Exploring factors affecting workers' acquisition of exercise habits using machine learning approaches

Tsukuba, Japan—Physical inactivity is the fourth leading mortality risk factor, following hypertension, smoking, and hyperglycemia. Therefore, acquiring an exercise habit is crucial to maintain and improve health. In Japan, Specific Health Guidance is provided to support the improvement of lifestyle habits, including exercise habits. To develop more efficient health guidance, it is important to identify factors that influence its effectiveness (e.g., characteristics and lifestyle of the target population). In this study, data from middle-aged workers who received Specific Health Guidance were analyzed using machine learning to explore the factors associated with the acquisition of exercise habits, and the importance of each factor was evaluated.

The researchers conducted a secondary analysis of data obtained by health insurance societies and other organizations through health projects in 2017-2018. They found that the most critical factor associated with the acquisition of exercise habits was "the higher stages of behavioral change toward lifestyle improvement," followed by "high level of physical activity" and "high density lipoprotein cholesterol level being within the reference range." In contrast, "daily alcohol consumption of ≥60 g" had a negative effect on the acquisition of exercise habits.

This study revealed the factors related to the characteristics and lifestyles of middle-aged workers who received Motivational Health Guidance under the Specific Health Guidance program that positively associate with the acquisition of exercise habits. The results of this study may contribute to developing more efficient health guidance.

###
This work was supported by the Japan Agency for Medical Research and Development (grant numbers 21ek0210124h9903 and JP23rea522107).

_{Original Paper}

Title of original paper:
Factors associated with acquiring exercise habits through health guidance for metabolic syndrome among middle-aged Japanese workers: A machine learning approach

Journal:
Preventive Medicine Reports

DOI:
10.1016/j.pmedr.2024.102915

_{Correspondence}

Professor NAKATA, Yoshio
Institute of Health and Sport Sciences, University of Tsukuba

Specially Appointed Professor TSUSHITA, Kazuyo
Faculty of Nutrition, Kagawa Nutrition University

Lecturer of hospital ONOUE, Takeshi
Department of Endocrinology and Diabetes, Nagoya University Graduate School of Medicine

Lecturer WAKABA, Kyohsuke
Faculty of Human Life, Jumonji University

_{Related Link}

Institute of Health and Sport Sciences