SRA-E Lisbon 2017

How many people get injured in a year? This question is simple but difficult to answer and not always available from the existing mortality and morbidity statistics. European Union publishes this estimate every year in a pyramid like format by “EuroSafe” report. On top of the pyramid there is the number of accidental deaths. The second layer is the number of people got injured and hospitalized. The bottom layer is the number of outpatients. While in Japan, no efforts has been made so far to make this estimate, the authors has made an attempt to produce it for the first time in Japan. There is a few sources of data. The number of death is taken from mortality statistics, where all kinds of fatal accident casualties, traffic accident fatalities and other unintentional injuries at home, school, etc., homicides, assaults, self-suicides are included. The second layer estimate is taken from Hospital Statistics, an annual survey by the Ministry of Health, Labor and Welfare (MHLW) of Japan, which gives the number of inpatients at a chosen date in October every year. As average length of days of hospitalization is available from the same statistics, the total number of inpatients in a year can be calculated from these two figures. The bottom layer estimate is the most difficult to get. Authors calculated it using Hospital Statistics and Medical Care Insurance Payment database. Hospital Statistics gives the number of outpatients at a chosen date and an average number of times of hospital visits, but the estimate of outpatients from these two figures seems overestimated, because of possible duplicated counts of those patients who took cared by more than two hospitals, in parallel or consecutively. In order to avoid this duplication of counts, the authors analyzed Medical Care Insurance Payment database. Every year, billions of personal medical care payments records are produced. These are stored at the central database, which has been inaccessible to the public for long years. The Japanese government, however, recently decided to allow access to this database to promote the useful datamining efforts by private sectors, local governments, and academic researchers. All records are anonymized but the hash-ID is given for all patients and it makes possible to identify personal medical care history. By using this database, the authors estimated average times of hospital visits of outpatients. The study gives this estimates, methodology and comparisons with European Union.