ABSTRACT
The built environment is a structural determinant of health. Here we reveal spatially heterogeneous associations of built environment indicators with objective health outcomes (morbidity) by combining a random forest (RF) approach and a multiscale geographically weighted (MGWR) regression method. Using data from six Japanese cities, we found that the ratio of morbidity has obvious spatial agglomerations. The mixed land-use diversity with 1000 m buffer, distance to hospital, proportion of park area with 300 m buffer, and house price with 2000 m buffer, negatively affect health outcomes at all locations. For most locations, high PM2.5 or high floor area ratio with 2000 m buffer are linked to a high ratio of morbidity. Our findings support the use of such data for long-term urban and health planning. We expect our study to be a starting point for further research on spatially heterogeneous associations of the built environment with comprehensive health outcomes.
Acknowledgement
This research is a part of the research project, titled “Development of a Health Education System Through Risk Prediction and Targeting Using Artificial Intelligence Based on National Health Insurance Claim Data and Health Information”, funded by the Artificial Intelligence-Utilization Project for Health Direction System, Japan Agency for Medical Research and Development (Project ID. 19ls0210002h0003). All authors have contributed significantly and that all authors are in agreement with the content of the manuscript.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Ethical considerations
This study was conducted with the approval (C-203 - C-203- 7) by the Research Ethics Committee of Hiroshima University, Japan. It conforms to the provisions of the Declaration of Helsinki. Since we did not obtain any consent directly from respondents, we opted out of Hiroshima University “Information Disclosure of Research on Medical Ethics” on the website.