A new spirometry-based algorithm to predict occupational pulmonary restrictive impairment.

Area of Science:

• Occupational Medicine
• Pulmonary Medicine
• Diagnostic Accuracy

Background:

• Spirometry is a common tool in workplace respiratory surveillance.
• Current spirometry algorithms have poor performance in identifying restrictive lung disease, especially at low prevalence.
• This leads to a high proportion of false positives and unnecessary referrals.

Purpose of the Study:

• To enhance the specificity and positive predictive value of spirometry algorithms for restrictive pulmonary impairment.
• To reduce false positive findings and subsequent unnecessary lung volume testing in occupational settings.

Main Methods:

• Re-analysis of two studies involving 376 patients to derive and validate a new spirometry algorithm.
• Utilized true restrictive lung cases as a reference standard.
• Compared the new algorithm against existing ones using receiver operating characteristic curves in a simulated working population (prevalence 1-10%).

Main Results:

• The proposed algorithm (FVC < 70% predicted and FEV1/FVC ≥ 70%) achieved 96% specificity and 71% sensitivity.
• Positive predictive values were 67% (10% prevalence) and 15% (1% prevalence).
• Resulted in only 4% false positives and correctly classified 91% of restrictive cases.

Conclusions:

• A novel spirometry-based algorithm demonstrates high accuracy for excluding pulmonary restriction.
• This algorithm has the potential to decrease unnecessary lung volume testing in occupational health.
• It offers a more reliable method for screening restrictive lung conditions in the workplace.