The opioid epidemic is devastating individuals and communities across America. While the opioid epidemic affects all states, Colorado has been particularly hard hit as is evident by a 68% increase in opioid overdose deaths between 2002 and 2014. Several of Colorado's counties have among the highest death rates from opioids in the country. Recent research suggests marijuana could serve as a substitute for opioids, and diminish the harms from opioid misuse. Although Colorado legalized medical marijuana in 2000, dispensaries were not widespread until 2010. Recreational stores, made legal by a popular vote in 2012, started opening in January 2014. Colorado is therefore an excellent environment in which to evaluate whether increasing access to marijuana does in fact reduce opioid abuse. In this study we exploit the increase in access that occurred with the legalization of recreational marijuana to assess the impacts on opioid utilization and opioid related harm (abuse, dependence, poisonings and mortality). While individuals with specific medical conditions had access prior to recreational legalization, that access was not uniform throughout the state as local jurisdictions could choose to ban dispensaries. Similarly, even with legalization, jurisdictions could ban stores. We make use of this geographic variation with respect to bans on medical and recreational marijuana outlets as well as differential access to medical marijuana stores based on underlying health conditions to assess the impact of the policy change and growth in the number of stores on prescription drug use and harm within these markets. By combining information on the number of registered medical and recreational outlets with information on sales tax revenue, we are able to create a new measure of exposure that improves upon prior measures by also reflecting the size of the market. We link this information on marijuana market exposure to prescription drug utilization and harms available in the Colorado All Payers Claims Database (APCD). Difference-in-difference methods allow us to identify the effects of living in exposed and growing markets while controlling for broader state-level trends in opioid prescribing and abuse. Given weaknesses of the APCD, we also evaluate effects using other data, including the National Vital Statistics Survey, Hospital Emergency Department data, Hospital Discharge data, and poison center calls. Our study will be the first to provide a comprehensive assessment of whether increasing access to legal marijuana actually reduces opioid utilization and harm.

Public Health Relevance

Recent studies suggest marijuana liberalization policies may reduce opioid related harm by encouraging some users to switch to marijuana. Limitations of these prior studies make the evidence highly questionable, however. This study will overcome many of those limitations and provide direct evidence of whether and how local access to marijuana influences opioid utilization and harm. This research will shed new light on whether marijuana really is a substitute for opioids, and if so, whether it is among individuals getting their own prescriptions or recreational users.

National Institute of Health (NIH)
National Institute on Drug Abuse (NIDA)
Research Project (R01)
Project #
Application #
Study Section
Social Sciences and Population Studies B Study Section (SSPB)
Program Officer
Kimmel, Heather L
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Rand Corporation
Santa Monica
United States
Zip Code