November 9, 2020
Written by

Nyx Robey

Race in Spirometry (Part II)

A little over a week ago I summarized some of America’s history that disadvantaged marginalized racial groups. Today I’ll dive into how spirometry equations are created, and next week we’ll dive into how history embeds itself into spirometry as well.

What is Spirometry?

Spirometry is a commonly used test for aiding the determination of lung health functioning and various diagnoses related to respiratory health (like asthma).¹ Typically along with spirometry, the patient would fill out a clinical history and may take additional tests depending on their symptoms. It functions by measuring how much air you can blow out of your lungs in a short period of time.

Why are there predicted values in the first place? Doctors usually compare a person’s spirometry results to these expected values for spirometry to get an idea of how they are doing. That is, they compare someone to results for a similar person. However, there are some issues here, like how do we classify similarity across people? How do we define identity for someone else, and is that a place for a predictive formula?

Spirometry is affected by a lot of variable factors, coaching for example is one area that may affect the quality of data that our company is beginning to research. The raw values produced by spirometry, are typically compared to predictive values pre-determined by a formula.

Reference Equations & Predicted Values

These formulas are widely used and accepted, endorsed by national organizations including the American Thoracic Society,² the Center for Disease Control and Prevention,³ in addition to major international respiratory organizations.⁴ With ringing endorsements, it can be a big step to think critically about these values.

However at VitalFlo, we found that many of our own employees were unaware of the history behind how these formulas develop. As a result, we felt it important to make that history better known in our own capacity so that individuals can understand and take charge of their own health and better understand how these formulas may impact them and their diagnoses.

Unearthing the Formulas

Image for post

Image for post

I’m going to rewind a bit to how I stumbled upon this story. Part of my job is to better understand how we can look at variables that affect and predict asthma attacks and emergency room visits so that we can better prevent these occurrences for our patients. As part of compliance with federal organizations, our spirometer must offer predicted values by reference equations.

The most widely used and far-reaching modern reference equation comes from the Global Lung Function Initiative from a 2012 paper,⁵ expanding on previous attempts at references by widening age and ethnic categories.⁶ These formulas are predicated on a few well-established variables as proxies to estimating lung size: age, height, sex, and race or ethnicity. Using statistics, the authors developed weighted formulas based off of the samples used within their papers to understand how these variables relate to one another. The formulas that are more widely used in modern spirometry include:

If you don’t know what the statistical models are, that’s okay. The important aspect to understanding these models is that they directly take those four variables I mentioned earlier and use them in a formula for predicting lung health. As a data scientist, it is my responsibility to question and verify the validity of each variable. For the purpose of this story, we’ll focus on the GLI equations.

The Variables

Age is used to the nearest quarter of a year in the GLI equations with height in centimeters.⁵ These two may be a bit less controversial if taken into consideration with past clinical history or other aspects that may directly impact an individuals lung size through physical development. That being said, recent research reveals body mass index (BMI) may be a better predictor of pulmonary health than height alone with additional particularity for obese patients.⁷ ⁸ ⁹

Sex and race or ethnicity are a little more complicated. Sex in these formulas is only determined as male or female, leaving no option for intersex patients or individuals who undergo sex reassignment surgeries. Race or ethnicity by these formulas is grouped into only four categories and an “other” category:

  1. Caucasian
  2. Black
  3. South-east Asian
  4. North-East Asian

Image for post

This was determined after extensive worldwide spirometry testing. However there are a couple notes, despite individuals from Mexico City and Israel performing better than the “Caucasian” group, they were lumped in to Caucasian as the best performing group and the reference for the equation.⁶ The only ethnicity included in the “Black” group is African Americans. The groups are based on locational nationalities that were available to the researchers. For any individual that does not qualify as Caucasian, Black or originating from east Asia is given an average score across groups without further questions.

Problems with Qualification in Race or Ethnic Identity

While this is the broadest reaching comprehensive study done in regards to ethnicity and the intent of the research is to better encompass racial and ethnic identities beyond the three groups identified by the previous standard in NHANES III (Caucasian, African American, or Mexican American),⁵ the equations do not account for intersectionality in racial or ethnic identity (biracial, multiracial) or varying levels of identity. One study examined how problematic this may be for biracial children by analyzing how that predictive formulas change depending on what race is entered.¹⁰

Additionally it may be possible that individuals identify their racial identity differently in different spaces, or are technically from a certain region by ancestry, but have a more complicated geographic history.

