Geographical Variation in Psychiatric Admissions Among Recipients of Public Assistance

Background: Understanding the area-specific resource use of inpatient psychiatric care is essential for the efficient use of the public assistance system. This study aimed to assess the geographical variation in psychiatric admissions and to identify the prefecture-level determinants of psychiatric admissions among recipients of public assistance in Japan. Methods: We identified all recipients of public assistance who were hospitalized in a psychiatric ward in May 2014, 2015, or 2016 using the Fact-finding Survey on Medical Assistance. The age- and sex-standardized number of psychiatric admissions was calculated for each of the 47 prefectures, using direct and indirect standardization methods. Results: A total of 46,559 psychiatric inpatients were identified in May 2016. The number of psychiatric admissions per 100,000 population was 36.6. We found a 7.1-fold difference between the prefectures with the highest (Nagasaki) and lowest (Nagano) numbers of admissions. The method of decomposing explained variance in the multiple regression model showed that the number of psychiatric beds per 100,000 population and the number of recipients of public assistance per 1,000 population were the most important determinants of the number of psychiatric admissions (R-2 = 28% and R-2 = 23%, respectively). The sensitivity analyses, using medical cost as the outcome and data from different survey years and subgroups, showed similar findings. Conclusions: We identified a large geographical variation in the number and total medical cost of psychiatric admissions among recipients of public assistance. Our findings should encourage policy makers to assess the rationale for this variation and consider strategies for reducing it.