Reference equations are the backbone of pulmonary function testing (PFT). When a patient blows into a spirometer, the raw numbers mean very little without a reliable comparison point. Customizable normal values libraries give clinicians the ability to select the most appropriate reference population for each patient, directly improving diagnostic accuracy, reducing misclassification, and ultimately supporting better patient outcomes.

TL;DR

• Normal values libraries define what "healthy" looks like for a given patient population; choosing the wrong one introduces systematic error.

• GLI and NHANES III reference equations are the two most widely used standards in PFT, each with distinct strengths and appropriate use cases.

• Research consistently links diagnostic accuracy to patient safety and treatment efficacy.

• Customizable libraries allow labs to switch between equation sets without manual recalculation, reducing clinical risk.

• Rezibase includes a pre-configured, regularly updated normal values library designed to meet evolving ATS standards.

What Are Normal Values Libraries in Pulmonary Function Testing?

A normal values library is a structured database of reference equations used to calculate predicted lung function values for a patient based on demographic variables such as age, height, sex, and ethnicity. These predicted values establish the threshold between "normal" and "abnormal" lung function.

Without the right reference equation set, a result that appears normal may actually indicate disease, or a healthy patient may be incorrectly flagged. The choice of equation set is not a minor administrative decision; it is a clinical one.

Key reference equation sets used in respiratory labs include:

Why Does the Choice of Reference Equation Matter Clinically?

The reference equation set a lab uses directly determines the rate of false positives and false negatives in PFT interpretation. A 2025 systematic review published in the Journal of Multidisciplinary Healthcare (Alharbi, 2025) found that accurate diagnosis improves treatment efficacy, enhances patient safety, and reduces unnecessary procedures. The inverse is equally true: inaccurate diagnostic classification leads to under-treatment, over-investigation, and avoidable harm.

Misclassification in spirometry is a well-documented risk when:

• An ethnicity-specific equation is applied to a patient from a different population

• An outdated equation set is used for a contemporary patient cohort

• A lab applies a single fixed equation universally without clinical review

This is precisely why the American Thoracic Society (ATS) and the Global Lung Initiative recommend that labs maintain the flexibility to select and update equation sets based on their patient population.

What Is the Difference Between GLI and NHANES III Reference Equations?

GLI-2012 was developed using data from over 74,000 healthy non-smokers across multiple countries and ethnicities. It uses continuous age-based equations (LMS method) that eliminate the artificial discontinuities seen in older equation sets at age transitions. GLI is now the preferred standard recommended by the ATS for most spirometry applications.

NHANES III reference equations were derived from the Third National Health and Nutrition Examination Survey, conducted in the United States between 1988 and 1994. They remain widely used, particularly in North American clinical settings, and are still embedded in many legacy spirometry devices. NHANES III provides separate equations for White, African American, and Mexican American populations.

Key practical differences:

• Age range: GLI covers 3 to 95 years continuously; NHANES III has separate adult and pediatric equations with transition points.

• Ethnic coverage: GLI includes more ethnic groups; NHANES III covers three US-defined racial groups.

• Adoption trend: GLI is increasingly preferred in new implementations; NHANES III remains relevant in labs with legacy workflows or specific population needs.

Neither is universally superior. The right choice depends on the patient population a lab serves and the clinical context.

How Do Data Quality Issues Affect PFT Interpretation?

Even with the correct reference equation, poor data quality undermines diagnostic accuracy. A 2025 study published in Frontiers in Artificial Intelligence (Jarmakovica et al., 2025) highlighted that missing values and anomalies in healthcare datasets can result in misdiagnoses and inefficient resource use. In the context of PFT, this translates to incomplete demographic data (such as missing height or incorrect age) producing unreliable predicted values.

A 2022 meta-research study in npj Digital Medicine (Jayakumar et al., 2022) also noted the importance of consistent quality assessment in diagnostic accuracy studies, pointing out that inconsistent application of standards undermines the reliability of results across settings.

Practical steps to protect data quality in PFT reporting:

• Verify patient demographics at point of data entry, not at reporting

• Use systems that flag missing or implausible inputs before calculations run

• Maintain audit trails for equation set selection to support accreditation requirements

What Makes a Normal Values Library "Customizable" and Why Does It Matter?

