Interpreting lung function data using 80% predicted and fixed thresholds misclassifies more than 20% of patients.

Area of Science:

• Pulmonary Medicine
• Diagnostic Accuracy
• Respiratory Physiology

Background:

• Current Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines use fixed thresholds (FEV1/FVC < 0.70 and FEV1 80% predicted) for diagnosing COPD.
• Many laboratories rely on 80% predicted values to determine abnormal lung function test results.
• Previous studies have not explored misclassification rates across a full spectrum of lung diseases using GOLD criteria versus the lower limit of normal (LLN).

Purpose of the Study:

• To compare the diagnostic accuracy of the GOLD guidelines (GOLD/PP) with the lower limit of normal (LLN) method for detecting airway obstruction and classifying COPD severity.
• To quantify misclassification rates in pulmonary function test (PFT) interpretation between the two methods.

Main Methods:

• Analysis of prebronchodilator pulmonary function test results from 11,413 adult patients in the UK, New Zealand, and the US.
• Comparison of classification outcomes using GOLD/PP criteria versus LLN (fifth percentile) thresholds.

Main Results:

• The GOLD/PP method misclassified 24% of patients.
• Ten percent of patients with normal lung function were falsely diagnosed, and 7% were falsely attributed with emphysema.
• GOLD/PP criteria showed significant bias, over-diagnosing airflow obstruction and restrictive defects in men and older individuals, while under-diagnosing them in younger patients.

Conclusions:

• Fixed thresholds (80% predicted) in PFT interpretation can lead to over 20% misdiagnosis rates.
• The GOLD/PP method introduces significant biases affecting disease status assessment and treatment allocation.
• Utilizing the LLN method based on fifth-percentile values effectively avoids these misclassifications and improves diagnostic accuracy.