OBJECTIVES: Psychotropic polypharmacy is common in clinical practice although supporting evidence is limited. In Japan, polypharmacy reduction policy for psychotropic drugs (anxiolytics, hypnotics, antidepressants, and antipsychotics) was introduced in 2012. This study aimed at examining the association of psychotropic polypharmacy with safety events of interest; hypertension, diabetes mellitus, pneumonia, extrapyramidal syndromes (EPS), and hyperlipidemia, based on the classification of the policy at the second revision in 2016. METHODS: Monthly administrative claims data between April 2011 and March 2016 were extracted from a large-scale Japanese health insurance claims database (MinaCare Co., Ltd.). Sequence symmetry analyses were conducted, which assessed asymmetry in the distribution of the safety event’s occurrence defined by both prescription for any therapeutic drugs and diagnoses of the event before and after initiation of psychotropic drugs. The adjusted sequence ratio (ASR) and its 95% confidence interval (CI) were calculated. RESULTS: The results for hypertension, diabetes mellitus, and pneumonia found no consistent signals in any categories of psychotropic drugs. On the other hand, all categories of psychotropic drugs were significantly associated with EPS, and the tendency was stronger in polypharmacy. Benzodiazepine receptor agonists (BZs) are mainstay anxiolytics/hypnotics, and a clearer association between polypharmacy of BZs and EPS was indicated. Furthermore, it was suggested that prescription of 2 or more BZs was associated with hyperlipidemia. In the analyses by subclasses of drugs, BZs, tetracyclic antidepressants, selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors, typical and atypical antipsychotics were significantly associated with EPS. The ASR of atypical antipsychotics (ASR 11.3; 95% CI 7.95–16.4) was higher than that of typical ones (ASR 2.96; 95% CI 2.06–4.33). CONCLUSIONS: The association between psychotropic polypharmacy and EPS, and between polypharmacy with BZs and hyperlipidemia was indicated. More strict categorization considering subclasses and potencies of drugs will be necessary in the Japanese polypharmacy reduction policy.