Effect of reductions in opioid prescribing on opioid use disorder and fatal overdose in the United States: a dynamic Markov model

Research output: Contribution to journalArticlepeer-review


Background and aims: Despite prescribing declines between 2011 and 2019, opioid morbidity and mortality in the United States continued to rise during this period. We estimated the relationship between opioid prescribing, opioid use disorder (OUD) and fatal opioid overdose in the United States. Design: Dynamic Markov model. Setting: United States, using data from the National Survey on Drug Use and Health, Centers for Disease Control and Prevention, National Health and Nutrition Examination Survey and National Epidemiologic Survey on Alcohol and Related Conditions III. Participants: Simulated US individuals 12+ years of age from the general population or with prescription opioid medical use, prescription opioid non-medical use, illicit opioid (e.g. heroin, illicit fentanyl) use, prescription OUD, illicit OUD with a history of prior prescription opioid non-medical use or non-fatal or fatal opioid overdose. Measurements: Active OUD cases and fatal prescription opioid overdoses. Findings: Between 2010 and 2019, opioid prescribing declined 42.5%. Although fatal opioid overdoses increased by 103.2%, these reductions in opioid prescribing averted an estimated 9600 [95% uncertainty interval (UI) = 7205, 15 478] deaths starting in 2011 relative to continued prescribing at 2010 levels—and are projected to avert another 50 918 (95% UI = 38 829, 79 795) overdose deaths between 2020 and 2029. The median time from initial opioid prescription to fatal opioid overdose was 5.2 years. Of the 2.4 million (95% UI = 2.2 million, 2.7 million) individuals in the United States with estimated active OUD in 2019, 65% (95% UI = 59%, 71%) were attributable to initial opioid use occurring prior to 2011, whereas 14% (95% UI = 12%, 17%) were attributable to initial opioid use occurring between 2017 and 2019. The impact, by 2029, of additional reductions in prescribing initiated in 2020 would be more than three times greater than that of similar reductions initiated in 2025. Conclusions: Observed reductions in opioid prescribing volume in the United States from 2010 to 2019 appear to have saved approximately 9600 lives by 2019 and are anticipated to avert more than 50 000 fatal overdoses by 2029.

Original languageEnglish (US)
Pages (from-to)969-976
Number of pages8
Issue number4
StatePublished - Apr 2022


  • Markov model
  • United States
  • mortality
  • opioids
  • prescribing patterns
  • temporal trends

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Psychiatry and Mental health


Dive into the research topics of 'Effect of reductions in opioid prescribing on opioid use disorder and fatal overdose in the United States: a dynamic Markov model'. Together they form a unique fingerprint.

Cite this