Association for Public Policy Analysis & Management

Recent work has begun to examine the link between work outcomes and opioid prescribing rates in a local area. For example, Krueger (2017) shows that labor force participation rates of prime age men are lower in areas with higher rates of opioid prescribing and Savych et al. (2017) find that injured workers receiving Workers’ Compensation (WC) benefits remain out of work longer if they reside in an area that has higher opioid prescribing rates. We extend this recent work to examine how a plausibly exogenous change in local area prescribing rates impacts work outcomes. We examine how work outcomes change following state adoption of a “strong” Prescription Drug Monitoring Program (PDMP) that requires a physician to check the PDMP before prescribing a controlled drug.

Prior work has shown that opioid use falls when a state adopts a prior access PDMP. Buchmueller and Carey (2017) use Medicare Part D claims to show that opioid prescriptions to Medicare recipients fall following state adoption of a prior access PDMP. Dave et al. (2017) show that admissions to a treatment facility for opioid abuse also fall following adoption.

We first present results confirming whether we observe a link between state adoption of a prior access PDMP and county level prescribing rates. We quantify opioid prescriptions using data from the Centers for Disease Control and Prevention from 2006 through 2016. We then examine the reduced form relationship between state adoption of a mandatory access PDMP and various labor market outcomes at the state and county level.

A worker using prescription opioids may have a harder time returning to work, so opioid use may impact labor force participation. We measure this using the county labor force participation rate from the Bureau of Labor Statistics Local Area Unemployment Statistics.

The National Institute for Occupational Safety and Health also cautions that workers using prescription opioids may be at greater risk of workplace injury, and there may be spillover effects to coworkers. We measure workplace injury rates using several different datasets. At the state level, we use the rate of nonfatal injuries from the Survey of Occupational Illness and Injury (SOII). At the county level, we compute the per capita rate of work-related Emergency Department discharges, which we identify as discharges that list an injury at an industrial site (Ecode 849.3) or list WC as payer. These data are available from the HCUP State Emergency Department Databases (SEDD) ; we have access to a subset of states data, thus this component does not encompass all states.

Together, our results provide important new evidence regarding the impact of opioid prescriptions and work outcomes. As researchers and policymakers are beginning to quantify the impact of the opioid epidemic, understanding the toll on workplace safety is especially important.