference equation shifts the LLN which results in an increase in people of color having values below the LLN and will decrease the number of white people being below the LLN. 

In a situation where restrictive disease is a possibility and the FVC is less than the LLN with a normal FEV1/FVC, more people of color might have further testing performed to evaluate the possible restrictive disease while white people would not be tested further.2 On the other hand, when considering the criteria used to qualify for lung transplantation, the new GLI Global reference equations would change percent predicted results in Black persons to lower numbers and would likely cause a referral for transplantation that would not have occurred with the previous race-specific equations.2 Also, under the new GLI Global reference equations, the may be an increase in Black persons qualifying for disability.

Education Is Needed 

The ATS statement points out that education is needed for the medical community, for those outside the medical community who utilize PFT results in their arena, and for patients. One issue is that assumptions concerning the effect of race and ethnicity on PFT findings have been taken as fact. For example, the idea that race can be based on appearance needs to be addressed in educating the medical community.2 Education needs to establish that a lower socioeconomic status may contribute to what is perceived as racial differences regarding lung growth and functional decline. Those entities that deal with work eligibility, qualification for disability, insurance coverage, and those who recommend standards for safety and hiring need to understand the changes that will come with the new reference equations.2 Education needs to emphasize that using strict thresholds for PFT results in making decisions needs to change, and that other means of assessing health and disability should be included in the evaluation. Patients need to know that the new reference equations may change the interpretation of their PFT and their status, and why this is a necessary change. 

Conclusion

The change to a race-neutral approach in using results from testing reference populations to establish predicted values is a major change in pulmonary function testing and interpretation. There has been a long history of racial/ethnic biases that is imbedded in the medical community due to the previous data collection (predominately white populations) and interpretation strategies that use strict thresholds for abnormal results and do not include consideration of health status and other factors such as socioeconomic status, geography, or exposure to harmful substances. This shift requires education and understanding, coupled with an increased investigation into the other factors that can affect PFT results.

However, continued collection of racial/ethnic data is needed to perform more research and evaluation of the impact of this change to the GLI Global reference sets. Continued and expanded testing of more groups of people are needed to provide more extensive, inclusive data so that the race-neutral reference equations can be more meaningful. This shift changes the PFT world, hopefully for the better.

RT 

Bill Pruitt, MBA, RRT, CPFT, FAARC, is a writer, lecturer, and consultant. Bill has over 40 years of experience in respiratory care in a wide variety of settings and has over 20 years teaching at the University of South Alabama in Cardiorespiratory Care. Now retired from teaching, Bill continues to provide guest lectures, participates in podcasts, and writes professionally. For more info, contact [email protected].


References

  1. Marciniuk DD, Becker EA, Kaminsky DA, McCormack MC, Stanojevic S, et. al. Effect of Race and Ethnicity on Pulmonary Function Testing Interpretation. Chest. 2023;164(2):461-75.
  2. Bhakta NR, Bime C, Kaminsky DA, McCormack MC, Thakur N, et. al. Race and ethnicity in pulmonary function test interpretation: an official American Thoracic Society statement. American journal of respiratory and critical care medicine. 2023 Apr 15;207(8):978-95.
  3. United States Department of Labor. Occupational exposure to cotton dust—final mandatory occupational safety and health standards. 1978. 43(122):27350-27463. United States  Government Printing Office website. https://www.govinfo.gov/content/pkg/FR-1978-06-23/pdf/FR-1978-06-23.pdf
  4. Stanojevic S, Kaminsky DA, Miller MR, Thompson B, Aliverti A, Barjaktarevic I, Cooper BG, Culver B, Derom E, Hall GL, Hallstrand TS. ERS/ATS technical standard on interpretive strategies for routine lung function tests. European Respiratory Journal. 2022 Jul 1;60(1).
  5. Bowerman C, Bhakta NR, Brazzale D, Cooper BR, Cooper J, Gochicoa-Rangel L, et.al. A race-neutral approach to the interpretation of lung function measurements. American journal of respiratory and critical care medicine. 2023 Mar 15;207(6):768-74.


Additional Reading

  • Ganguli I, Mackwood MB, Yang CW, Crawford M, Mulligan KL, O’Malley AJ, Fisher ES, Morden NE. Racial differences in low value care among older adult Medicare patients in US health systems: retrospective cohort study. BMJ. 2023 Oct 25;383.
  • Moffett AT, Bowerman C, Stanojevic S, Eneanya ND, Halpern SD, Weissman GE. Global, race-neutral reference equations and pulmonary function test interpretation. JAMA Network Open. 2023 Jun 1;6(6):e2316174-.
  • Sjoding MW, Ansari S, Valley TS. Origins of racial and ethnic bias in pulmonary technologies. Annual review of medicine. 2023 Jan 27;74:401-12.
  • JonesJ DS, Hammonds E, Gone JP, Williams D. Explaining Health Inequities—The Enduring Legacy of Historical Biases. New England Journal of Medicine. 2024 Feb 1;390(5):389-95.
  • Omiye JA, Lester JC, Spichak S, Rotemberg V, Daneshjou R. Large language models propagate race-based medicine. NPJ Digital Medicine. 2023 Oct 20;6(1):195.
  • Kanj AN, Scanlon PD, Yadav H, Smith WT, Herzog TL, Bungum A, et. al. Application of Global Lung Function Initiative global spirometry reference equations across a large, multicenter pulmonary function lab population. American Journal of Respiratory and Critical Care Medicine. 2024 Jan 1;209(1):83-90.