Title: Development of a Community Screening tool for COPD in India: The OSCAR study

Abstract

Background: COPD is a huge and growing burden in India and other low- and middle-income countries (LMICs),yet remains poorly diagnosed. Aimed to develop a community-based COPD screening tool,called OSCAR,which could be used by Community Health Workers for a nationwide COPD screening program in India.

Methods: 5420 male and female subjects ≥30 years were approached from randomly selected 11 urban slums and 7 rural villages from Pune district in India and administered a questionnaire containing 129 variables developed through literature review,in-depth interviews (COPD patients) and focused group discussion (pulmonologists). Pre and post bronchodilator spirometry along with peak flow metry were performed. Using post-bronchodilator FEV1/FVC<0.7 as the gold standard for COPD, the 129 variables were reduced using logistic regression and random forest analysis to the smallest set of questions that offered maximum sensitivity, specificity, positive predicted value, and negative predicted value.

Results: Clean data was obtained from 4258 participants.8 variables i.e.,age, gender,smoking status,shortness of breath,cough,wheeze,chest tightness and long-standing history of asthma along with a peak flow meter(<250L/min for males and <200 L/min for females) together yielded a SN,SP,PPV and NPV of 51.6%, 97.5%, 52.4% and 97.4% respectively, with an overall accuracy of 95.2%, for a cut-off value of ≥ 8 (max 14).

Conclusion: The OSCAR COPD Screening tool comprising of 8 item questions and a Peak Flow Metry has a reasonable accuracy, compared to other studies. This is the first community-based COPD Screening tool developed in India that has shown reasonably good accuracy and may have a wide application COPD screening in LMICs.