A customizable normal values library allows a lab to:

• Select different equation sets for different patient subgroups

• Update to new equation sets as clinical guidelines evolve (without vendor dependency)

• Apply site-specific configurations that reflect the local patient population

This flexibility is not a luxury; it is a quality control requirement. A paper published via Academia.edu on designing transferable diagnostic accuracy studies emphasized that estimates of test accuracy must be designed with transferability in mind, accounting for differences in patient populations across settings. A one-size-fits-all equation set directly contradicts this principle.

Labs that cannot update their reference equations without a software vendor intervention face a structural clinical risk. This is one of the core problems that platforms like Rezibase were built to solve. Rezibase includes a pre-configured and regularly updated normal values library, allowing respiratory scientists to apply the correct equation set for each patient without manual recalculation or external intervention.

How Should Labs Approach Transitioning Between Reference Equation Sets?

Switching from one equation set to another, for example moving from NHANES III to GLI, is a straightforward process when the right system is in place. The core steps are:

1. Audit current usage: Identify which equation sets are currently active and which patient groups they apply to.

2. Review clinical guidelines: Confirm ATS or local society recommendations for your patient population.

3. Configure the new equation set: In a flexible platform, this is a configuration change, not a data migration project.

4. Document the change: Update your quality management records to reflect the transition date and rationale.

5. Communicate with reporting clinicians: Ensure medical staff understand that predicted values may shift slightly for some patients.

The transition itself should not be a source of anxiety. With the right software, it is a controlled, auditable configuration update.

Frequently Asked Questions

What is the most widely recommended reference equation set for spirometry in 2026?
The GLI-2012 equations are currently recommended by the ATS and most international respiratory societies for spirometry. NHANES III reference equations remain in use in many North American labs and legacy systems.

Can a lab use more than one equation set?
Yes. Many labs apply different equation sets for pediatric versus adult patients, or for specific ethnic groups where one set is better validated.

Does switching reference equations affect historical patient data?
Historical raw data (measured values) is unaffected. Predicted values and percent-predicted calculations may change if recalculated under a new equation set.

How often are reference equation sets updated?
Major updates are infrequent but significant. GLI has released updates for specific tests beyond spirometry. Labs should monitor ATS and ERS publications for guidance.

What accreditation standards govern normal values selection?
TSANZ/NATA standards in Australia and ISO 15189 requirements both address the need for documented, validated reference ranges. Labs must be able to demonstrate their equation set selection is appropriate and current.

Is the NHANES III dataset still considered valid?
NHANES III reference equations remain statistically valid for the populations they were derived from. Their limitation is coverage, not validity; they do not cover all ethnic groups represented in modern clinical populations.

What happens if the wrong equation set is used?
Using an inappropriate equation set can result in systematic misclassification of patients as obstructed, restricted, or normal when they are not. This directly affects treatment decisions and patient safety.

About Rezibase

Rezibase is Australia's most advanced cloud-based respiratory and sleep reporting platform, built by respiratory scientists for respiratory scientists. It includes a pre-configured, regularly updated normal values library, vendor-neutral device integration, and full accreditation support for TSANZ/NATA and ISO 15189 standards. Rezibase is trusted by over 35 sites across Australia and the UK, including NSW Health and the NHS.

If your lab is ready to move beyond rigid, vendor-locked equation sets and adopt a flexible, clinician-configured normal values library, explore what Rezibase offers at rezibase.com.

References

• Jayakumar, S. et al. Quality assessment standards in artificial intelligence diagnostic accuracy systematic reviews: a meta-research study. npj Digital Medicine, 2022. https://www.nature.com/articles/s41746-021-00544-y

• Alharbi, T.A.F. The Impact of Accurate Diagnosis on Patient Safety: A Systematic Review. Journal of Multidisciplinary Healthcare, 2025. https://www.dovepress.com/diagnostic-challenges-and-patient-safety-the-critical-role-of-accuracy-peer-reviewed-fulltext-article-JMDH

• Jarmakovica, A. et al. Machine learning-based strategies for improving healthcare data quality: an evaluation of accuracy, completeness, and reusability. Frontiers in Artificial Intelligence, 2025. https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1621514/full

• Designing studies to ensure that estimates of test accuracy are transferable. Academia.edu. https://www.academia.edu/13509246/Designing_studies_to_ensure_that_estimates_of_test_accuracy_are_transferable