Association Between Opioid Dosage Tapering and Opioid Overdose Among Long-Term Higher-Dose Opioid Users

There has been a steady decline in the national opioid-dispensing rate in recent years. However, emerging evidence that suggests that abrupt discontinuation of opioids for pain may increase overdose risk. The authors examined associations between prescribed opioid tapering and short-term risk of overdose among adults who have received a new long-term high-dose opioid prescription.

This cohort study included adults (aged ≥18 years) without cancer who were newly prescribed a long-term high-dose opioid, that is, daily dose ≥50 morphine milligram equivalents for ≥183 days, between 2013 and 2018 in Kaiser Permanente Southern California. Opioid tapering within the evaluation period (3 months after being prescribed long-term high-dose opioid) was defined as a reduction ≥10% in monthly morphine milligram equivalents for at least 2 consecutive months compared with baseline morphine milligram equivalents. The association of opioid tapering with incident opioid overdose within 12 months was examined separately using multivariable Cox proportional hazard models with inverse probability of treatment weight. Sensitivity analyses were performed to differentiate the association by tapering rates.

Among 12,866 eligible individuals, 7,372 (57.3%) initiated tapering within 3 months. Seventy-five patients had a documented opioid overdose event within 12 months during follow-up, resulting in an incidence rate of 0.022 and 0.013 event per 1,000 person-years, respectively, among the nontapered and tapered cohorts. Tapering within the evaluation period was associated with a 0.48 (hazard ratio=0.52; 95% CI=0.33, 0.82) reduction in risk of overdose in the subsequent 12 months. Sensitivity analyses suggested that a tapering rate of 20%-40% reduction monthly was associated with lower risk.

Among patients newly prescribed long-term high-dose opioids, initiating tapering within 3 months was shown to be associated with reduced short-term risk of opioid overdose. However, more studies are needed for further confirmation.