Diagnostic adequacy of COPD in primary care: a population-based analysis of spirometry use and risk-factor documentation.

Abstract

Background: Diagnosing COPD in primary care remains challenging, with significant international evidence of underdiagnosis, misdiagnosis and miscoding. How reliably COPD diagnoses in real-world practice reflect objective diagnostic criteria is less well understood.

Aim: To evaluate the compatibility between recorded COPD diagnoses and key diagnostic criteria - spirometry and risk-factor documentation - and to identify patient and contextual factors associated with diagnostic adequacy.

Methods: A population-based, cross-sectional study including all patients with a recorded diagnosis of COPD across a large Portuguese primary care cluster (17 Family Health Units; 182,947 patients). Electronic Health Records data were extracted for demographics, comorbidities, symptoms, exposures, spirometry and clinical follow-up. Spirometry and risk-factor records were independently classified according to their compatibility with a diagnosis of COPD (compatible, uncertain or not compatible) and combined to derive diagnostic-certainty categories. Associations were analysed using non-parametric tests and multinomial logistic regression.

Results: Among 1817 included patients, only 3.63% had both spirometry and exposure records fully compatible with COPD at diagnosis; 24.05% had incompatible records. Using the most recent spirometry, compatibility improved only marginally (4.4%), while incompatible cases remained high (21.57%). Spirometry was absent or non-confirmatory in nearly 80% of patients, and smoking or other exposures documentation was missing or incomplete in almost 50%. Diagnostic incompatibility was strongly associated with older age, female sex, diagnosis made outside primary care, multimorbidity, and especially psychological comorbidities (OR 11.6; 95% CI 7.7-17.4). Asthma was the only respiratory comorbidity significantly associated with incompatible diagnoses. Records of symptoms, exacerbations and GOLD classification were infrequently documented.

Conclusions: COPD diagnoses recorded in routine primary care frequently lack spirometric confirmation and adequate documentation of risk factors, resulting in high rates of incompatible diagnoses. Diagnostic adequacy reflects demographic, social and clinical determinants, revealing important inequities. These findings highlight an urgent need to strengthen diagnostic pathways through systematic use of spirometry, structured recording of risk factors and symptoms, and enhanced electronic health records documentation. Improving these core elements is essential to ensure accurate, consistent and equitable COPD diagnosis in primary care.