The impact of the abuse-deterrent reformulation of extended-release T OxyContin on prescription pain reliever misuse and heroin initiation

Authors

Carolyn Wolff, Food and Drug Administration
William N. Dowd, RTI International
Mir M. Ali, US Department of Health & Human Services
Chandler McClellan, Agency for Healthcare Research and Quality
Angelica Meinhofer, Weill Cornell Medicine,
Lukas Glos, Food and Drug Administration
Ryan Mutter, Congressional Budget Office
Matthew Rosenberg, Food and Drug Administration
Andreas Schikc, Food and Drug Administration

Date of this Version

2020

Document Type

Article

Citation

U.S. government works are not subject to copyright.

Comments

Addictive Behaviors 105 (2020) 106268

https://doi.org/10.1016/j.addbeh.2019.106268

Abstract

The introduction of abuse-deterrent OxyContin in 2010 was intended to reduce its misuse by making it more tamper resistant. However, some studies have suggested that this reformulation might have had unintended consequences, such as increases in heroin-related deaths. We used the 2005–2014 cross-sectional U.S. National Survey on Drug Use and Health to explore the impact of this reformulation on intermediate outcomes that precede heroin-related deaths for individuals with a history of OxyContin misuse. Our study sample consisted of adults who misused any prescription pain reliever prior to the reformulation of OxyContin (n = 81,400). Those who misused OxyContin prior to the reformulation were considered the exposed group and those who misused other prescription pain relievers prior to the reformulation were considered the unexposed group. We employed multivariate logistic regression under a difference-in-differences framework to examine the effect of the re- formulation on five dichotomous outcomes: prescription pain reliever misuse; prescription pain reliever use disorder; heroin use; heroin use disorder; and heroin initiation. We found a net reduction in the odds of pre- scription pain reliever misuse (OR:0.791, p < 0.001) and heroin initiation (OR:0.422, p = 0.011) after the reformulation for the exposed group relative to the unexposed group. We found no statistically significant effects of the reformulation on prescription pain reliever use disorder (OR: 0.934, p = 0.524), heroin use (OR: 1.014p = 0.941), and heroin use disorder (OR: 1.063, p = 0.804). Thus, the reformulation of OxyContin appears to have reduced prescription pain reliever misuse without contributing to relatively greater new heroin use among those who misused OxyContin prior to the reformulation.