All-cause mortality in patients with long-term opioid therapy compared with non-opioid analgesics for chronic non-cancer pain: a database study

BACKGROUND: Hitherto only studies with selected populations have found an increased all-cause mortality of some selected opioids compared to selected non-opioids for chronic non-cancer pain (CNCP). We have examined the all-cause mortality for CNCP associated with all established opioids compared to non-opioid analgesic therapy (anticonvulsants, antidepressants, dipyrone, non-steroidal agents). METHODS: The study used the InGef (Institute for Applied Health Research Berlin) database which is an anonymized healthcare claims database including 4,711,668 insured persons who were covered by 61 German statutory health insurances between 2013 and 2017.The health insurance companies are the owners of the database. All-cause mortality was determined from death certificates. Adjusted hazard ratios (HRs) including age, gender, comorbidity index, and propensity score as covariates and risk differences (RD) in incidence of death between patients with long-term opioid therapy (LTOT) and control-drug therapy were calculated. RESULTS: The mean age of participants was 66 years; 55% were women. There were 554 deaths during 10,435 person-years for the LTOT patients, whereas there were 340 deaths during 11,342 person-years in the control group. The HR for all-cause mortality was 1.59 (95% CI, 1.38–1.82) with a risk difference of 148 excess deaths (95% CI 99–198) per 10,000 person-years. The elevated risk of death for LTOT was confined to the out-of-hospital deaths: LTOT patients had 288 out-of-hospital deaths during 10,435 person-years (276 per 10,000 person-years) whereas there were 110 deaths during 11,342 person-years (97 per 10,000 person-years) in the control group. HR was 2.29 (95% CI 1.86, 2.83). Although our propensity score matching model indicated a good classification, residual confounding cannot be fully excluded. The opioid group had a higher prevalence of heart failure and a higher use of anti-thrombotic and antiplatelet agents and of psycholeptics. CONCLUSIONS: LTOT for CNCP compared to non-opioid analgesics was associated with an increased risk for all-cause mortality. When considering treatment options for patients with CNCP, the relevant risk of increased all-cause mortality with opioids should be discussed. TRIAL REGISTRATION: ClinicalTrials.gov, NCT03778450, Registered on 7 December 2018