Background Occupational exposures are important and preventable causes of COPD. In a cross-sectional study of current occupation among over 220,000 workers in the UK Biobank cohort (over 500,000 subjects) we previously reported that 14 jobs were associated with increased COPD risk (De Matteis S, et al., OEM, 2016). To progress these findings we developed OSCAR, a new web-based tool for efficient self-reporting and automatic coding of job-histories in large population-based studies. Our aim was to identify the occupations at increased COPD risk taking into account lifetime job-histories in the UK general population.

Methods We administered OSCAR to all UK Biobank participants with an available email address (n = 324,653) between June-February 2016. Paid jobs held for at least six months were collected and coded by OSCAR using the UK Standard Occupational Classification (SOC), v.2000. COPD was defined as FEV1/FVC<LLN based on spirometry performed at recruitment. Prevalence ratios (PRs) for ever-exposure to each of the 353 SOC-coded jobs using lifetime office workers as reference category were estimated using Poisson regression with adjustment for age, sex, recruitment centre and lifetime tobacco smoking. In addition, we used lifetime cumulative job durations to test for exposure-response trends.

Results Among the 116,375 participants who completed OSCAR (response rate: 34%), 94,551 had acceptable and repeatable spirometry data (according to ERS/ATS criteria) and smoking information and were included in the analyses. Taking into account individual lifetime job-histories, 19 jobs significantly increased COPD risk (e.g. ‘fishing and agriculture related occupations’: PR: 1.69; 95% confidence interval (CI): 1.18–2.42, and ‘food, drink and tobacco process operatives’: PR: 1.42; 95% CI: 1.02–1.97). The majority were also confirmed by positive exposure-response trends for cumulative years of employment in each job/lifetime.

Conclusions Compared to our previous cross-sectional study, some jobs were confirmed (e.g. ‘food, drink and tobacco process operatives’, and ‘horticultural trades’) while others were not (e.g. ‘coal miners’), likely due to the few employed in these jobs among the OSCAR responders. OSCAR is still collecting job-histories in the UK Biobank cohort; further analyses on a larger sample, as well analyses restricted to never-smokers and never-asthmatics to investigate the possibility of residual confounding, are planned.