S150 GOLD VERSUS NICE DIAGNOSTIC CRITERIA FOR CHRONIC OBSTRUCTIVE PULMONARY DISEASE: IMPACT ON DISEASE PREVALENCE AND MORTALITY RISK WITHIN THE RENFREW/PAISLEY STUDY

¹HJ Starkie, ¹AH Briggs, ¹CL Hart, ¹M Gillies, ¹K MacIntyre, ²MC Shepherd. ¹Section of Public Health and Health Policy, University of Glasgow, Glasgow, UK, ²Division of Immunology, Infection and Inflammation, University of Glasgow, Glasgow, UK

Objective: To compare chronic obstructive pulmonary disease (COPD) prevalence and mortality risk in a single population using alternative diagnostic criteria.

Methods: The global initiative for chronic obstructive lung disease (GOLD) is based on lung function alone, whereas the UK National Institute for Health and Clinical Excellence (NICE) criteria include lung function, respiratory symptoms and other risk factors. These criteria were applied to a Scottish prospective cohort (Renfrew/Paisley (Midspan) study) of 15 402 men and women aged 45–64 years at baseline and followed for over 30 years. All-cause and COPD mortality were modelled using Cox regression analysis.

Results: Overall COPD prevalence for men (women) was 31% (20%) applying GOLD criteria compared with 12% (5%) applying NICE criteria. Prevalence was strongly related to age (see table). Kaplan–Meier curves for COPD mortality by disease severity showed greater separation under NICE criteria compared with GOLD. Following Cox regression analysis (adjusting for age, smoking pack years, diastolic blood pressure, cholesterol, body mass index and social class), participants meeting NICE diagnostic criteria were found to have increased hazard ratios (HR) by disease severity for both all-cause and COPD mortality, compared with those meeting the GOLD criteria. The HR for COPD mortality for men in the most severe COPD group, compared with those without COPD, was 92 (95% CI 58 to 148) applying GOLD and 110 (95% CI 70 to 171) applying NICE. A separate analysis showed respiratory symptoms and pack years, independent of lung function, to be significant contributors to mortality risk for all-cause and COPD mortality.

Conclusion: COPD prevalence in this population …