Predicted values, Z-scores, LLN, and ATS/ERS severity bands are already documented on the PFT calculator page. This guide is about the judgment calls around those numbers — especially the places U.S. learners and busy clinicians get tripped up.
Clinic vignette: two reports, one patient
A 62-year-old man returns for COPD follow-up. Last year's report used an older adult equation set; this year's lab switched to GLI-2012. His raw FEV1 barely changed, but percent predicted and the “below LLN” flags shifted. The patient asks why he is suddenly “worse on paper.” He isn't — the reference frame moved.
When you compare serial PFTs, confirm that ethnicity group, height, and equation set match. Otherwise you will counsel on noise.
Mistake: trusting a fixed 0.70 ratio for everyone
Fixed cutoffs such as FEV1/FVC < 0.70 are easy to remember and still appear in some pathways. They tend to over-label obstruction in older adults and under-label it in younger ones. GLI-style LLN thinking asks a different question: is this ratio unusual for someone of this age, sex, height, and ancestry? Borderline older patients are where the two approaches disagree most often — and where a quick “0.70 rule” mis-teaches students.
Mistake: wrong ethnicity or stale height
U.S. EHRs and PFT software may label GLI groups as White/Caucasian, African American/Black, Northeast Asian, Southeast Asian, or Other/mixed. Choosing the wrong group — or leaving a default that does not match how the lab codes ancestry — can move Z-scores across the LLN line on borderline tests.
Height errors are quieter but common. Chart height from three years ago, shoes-on measurement, or a half-inch rounding difference changes predicted FEV1 enough to matter when the patient sits near LLN. Use today's standing height for today's interpretation.
Edge case: “restriction” that isn't confirmed yet
Low FVC with a preserved FEV1/FVC often gets casually called restriction. Spirometry can only suggest it. Poor effort, air trapping, and true reduced TLC can look similar on a forced maneuver. Before you lock in a restrictive diagnosis, ask whether lung volumes (TLC) are indicated and whether the blows met quality criteria.
Edge case: the nonspecific pattern
Sometimes FEV1 sits below LLN while both the ratio and FVC do not. That nonspecific pattern frustrates people who want a single label. It is a reminder not to force every report into “obstruction” or “restriction” from spirometry alone — and not to treat one percent-predicted number as a diagnosis.
Quality before cleverness
Acceptability and repeatability failures produce beautiful Z-scores that are wrong. Early termination, cough, leak, or extra breaths can fake obstruction or restriction. If the tech notes are soft, fix the test before debating GLI versus NHANES nostalgia.
What to do next
For transparent educational GLI-2012 predicted values and Z-scores, open the PFT calculator. For patient care, interpret within accredited lab software, clinical context, and professional standards. See also the site disclaimer.