[Self-reported obituary columns in a local daily newspaper]

Objectives　To reveal the features of death and subsequent funeral services, we analyzed self-reported obituaries in The Shimotsuke, a local daily newspaper in Tochigi, Japan. In addition, the usefulness and disadvantages of such a database of deaths based on the obituaries were discussed.Method　For a 5-year period, from January 2011 through December 2015, all the obituary columns in The Shimotsuke were computerized as a database and analyzed. Some results were compared with the vital statistics data in the corresponding period of time. Data analyzed were: dead persons' address (municipalities), name, sex, cause of death, age at death, date of death and funeral services, the chief mourner (Moshu), and so on.Results　During the observed 5-year period, 69,793 deaths appeared in the columns, which were 67.6% of the all deaths in the vital statistics. No difference was observed in the proportions between the sexes in the published deaths in comparison with the vital statistics. In both sexes, the proportion was low in 0-9 years old, then high among 10-19 years, decreasing in 20s, and after then the proportion increased according to the age. The proportion was low in city areas, such as Utsunomiya and Oyama, and high in some cities and towns in the northern or eastern parts of Tochigi prefecture. The highest was in Motegi Town (88.0%), and the lowest was Nogi Town (38.0%), which is located in the southern-most part of the prefecture and is closest to the Tokyo metropolitan area. Almost all the funeral services were conducted within a week of the death, and no delays were observed to have been caused by the short supply of cremation services, which exists in large cities in Japan. In case where the chief mourner was a child, a parent, or a spouse of a child, the male sex was dominant. Analyses of death from senile decay, suicide, and homicide indicated that the accuracy of the published cause of death was low. The date of the funeral services was strongly influenced by superstitions in this country.Conclusions　Observations of the self-reported obituaries in a local daily newspaper over five years revealed the situations of the deaths in the area that could not be observed using vital statistics. Despite limitations, the information was partially useful as a database.