Association for Public Policy Analysis & Management

Importance: Pharmaceutical industry marketing of opioid medications has been largely blamed for recent unprecedented increases in opioid overdoses. Hadland et al. (2019) identified positive associations between direct marketing to physicians and both opioid prescribing rates and prescription mortality. These findings support calls to reduce opioid prescribing, which were originally predicated on the observation that areas with higher prescribing rates also have higher opioid-related mortality rates. However, further analysis shows regions with the largest decreases in prescribing also had the largest increases in drug-related mortality.

Objective: To determine if Hadland et al. (2019) is robust to various adjustments.

Design, Setting, and Participants: County-level analysis of industry marketing information using data from the CMS Open Payments database linked with 1-year lagged data from the CDC and Prevention on opioid prescribing and mortality.

Main Outcomes and Measures: County-level mortality counts, opioid prescribing rates, marketing interactions with physicians, and sociodemographic factors.

Results: Of six exposure variable estimates, only one was replicated. Updated estimates showed a 1-SD increase in marketing increased deaths from prescription opioids, but decreased deaths from heroin and all drug overdoses. Opioid prescribing also increased with marketing and partially mediated these effects on drug overdoses.

Conclusions and Relevance: Across 3,139 US counties, opioid-related marketing was associated with marginal increases in prescribing and opioid-related mortality, but decreases in black-market drug overdose mortality. The central strategy to reduce the number of opioids prescribed by physicians may have decreased prescription-related deaths at the expense of increasing all drug-related deaths.