Abstract
A total fertility rate (TFR) of 2.07 is considered necessary for demographic stability in developed countries, but Japan’s TFR (1.30) ranked 35th among 38 OECD countries in 2021. There is a great need for studies about TFR recovery. However, one inhibiting factor is how to adjust for the unstable TFR in small municipalities such as islands. Previous studies proposed methods using standard population or Bayesian correction. A second factor is the nationwide major mergers inhibiting continuous long-term observation of municipalities.
In this study, all municipalities (2000-2020) were divided into 335 secondary medical areas (SMAs). These were divided into two groups: the 26 highest-TFR SMAs showing an increasing trend, and the 12 lowest-TFR SMAs showing a decreasing trend. A total of 18-49-year-old 700 women were selected and administered an anonymous questionnaire. Binary logistic regression analysis revealed that respondents living on islands (odds ratio=26.75) and those whose final education level was high school (1.47) were more likely to belong to highest-TFR group, and that respondents who lived in cities (0.47), were never married (0.35), were married (first marriage) (0.54), and were housewives (0.53) less likely to belong to the highest-TFR group. In addition, the synergy analysis between co-occurrence network words and quantitative analysis showed that Kyushu Region and Okinawa Prefecture, where TFR values are high or rising over time, could be more accurately studied by the new method of ensuring that all the data, including that of small municipalities, is used rather than applying statistical adjustments. The results indicated that women in the highest-TFR group had high expectations for financial subsidies provided by local governments for each stage of pregnancy, childbirth, and childcare. There was no difference in the values toward children, suggesting the need to enhance financial support and maintain a detailed, multi-layered service system surrounding pregnancy, childbirth, and childcare.
