Ann Epidemiol

Background Participation in health studies may be inversely associated with employment and stress. We investigated whether employment, perceived stress, work-related stress, and family caregiving were related to participation in a longitudinal US community-based health study of black and white men and women aged ≥45 years. Methods Prevalence ratios and confidence intervals were estimated for completion of the second-stage (S2) of a two-stage enrollment process by employment (status, type), and stress (perceived stress, work-related stress, caregiving), adjusting for age, sex, race, region, income, and education. Eligibility and consent for a follow-up occupational survey was similarly evaluated. Results Wage- but not self-employed participants were less likely than the unemployed to complete S2. Among the employed, S2 completion did not vary by stress; however, family caregivers with a short time burden of care (<2-hours/day) were more likely to complete S2, compared to non-caregivers. Eligibility and participation in the follow-up occupational survey was higher among those employed (versus unemployed) at enrollment, but was not associated with enrollment stress levels. Conclusions Limited evidence of selection bias was seen by employment and stress within a large US community-based cohort, but findings suggest the need for enrollment procedures to consider possible barriers to participation among wage-employed individuals.

28890281

PMC5632192

CDC Public Access