A comparison of spirometry in general practice and a pulmonary function laboratory

Abstract

Aims: : To compare the results of spirometry testing in primary care with those obtained at a pulmonary function laboratory and to explore whether differences were due to technique or equipment.

Methods: : Patients on the waiting list for spirometry in six participating practices had the test performed in their own practice and at the pulmonary function laboratory (PFL).

Results: : A total of 45 patients had spirometry performed at both locations. Practice nurses underestimated FEV(1) and FVC. The mean difference in FEV(1) was 0.109litres (6.69%, 95% CI 2.88, 9.51) compared with a bellows spirometer, and 0.07litres (6.2% 95% CI 0.89, 8.25) when the same type of spirometer was used. The mean difference in FVC was 0.413litres (15.0% 95% CI 9.3, 20.6) when compared with bellows, and 0.267litres (10.2% 95% CI 4.1, 16.2) when the same type of spirometer was used. All differences were statistically significant (p<0.05, paired t test). Agreement on categorization of COPD was moderate (Kappa 0.46) with practice nurses overestimating severity. Compared to PFL categorisation for the presence or absence of COPD using bellows spirometers, the sensitivity of practice nurse spirometry was 0.93 (95% CI: 0.76, 0.99) and specificity 0.65 (95% CI: 0.38, 0.86).

Conclusion: : Spirometry results obtained by practice nurses were lower than those obtained in a PFL, leading to over-diagnosis of COPD severity.