Lastly, spirometry exams may use self-reported racial or ethnic identity (most accurate), but they also may assume racial identity by past clinical exams, or by what the parent reports for children or when they first received their diagnoses. Racial or ethnic identity may be observed by staff or physicians and reported. These formulas were even used by Occupational Safety and Health Association (OSHA) for baseline reporting of respiratory health on the job, where previous standards told the official that if a self-reported value was omitted, it could be taken from their personal record or assumed visually.¹¹

While standards for these processes are better emerging, in terms of more accurate self-report measures and collection,¹² there is also understandable pushback, concern or hesitation from patients (and staff)¹² at times on their use of race in their health decisions.¹³ ¹⁴ ¹⁵ Race and ethnicity may also be more variable than its understood in its use in predictive equations. Given race and racial identity are socially constructed, associating them with genetics and influencing health decisions can be incredibly dangerous, and echo previous efforts to cleave apart individuals into distinct races to justify its use in health (see Part I: History of Race in Medicine).¹⁶ Patients may identify differently depending on the race or ethnic identity of their physician,¹⁴ experiences of bias or racial stereotyping¹⁴ or fear of future discrimination.¹⁷ This is also the case for gender and social class, which melts into the patient-physician communication and impacts other aspects of self-report that are often tied to social determinants of health.¹⁴

How the Respiratory Community Defines Race is Varied

There also may be confusion for patients on how to define race and ethnicity by both patients¹⁸ and the medical community.¹⁹ A comprehensive literature review shows that lung function research does not clearly and consistently define race and ethnicity nor consistently attribute these differences to the same source.¹⁹ The authors examined 226 articles published from 1922 to 2008.They noted that over 90% of articles failed to look at socioeconomic status as an additional proxy, often grouping race into “white” or “other”.¹⁹ While definitions have improved since 2000 (about 70% of articles provided a definition of how they classified race or ethnicity),¹⁹ there is still a lot of room for improvement.

Up until this past summer when I began at VitalFlo, the American Medical Association had little consistency for reporting race and ethnicity beyond capitalization of sociocultural origin in the 11th edition.²⁰ In July, they cite recent events and discussions leading to the final usage of capitalization for racial and ethnic groups like Black and White ²¹ that align with other journal styles like APA.²² AMA style also now requires reporting of classification (patient self-report, observed, etc.) for sex and gender²³ as well as race/ethnicity.²⁴

However these strides are recent and do not exactly shape the history that has already passed and led to the reference equations still in use today developed in the 90s and in 2012. So where did these differences come from to begin with? How did medical history start to put weighted numbers to racial groups given our genetic diversity is at times more dissimilar to members of the same race than to members of other racial or ethnic groups?²⁵ The answer, unsurprisingly, follows similar themes to Part I, the history of race in America, and a social fostering of preference towards lighter skin worldwide.²⁶

In Part I, I mentioned this would be a two-part series but after writing the full length second part I thought I’d break it up into one to two more posts in order to keep them approximately similar in length. Plus, who doesn’t love a good cliffhanger? Hope you join me next week as we hone in on the history of race in spirometry.

P.S. If you’re interested in seeing how some of these formulas play out with your own personal data, you can check out


