Risk of opioid-related adverse events among Medicaid and workers' compensation patients receiving long-term opioid therapy
Death from opioid analgesics has become a national epidemic after a change in federal regulations in 1999 liberalized the use of opioids in long-term treatment for chronic non-cancer pain (CNCP). Opioids were increasingly prescribed for CNCP in the last decade despite weak evidence on the safety and effectiveness of long-term opioid therapy. In 2007, Washington State (WA) agencies implemented an opioid dosing guideline on safe prescribing for CNCP as an educational tool for providers to safely prescribe opioids for CNCP at effective doses. We conducted two population-based observational studies in WA state-funded programs after implementation of the opioid dosing guideline. The objective of the first study was to determine whether opioid dose is associated with the risk of opioid-related mortality using a population-based retrospective cohort study design among opioid users who were enrolled in WA Medicaid. We also investigated the risk of opioid-related mortality associated with patterns of opioid use such as duration of use, chronic use, opioid duration of action, and other prescription drug use. The objective of the second study was to evaluate opioid use and dosing before and after Guideline implementation in the WA workers’ compensation population.
Study 1 Methods and Results
We identified 150,821 patients ages 18-64 years with non-cancer pain who had >1 opioid prescriptions in WA Medicaid during April 1, 2006-December 31, 2010. We used bivariate analysis to examine indicators of opioid use with opioid-related death. Overall and stratified rates of opioid-related deaths per 100,000 person-years were calculated among the WA Medicaid study population. Cox proportional hazards models were used to calculate unadjusted and demographic-adjusted hazard ratios (HRs) and 95% Confidence Intervals (CIs) for risk of opioid overdose death with opioid use and dosing factors. The majority of patients were short-term users with <3 months of opioid use overall (76.5%) but almost three-quarters of the opioid deaths (73.4%) occurred in patients who were chronic users. The median average daily dose was 40 mg/day morphine equivalent dose (MED). The majority of patients who died from an opioid overdose had a current opioid prescription (76.9%). Only a small fraction of patients (4.4%) had doses >200 mg/day MED at last use, but this group constituted more than a quarter of the opioid overdose deaths (27.8%). The overall rate of opioid-related death was 125.8 per 100,000 person-years, and as high as 684.0 per 100,000 person-years among patients with >200 mg/day MED. The overdose death rates among chronic users were higher than the rates among the overall study population, even at moderate doses. Among patients at 50-89 mg/day MED, the rate in chronic users was 425.7 per 100,000 person-years, whereas the rate in all users was 261.8 per 100,000 person-years. The risk of opioid overdose death significantly increased with increasing doses at or above 50 mg/day MED in all users as well as in chronic users only. Among all users, compared with patients with low doses of 1-19 mg/day MED, patients with 50-89 mg/day MED had almost 3 times the risk of overdose death (adjusted HR: 2.7, 95% CI: 1.5 - 4.8) and patients with >200 mg/day MED had more than 5 times the risk (adjusted HR: 5.3, 95% CI: 3.1 - 9.1). The HRs among chronic users were higher than among all users, particularly at doses of 120-199 mg/day MED (adjusted HR: 4.8, 95% CI: 2.3- 10.1) and >200 mg/day MED (adjusted HR: 5.9, 95% CI: 3.0-11.4). Chronic users had a 6-fold increased risk of opioid overdose death relative to non-chronic users (adjusted HR: 6.3, 95% CI: 4.8-8.2). Opioid overdose risk increased with cumulative duration of opioid use (continuous and non-continuous days), even at shorter use. Users at 31-89 cumulative days of opioid use were 4 times more likely to have an opioid-related death than users with <30 cumulative days of use (adjusted HR: 4.3, 95% CI: 2.7-6.9). Opioid controlled substance Schedule and duration of action were also associated with the risk of opioid overdose death. The greatest risk was among patients with prescriptions for both long-acting and short-acting Schedule II opioids versus those with non-Schedule II opioids only (adjusted HR: 4.7, 95% CI: 3.2-6.8).
Study 2 Methods and Results
We identified 161,283 workers aged 18-64 years with non-cancer pain who had >1 opioid prescriptions in WA Workers’ Compensation during April 1, 2004-December 31, 2010. Prevalence and incidence rates of opioid use were assessed. We compared pre- and post-Guideline chronic and high-dose use (>120 mg/day) among incident users. The mean monthly prevalence of opioid use declined by 25.6% between 2004 (14.4%) and 2010 (10.7%). Fewer incident users went on to chronic opioid therapy in the post-Guideline period (4.7%; 95% CI, 4.5%-5.0%) than in the pre-Guideline period (6.3%; 95% CI, 6.1%-6.6%). Compared with pre-Guideline incident users, post-Guideline incident users were 35% less likely to receive high doses (adjusted odds ratio=0.65; 95% CI, 0.59-0.71).
The results of these studies demonstrate significantly increased risk of opioid overdose death among various groups of opioid users and emphasize the importance of continuous monitoring throughout the course of therapy. The study findings suggest that opioid dosing guidelines that specify a “yellow flag” dosing threshold may be a useful tool in preventing escalation of doses into ranges associated with increased mortality risk. Although the extent to which decreases were due to the Guideline is uncertain, to our knowledge, this was the first report of significant decreases in chronic and high-dose prescription opioid use among incident users. Dosing guidelines have generally targeted CNCP patients, but patients with short-term opioid therapy for acute pain may also be at increased risk of opioid overdose death. Evidence-based strategies for opioid risk management are needed to help abate the epidemic of opioid-related morbidity and mortality.