Cause-specific mortality among people previously hospitalized with opioid-related conditions: a retrospective cohort study

PurposeAlthough numerous studies have reported on all-cause mortality among opioid users, few have provided estimates for specific causes of death. We report all-cause and cause-specific mortalities for a large cohort of individuals with a history of opioid-related disorders.MethodsWe used administrative data on people discharged from California hospitals between 1990 and 2005 with diagnoses of opioid dependence, abuse, poisoning, or withdrawal. We calculated crude and standardized cause-specific mortality rates and ratios.ResultsData included 68,066 individuals and 390,438 person-years of follow-up. There were 13,107 deaths (a crude mortality rate of 33.6 per 1000 person-years). The standardized mortality rate was 37.6 per 1000, and the standardized mortality ratio (SMR) was 4.36 (95% confidence interval [CI], 4.30–4.43). Standardized mortality ratios were greater than 1 for all specific causes except Alzheimer's disease, and were the highest for drug-induced deaths (62.2, 95% CI, 58.2–66.1), human immunodeficiency virus (24.6, 95% CI, 22.3–26.9), and endocarditis (23.2, 95% CI, 15.3–31.3). The broad causes associated with the most excess deaths were drug- or alcohol-induced death (23% of excess deaths), cancer (18%), heart disease (14%), and lower respiratory disease (7%).ConclusionsMortality in this cohort was very high. Most excess deaths were not directly due to drug use, but many were from causes associated with the use of tobacco or alcohol. Mortality for cancers not linked to tobacco or alcohol was also elevated, suggesting that admission for opioid-related causes serves in part as a risk marker for preexisting disease.