PT - JOURNAL ARTICLE AU - Huynh, Tran AU - Burstyn, Igor AU - MacDonald, Leslie AU - McClure, Leslie AU - Safford, Monika AU - Howard, Virginia TI - O-110 Occupation and incident stroke in a U.S. general population cohort 45 years of age or older (REGARDS study) AID - 10.1136/OEM-2021-EPI.22 DP - 2021 Nov 01 TA - Occupational and Environmental Medicine PG - A8--A9 VI - 78 IP - Suppl 1 4099 - http://oem.bmj.com/content/78/Suppl_1/A8.2.short 4100 - http://oem.bmj.com/content/78/Suppl_1/A8.2.full SO - Occup Environ Med2021 Nov 01; 78 AB - Objective To identify occupational groups with greater risk of stroke within a cohort of older adults in the United States.Methods Participants with occupational history data were selected from the REasons for Geographic and Racial Differences in Stroke (REGARDS) study (n=17,333) - a population-based prospective cohort of black and white men and women aged ≥45 years enrolled between 2003 and 2007. Self-reported occupational data were collected between 2011 and 2013. Participants were contacted every 6 months by telephone for self- or proxy-reported strokes later adjudicated by medical records through October 2018. Analyses focused on participants with a coded occupation who were employed in their longest held job prior to enrollment (n=15,016). Modified Poisson regression analyses were performed to estimate the relative risk (RR) for each two-digit Standard Occupational Classification (SOC) (2010 version) compared to all other occupation groups combined to highlight occupation groups at greatest risk of stroke.Results We observed 735 strokes (5%) during 15 years of follow-up. Results showed that participants employed in ‘food preparation and serving’ (22 cases; 8%;), ‘sales’ (69 cases; 7%), and ‘production’ occupations (80 cases; 6%;) experienced a greater occurrence of stroke, compared with all other occupational groups combined. These associations were robust to adjustment for Framingham stroke risk score (that includes age, sex, smoking status), race, region, and BMI with adjusted RRs 1.5 (95% CI, 1.0, 2.3), 1.4 (95% CI, 1.1, 1.7) and 1.2 (95% CI, 1.0, 1.5), respectively. Findings may be biased by incomplete job histories, crudeness of exposure assessment, latent confounding, and survivor effect.Conclusions We found evidence of an association between three two-digit occupation groups and incident stroke. We will better contextualize these results by refining the exposure assessment by examining associations with more detailed three-digit occupation groups, include all jobs held prior to enrollment, and incorporate employment duration.