Combining case-finding methods for COPD in primary care: a large, two-stage design study.

Abstract

Underdiagnosis of chronic obstructive pulmonary disease (COPD) is common. We aimed to assess the effectiveness of using the Chronic Obstructive Pulmonary Disease-Population Screener (COPD-PS) questionnaire with pre-bronchodilator (BD) peak expiratory flow (PEF) measurements as a case-finding strategy for COPD in primary care. This was a two-stage, cross-sectional study comprising a population survey in a primary care population aged 35 years without previous COPD, followed by a validation study using COPD-PS 4 or PEF 2.2 l/s·m2, and confirmed by spirometry (post-BD forced expiratory volume in 1 s/forced vital capacity [FEV1/FVC] <0.70). The predictive capacity of the strategy was assessed in a case-control sub-study. Of a total of 10 071 individuals, 6969 (69.2%) participants were included. Both tests were positive in 4.3% subjects, PEF only in 2.1% and COPD-PS only in 5.1%. Of the 802 with positive screening results, COPD was confirmed using spirometry in 130, accounting for 1.9% of all participants and 16.2% of those who tested positive on COPD-PS or PEF. Among the 130 true-positives, the mean score for the COPD-PS questionnaire was 5.1 l/s·m2 (± standard deviation [SD] 1.7) and 1.9 l/s·m2 (±SD 0.8) for pre-BD PEF, both significantly worse than in the 672 false-positives. The combined use of both screening tests had a sensitivity of 67.5%, a specificity of 71.3% and a diagnostic accuracy of 69.6%. Case finding for COPD using COPD-PS + PEF led to a 90% reduction in the number of spirometry tests performed.