The Causes and Consequences of Opioid Use among Older Americans: A Panel Survey Approach

Abstract

This study examines the effects of prescription opioid analgesic use for older Americans, specifically with regard to work disability and disability program participation. We draw on the long-panel structure of the Health and Retirement Study and a newly available 2009 survey module measuring prescription drug use and initiation. We pursue regression-adjustment and nearest neighbor matching approaches, using rich 2008 HRS measures on health, disability, sociodemographic characteristics, and economic status, to account for selection into prescription opioid use, since supply-side instruments used in the opioid literature have little relevance to opioid use for this population in 2009. Pre-2008 comparisons between individuals with 2009 opioid prescriptions and controls demonstrate face validity of the analytic approach; we then estimate opioid use effects on mortality, self-reported health, labor force participation, work-limiting health conditions, and disability program participation, spanning from 2010 to 2018. We find substantial and significant mortality effects starting in 2010; in estimating effects on other outcomes, we account for differential attrition through mortality via inverse probability reweighting. Our findings are significant, both statistically and economically: up through 2018, individuals with 2009 opioid prescriptions were nearly 40% more likely to develop a health condition that limited their ability to work than those without a prescription. This difference in work disability led to substantial differences in disability program participation: those using opioids were nearly 300% more likely to apply for or receive Social Security disability benefits by 2018.