Article Title

A mixed methods study to compare models of spirometry delivery in primary care for patients at risk of COPD


Background: To increase recognition of airflow obstruction in primary care, we compared two models of spirometry delivery in a target group at risk of chronic obstructive pulmonary disease (COPD).

Methods: A 6 month qualitative/quantitative cluster randomised study in eight practices compared opportunistic spirometry by “visiting trained nurses” (TN) with optimised “usual care” (UC) from general practitioners (GPs) for smokers and ex-smokers, aged over 35 years. Outcomes were: spirometry uptake and quality, new diagnoses of COPD and GPs’ experiences of spirometry.

Results: In the eligible target population, 531/904 (59%) patients underwent spirometry in the TN model and 87/1130 (8%) patients in the UC model (p<0.0001). ATS spirometry standards for acceptability and reproducibility were met by 76% and 44% of tests in the TN and UC models, respectively (p<0.0001). 125 (24%) patients tested with the TN model and 38 (44%) with the UC model reported a pre-existing respiratory diagnosis (p<0.0001). Three months after spirometry, when the ratio of forced expiratory volume in 1 s/forced vital capacity (FEV1/FVC) was <0.7 and no prior COPD diagnosis was reported, nine (8%) participants had a new doctor recorded COPD diagnosis in practices with the TN model and two (8%) participants in practices with the UC model. Mislabelling of participants with a diagnosis of COPD when FEV1/FVC was ≥0.7 was present in both models prior to and after spirometry. GPs valued high quality spirometry and increased testing of patients at risk of COPD in the TN model. They identified limitations, including the need for better systematic follow-up of abnormal spirometry and support with interpretation, which may explain persisting underdiagnosis of COPD in practice records.

Conclusions: Although opportunistic testing by visiting trained nurses substantially increased and improved spirometry performance compared with usual care, translating increased detection of airflow obstruction into diagnosis of COPD requires further development of the model.




Link to Publisher Version (DOI)