  1. American Lung Association. (2020, February 19). Spirometry. Retrieved November 03, 2020, from
  2. Culver BH, Graham BL, Coates AL, et al. Recommendations for a Standardized Pulmonary Function Report. An Official American Thoracic Society Technical Statement. Am J Respir Crit Care Med. 2017;196(11):1463–1472. doi:10.1164/rccm.201710–1981ST
  3. Center for Diseases Control and Prevention. NHANES III (1988–1994). National Center for Health Statistics. Published Aust 2020. Accessed November 3, 2020.
  4. Cooper BG, Stocks J, Hall GL, et al. The Global Lung Function Initiative (GLI) Network: bringing the world’s respiratory reference values together. Breathe (Sheff). 2017;13(3):e56-e64. doi:10.1183/20734735.012717
  5. Quanjer, P.H., Stanojevic, S., Cole, T.J., Baur, X., Hall, G.L., Culver, B.H., Enright, P.L., Hankinson, J.L., Ip, M.S., Zheng, J. and Stocks, J., 2012. Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations. doi: 10.1183/09031936.00080312
  6. Quanjer, P. H., Tammeling, G. J., Cotes, J. E., Pedersen, O. F., Peslin, R., & Yernault, J. C. (1993). Lung volumes and forced ventilatory flows. | PFT Adjusted Predicted Values for Men MultiCalc. Accessed November 3, 2020.
  7. Bhatti U, Laghari ZA, Syed BM. Effect of Body Mass Index on respiratory parameters: A cross-sectional analytical Study. Pak J Med Sci. 2019;35(6):1724–1729. doi:10.12669/pjms.35.6.746
  8. Kamal R, Kesavachandran CN, Bihari V, Sathian B, Srivastava AK. Alterations in Lung Functions Based on BMI and Body Fat % Among Obese Indian Population at National Capital Region. Nepal J Epidemiol. 2015;5(2):470–479. doi:10.3126/nje.v5i2.12829
  9. Peralta GP, Marcon A, Carsin A-E, et al. Body mass index and weight change are associated with adult lung function trajectories: the prospective ECRHS study. Thorax. 2020;75(4):313–320. doi:10.1136/thoraxjnl-2019–213880
  10. Pittman JE, Van Rie A, Davis SD. Spirometry in Biracial Children: How Adequate Are Race-Based Reference Equations? Arch Pediatr Adolesc Med. 2011;165(6). doi:10.1001/archpediatrics.2011.76
  11. Occupational Safety and Health Administration. (2013). Spirometry Testing in Occupational Health Programs. Best Practices for Healthcare Professionals. Osha 3637–03 2013. Washington, DC: OSHA.
  12. Baker DW, Cameron KA, Feinglass J, et al. Patients’ attitudes toward health care providers collecting information about their race and ethnicity. J Gen Intern Med. 2005;20(10):895–900. doi:10.1111/j.1525–1497.2005.0195.x
  13. Institute of Medicine (US) Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care; Smedley BD, Stith AY, Nelson AR, editors. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington (DC): National Academies Press (US); 2003. PATIENT-PROVIDER COMMUNICATION: THE EFFECT OF RACE AND ETHNICITY ON PROCESS AND OUTCOMES OF HEALTHCARE. Available from:
  14. Petkovic J, Duench SL, Welch V, Rader T, Jennings A, Forster AJ, Tugwell P. Potential harms associated with routine collection of patient sociodemographic information: A rapid review. Health Expect. 2019 Feb;22(1):114–129. doi: 10.1111/hex.12837. Epub 2018 Oct 19. PMID: 30341795; PMCID: PMC6351414.
  15. Gannon, M. Race Is a Social Construct, Scientists Argue. Scientific American. Published February 5, 2016. Accessed November 3, 2020.
  16. Witzig R. The Medicalization of Race: Scientific Legitimization of a Flawed Social Construct. Ann Intern Med. 1996;125(8):675–679. doi:10.7326/0003–4819–125–8–199610150–00008
  17. Kiran T, Sandhu P, Aratangy T, Devotta K, Lofters A, Pinto AD. Patient perspectives on routinely being asked about their race and ethnicity: Qualitative study in primary care. Can Fam Physician. 2019;65(8):e363-e369.
  18. Baker DW, Hasnain-Wynia R, Kandula NR, Thompson JA, Brown ER. Attitudes toward health care providers, collecting information about patients’ race, ethnicity, and language. Med Care. 2007 Nov;45(11):1034–42. doi: 10.1097/MLR.0b013e318127148f. PMID: 18049343.
  19. Braun L, Wolfgang M, Dickersin K. Defining race/ethnicity and explaining difference in research studies on lung function. Eur Respir J. 2013;41(6):1362–1370. doi:10.1183/09031936.00091612
  20. Gregoline B, The JAMA Network Editors. Capitalization: Sociocultural Designations. In: AMA Manual of Style: A Guide for Authors and Editors. 11th ed. AMA Manual of Style. Oxford University Press; 2020:1256.
  21. AMA Manual of Style Committee. Updates to the Manual. AMA Manual of Style 11th Edition: A Guide for Authors and Editors. Published July 3, 2020. Accessed November 3, 2020.;jsessionid=E9507D8DB95536A954402965CDF4E474
  22. Racial and Ethnic Identity. Accessed November 3, 2020.
  23. Clayton JA, Tannenbaum C. Reporting Sex, Gender, or Both in Clinical Research? JAMA. 2016;316(18):1863. doi:10.1001/jama.2016.16405
  24. Journal of American Medical Association (JAMA). (2020, October 28). Instructions for Authors. Retrieved November 03, 2020, from
  25. Witherspoon DJ, Wooding S, Rogers AR, et al. Genetic Similarities Within and Between Human Populations. Genetics. 2007;176(1):351–359. doi:10.1534/genetics.106.067355
  26. Glenn EN. Shades of Difference: Why Skin Color Matters. Stanford University Press; 2009.
  27. Quanjer, P. H., Tammeling, G. J., Cotes, J. E., Pedersen, O. F., Peslin, R., & Yernault, J. C. (1993). Lung volumes and forced ventilatory flows. | PFT Adjusted Predicted Values for Men MultiCalc. Accessed November 3, 2020.
  28. National Institute for Occupational Safety and Health. CDC — Spirometry — Spirometry Training Program: NHANES III Reference Values — NIOSH Workplace Safety and Health Topic. Center for Disease Control and Prevention. Published December 22, 2011. Accessed November 6, 2020.
  29. European Respiratory Society. ERS — Online Calculator. ERS. Accessed November 6, 2020.

Written by

Nyx Robey