Logistic Regression Analysis of the Effect of Third Places on the Risk of Needing Long-Term Care

Sustainable Built Environment Conference 2019 Tokyo Built Environment in an era of climate change: how can cities and buildings adapt?

生态环境科学

Hisamatsu, Shinya^1 ; Ando, Shintaro^1 ; Sakamoto, Sho^1

University of Kitakyushu, Fukuoka, Japan^1

关键词: Analysis models;    Educational attainments;    Elderly people;    Health condition;    Logistic regression analysis;    Long term care;    Questionnaire surveys;    Social isolation;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/294/1/012086/pdf
DOI  :  10.1088/1755-1315/294/1/012086

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

In Japan, late elderly people (≥75 years old) tend to be unable to create social groups, because they have decreased body functions and structures. Previous studies have shown that those who are able to visit so-called "third places", places such as cafes, libraries, and parks where people can go and spend time freely, are significantly more outgoing and have more conversations with people. The aim of our survey was to clarify the effects of the third places on the need for long-term care and support in the late elderly. We conducted a questionnaire survey for 3 weeks from September to October 2016. The subjects were late elderly people located in Kitakyushu, Fukuoka. Survey questions included their utilisation of third places, their neighbourhood environment and their health condition. A total of 629 questionnaires were distributed by mail, and 349 were correctly completed (valid response rate = 55.5%). A total of 58.6% of respondents could identify their third place. Of those, 75.6% visit their third place more than once a week. We conducted logistic regression analysis to clarify the effect of a third place on the risk of being homebound (i.e., going out less than once a week), social isolation (having a conversation less than once a week), and long-term care/support needs. The analysis model considered age, BMI, marital status, body pain, household, educational attainment, and length of residence in the target area. The resulting odds ratio (OR) for the risk of being homebound was 8.46 (p