Abstract
Introduction: Over diagnosis is a growing concern for a wide range of diseases. We aimed to assess the magnitude of COPD over-diagnosis in the UK primary care settings and examine the characteristics of patients potentially over-diagnosed with COPD. Methods: We analysed data on 1,473 GP diagnosed COPD patients aged 40 years and over who participated in the Birmingham COPD cohort study-UK. Patients were classified as non-COPD or confirmed-COPD based on post-bronchodilator spirometry results. Characteristics were compared using logistic regression adjusted for age, sex and smoking status. Results: Based on GOLD, LLN (GLI-2012 equations) and NICE 2004 definitions, 13.7%, 28.1% and 32.3% of participants were potentially over-diagnosed with COPD respectively. Restrictive pattern of lung disease was observed in 18.9% of non-COPD. Compared to confirmed-COPD, non-COPD were younger (mean age 67.2 versus 69.5 years, OR 0.98; 95% CI 0.96 - 0.99), more likely to be female (52.2% versus 35.4%, OR 0.5; CI 0.4 - 0.7), never smokers (22.9% versus 13.8%, OR 0.5; CI 0.4 - 0.7), obese (39.3% versus 31%, OR 1.6; CI 1.04 - 2.5), with multiple co-morbidities (23.9% versus 16.4%, OR 1.7; CI 1.1 - 2.8) and showed less FEV1 reversibility (10% versus 21.4%, OR 0.4; CI 0.2 - 0.7). Non-COPD participants were more likely to report previous asthma (47.3% versus 38.7%), coronary heart diseases (18.4% versus 14.2%), diabetes (19% versus 14%) and depression (21.4% versus 16.8%). But the difference between groups was insignificant. Conclusion: Over diagnosis was common. We identified female sex, obesity, restrictive lung pattern and multiple comorbidities as potential predictors for COPD over-diagnosis but recommend a follow-up assessment for spirometry variability.