Predictive value of primary care made clinical diagnosis of chronic obstructive pulmonary disease (COPD) with secondary care specialist diagnosis based on spirometry performed in a lung function laboratory

Aim To define the predictive value of clinical diagnosis of chronic obstructive pulmonary disease (COPD) or suspected COPD in primary care patients with spirometric criteria for diagnosis. Background The diagnosis of COPD is usually made clinically but often not confirmed by diagnostic testing. Recent initiatives have called for universal spirometry testing in primary care to diagnose and monitor such patients the implications of this policy on diagnostic accuracy are not as yet known. Methods Retrospective comparative analysis of 677 consecutive primary care referrals to a district general hospital lung function laboratory for spirometry, March 1998 to December 2006. Findings Five hundred and three of 677 patients referred for open access spirometry had a primary care clinical diagnosis or suspected diagnosis of COPD. When compared with NICE spirometric criteria for diagnosis of COPD, 141 patients (28%) had normal spirometry, 46 (9%) had reversible airflow obstruction and 14 (3%) a restrictive pattern of spirometry. The positive predictive value of a primary care clinical diagnosis of COPD was 0.62 for patients referred for assessment of severity and 0.56 for those referred for diagnostic testing. Clinical suspicion of COPD in this sample was not confirmed by spirometry in a high proportion of referred patients. The introduction of the widespread use of spirometry for confirmation of primary care clinician made COPD diagnosis have important implications for both individual patients and primary care service planning.