Date Presented 03/23/24
This abstract focuses on classifying lifestyle types and examining the influence of personal values among older adults using machine learning. This topic contributes to both academic and practical concepts within a personal value-centered approach.
Primary Author and Speaker: Seungju Lim
Additional Authors and Speakers: Lim Young-Myoung, Ah-Ram Kim, Sanghun Nam, Sooyeon Yoo, Jungmin Han
Contributing Authors: Ji-Hyuk Park
PURPOSE: This study aims to categorize different types of healthy lifestyles among older adults and to identify the personal values that influence these healthy lifestyle types using machine learning.
DESIGN: This study is a cross-sectional study targeting middle-aged and older adults aged 55 and above living in local communities in South Korea.
METHOD: The study utilized data from 300 participants, collected through online surveys. The outcome variable for machine learning is the latent group of the Healthy Lifestyle Practice Index (HLPI), which consists of four lifestyle domains: Physical activity, nutrition management, social relationships, and activity Participation. These groups were determined through latent profile analysis. The predictor variables for machine learning were items from the Lifestyle-Value Index. Four machine learning algorithms were employed to classify the types of healthy lifestyle practices and identify the influence of personal values on these types.
RESULTS: The study classified lifestyle practice types into two categories: ‘Healthy Lifestyle Group’ at 48.87% and ‘Unhealthy Lifestyle Group’ at 51.13%. A significant mean difference between these two groups was observed in the ‘social relationships’ domain of HLPI. The machine learning model that exhibited the best fit was the support vector machine with an accuracy of 96% and an Area under the ROC Curve of 95%. Moreover, the most significant value for classifying their lifestyle types is the degree of interest in healthy food, health-related media, and leisure activities.
CONCLUSION: These findings emphasize the importance of a personal value-centered approach in managing the healthy lifestyle of older adults. The study highlights the necessity for healthy lifestyle interventions centered around the key personal values identified in this research.
References
Malcolm, M., Frost, H., & Cowie, J. (2019). Loneliness and social isolation causal association with health-related lifestyle risk in older adults: a systematic review and meta-analysis protocol. Systematic Reviews, 8(1), 1–8. https://doi.org/10.1186/s13643-019-0968-x
Huang, W., Ying, T. W., Chin, W. L. C., Baskaran, L., Marcus, O. E. H., Yeo, K. K., & Kiong, N. S. (2022). Application of ensemble machine learning algorithms on lifestyle factors and wearables for cardiovascular risk prediction. Scientific Reports, 12(1), 1033. https://doi.org/10.1038/s41598-021-04649-y
Poudel, G. R., Barnett, A., Akram, M., Martino, E., Knibbs, L. D., Anstey, K. J., . . . & Cerin, E. (2022). Machine learning for prediction of cognitive health in adults using sociodemographic, Neighbourhood environmental, and lifestyle factors. International Journal of Environmental Research and Public Health, 19(17), 10977. https://doi.org/10.3390/ijerph191710977
Mareschal, J., Genton, L., Collet, T. H., & Graf, C. (2020). Nutritional intervention to prevent the functional decline in community-dwelling older adults: a systematic review. Nutrients, 12(9), 2820. https://doi.org/10.3390/nu12092820
