Associations between cannabis use and availability on prescription opioid use and related outcomes among injured workers

INTRODUCTION: Work-related injuries are a significant contributor to the high prevalence of acute and chronic, non-cancer pain in the U.S. While opioids are often prescribed for work- related injuries, inappropriate prescribing of opioids for pain management has contributed to the rise in opioid dependence, overdose, and death. Increasing access to medical and nonmedical (also called “recreational”) cannabis has been promoted as a strategy to combat the dual crises of pain and opioid-related harms in recent years. However, the health impact of cannabis use for pain and the impact of the increased availability to cannabis remains unclear. In this dissertation, we examined 1) the longitudinal relationships between cannabis use for pain and prescription opioid use, physical function, and psychological outcomes, 2) associations between cannabisretail outlet availability and cannabis use for pain, and 3) county-level associations between cannabis retail outlet availability and occupational injury rates. METHODS: We used data from the Washington and Ohio Worker (WOW) study, which aimed to examine the effectiveness of workers’ insurance programs in reducing unsafe opioid prescribing among injured workers who received an opioid prescription during the first 6 weeks of occupational injury. Participants completed surveys on pain intensity, pain interference, pain management strategies, and use of prescription opioid medications over a one-year period after injury. In Chapter 2, we assessed longitudinal associations between cannabis use for pain on prescription opioid use, anxiety, depression, pain intensity, and pain interference by using surveys conducted at around 6-weeks (baseline), 6-months, and 12-months after injury. Chapter 3 examined the cross-sectional association between cannabis retail outlet availability and cannabis use for pain. In Chapter 4, we conducted a retrospective, longitudinal county-level analysis examining the association between cannabis outlet retail availability and subsequent occupational injury rates by using workers’ compensation claims data for all injured workers covered under state-funded workers’ compensation insurance programs from 2014 to 2019. RESULTS: Approximately 20% of injured workers reported using cannabis for pain at baseline. Self-reported use of prescription opioids in the past week significantly declined over time. There was a significant group by time interaction, such that the reduction in the prevalence of self- reported prescription opioid use over time was slower among participants who used cannabis for pain, compared to those who did not use cannabis for pain (interaction adjusted prevalence ratio [aPR]: 1.31, 95% confidence interval [CI]: 1.02, 1.69). Over time, there were no differences in anxiety symptoms (adjusted count ratio [aCR]: 1.03, 95% CI: 0.97, 1.09), depressive symptoms (aCR = 1.05, 95% CI: 0.98, 0.13), pain intensity [aCR: 1.02, 95% CI: 0.97, 1.06), pain interference with enjoyment of life (aCR = 1.01, 95% CI: 0.95, 1.07), and pain interference with general activities (aCR = 0.99, 95% CI: 0.93, 1.05) between people who used cannabis for pain and those who did not. In Chapter 3, we found that participants living within a 1.0-mile road network buffer had an elevated likelihood of using cannabis for pain after adjusting for both individual and neighborhood-level covariates; however, this association was not statistically significant (aPR = 1.11; 95% CI: 0.94, 1.30). Sensitivity analyses suggested stronger associations with the larger spatial scales (1.5-mile buffer aPR = 1.17; 95% CI: 1.00, 1.37) compared to the more proximal scale (0.5-mile buffer aPR = 1.04; 95% CI: 0.81, 1.33). Lastly, in Chapter 3, we found little evidence of an association between county-level cannabis retail outlet availability and overall rates of occupational injuries (adjusted incidence rate ratio [aIRR] = 1.00; 95% CI: 0.97, 1.04). Findings suggest county-level associations of cannabis availability with occupational injuries may vary by industry. A within-county increase of 1 retail outlet per 10,000 residents was associated with an 12% increase in injury rates among workers in transportation and warehousing, although results were not statistically significant (aIRR = 1.12; 95% CI: 0.99, 1.27). CONCLUSIONS: Cannabis use for pain was prevalent among injured workers in WA. Findings from this dissertation suggest a cautionary approach in the recommendation and use of cannabis for pain, as well as the need to consider regulatory approaches that may limit excessive cannabis commercialization. As more states legalize cannabis for use, understanding the health impacts of cannabis use for pain and the increased availability to cannabis is necessary in order to prevent unintended harms.