**2. Materials and Methods**

#### *2.1. Study Overview*

A cohort of all hourly employees newly hired from 2 January 2012 through 25 November 2016 by a large manufacturing facility in the US Midwest was established using the employer's human resources records database. We did not include employees newly hired into administrative, managerial, or other salaried positions (e.g., engineering). Occupational injury data from the same time period were extracted from a separate database maintained by the facility's on-site occupational health center. The two datasets were linked using an identification number assigned to each employee. The employer redacted all personal identifiers from the merged dataset prior to delivery to the research team. Written approval was obtained from the employer for use of the merged (de-identified) dataset, and the University of Iowa Institutional Review Board determined that the study did not meet the regulatory definition of human subjects' research.
