BackgroundPremature mortality, frequent relapse that easily leads to hospitalization, and discontinuous employment are key challenges for the treatment of schizophrenia. We evaluated risk factors for important clinical outcomes (death, hospitalization, resignation, and sick leave from work) in patients with schizophrenia in Japan.MethodsA nested case–control study was conducted for patients with schizophrenia identified in a Japanese claims database. For each outcome, the case was matched with up to four controls of the same age, sex, index year, and enrollment status (employee or dependent family). Potential risk factors were defined by prescriptions or diagnoses within 3 months prior to or in the month of the event. The association among potential risk factors and each outcome was evaluated using multivariable conditional logistic regression analysis with stepwise variable selection.ResultsThe number of cases and eligible patients for each outcome were 144 and 38,451 (death), 1,520 and 35,225 (hospitalization), 811 and 18,770 (resignation), and 4,590 and 18,770 (sick leave), respectively. Depression was a risk factor for death (odds ratio [OR]: 1.92; 95% confidence interval [CI]: 1.12, 3.29), hospitalization (OR: 1.22; 95% CI: 1.05, 1.42), and sick leave from work (OR: 1.46; 95% CI: 1.36, 1.57). Other risk factors for death were hospitalization history, Charlson Comorbidity Index (CCI) score, and prescription for laxatives. Prescriptions for hypnotics, laxatives, and anticholinergics were risk factors for hospitalization. Prescriptions for hypnotics and anticholinergics were risk factors for resignation. CCI score, prescription for hypnotics, laxatives, and antidiabetics were risk factors for sick leave from work.ConclusionsOur findings suggest that depression and some physical symptoms, such as constipation and extrapyramidal symptoms, are risk factors for important clinical outcomes in patients with schizophrenia. Attention should be paid to both depression and physical symptoms for the treatment of schizophrenia.