Therapeutic Science for Rehabilitation (재활치료과학)

Korean Society of Neurological Occupational Therapy (대한신경계작업치료학회)
권호 표지
DOI QR코드

DOI QR Code

Identifying Personal Values Influencing the Lifestyle of Older Adults: Insights From Relative Importance Analysis Using Machine Learning

중고령 노인의 개인적 가치에 따른 라이프스타일 분류: 머신러닝을 활용한 상대적 중요도 분석

  • Lim, Seungju (Dept. of Occupational Therapy, Graduate School, Yonsei University) ;
  • Park, Ji-Hyuk (Dept. of Occupational Therapy, College of Software and Digital Healthcare Convergence, Yonsei University)
  • 임승주 (연세대학교 일반대학원 작업치료학과) ;
  • 박지혁 (연세대학교 소프트웨어디지털헬스케어융합대학 작업치료학과)
  • Received : 2024.01.28
  • Accepted : 2024.03.18
  • Published : 2024.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

Objective : This study aimed to categorize the lifestyles of older adults into two types - healthy and unhealthy, and use machine learning to identify the personal values that influence these lifestyles. Methods : This cross-sectional study targeting middle-aged and older adults (55 years and above) living in local communities in South Korea. Data were collected from 300 participants through online surveys. Lifestyle types were dichotomized by the Yonsei Lifestyle Profile (YLP)-Active, Balanced, Connected, and Diverse (ABCD) responses using latent profile analysis. Personal value information was collected using YLP-Values (YLP-V) and analyzed using machine learning to identify the relative importance of personal values on lifestyle types. Results : The lifestyle of older adults was categorized into healthy (48.87%) and unhealthy (51.13%). These two types showed the most significant difference in social relationship characteristics. Among the machine learning models used in this study, the support vector machine showed the highest classification performance, achieving 96% accuracy and 95% area under the receiver operating characteristic (ROC) curve. The model indicated that individuals who prioritized a healthy diet, sought health information, and engaged in hobbies or cultural activities were more likely to have a healthy lifestyle. Conclusion : This study suggests the need to encourage the expansion of social networks among older adults. Furthermore, it highlights the necessity to comprehensively intervene in individuals' perceptions and values that primarily influence lifestyle adherence.

목적 : 노인의 건강한 삶의 방식으로서 라이프스타일에 대한 연구가 증가하고 있다. 라이프스타일이 개개인의 가치와 삶의 태도를 반영하는 개념임에도 불구하고, 아직까지 개인의 어떠한 가치가 라이프스타일을 건강하게 유도하는지 파악한 연구는 부족한 실정이다. 이에 본 연구는 노인의 라이프스타일 유형을 두 가지로 분류하고, 머신러닝을 활용하여 어떠한 개인적 가치가 건강한 라이프스타일에 우선적으로 작용하는지 파악하고자 한다. 연구방법 : 본 연구는 지역사회에 거주하는 55세 이상 중고령 노인 300명을 대상으로 횡단 연구를 수행하였다. 라이프스타일은 Yonsei Lifestyle Profile-Active, Balanced, Connected, Diverse (YLP-ABCD) 응답을 사용하여 잠재프로파일 분석을 통해 유형화하였다. 라이프스타일 유형을 예측하는 개인적 가치는 YLP-V (Values) 응답을 수집하여, 예측성능이 가장 높은 머신러닝 알고리즘을 선정한 후 상대적 중요도를 파악하였다. 결과 : 잠재프로파일 분석 결과, 라이프스타일은 건강한 라이프스타일 실천형(48.87%), 비실천형(51.13%)으로 분류되었다. 실천형에 속한 중고령 노인은 비실천형에 비해 사회관계가 활발한 특성을 나타내었다. 본 연구에 포함된 머신러닝 알고리즘 중 가장 우수한 성능을 보인 모델은 서포트 벡터 머신으로, 정확도 96%, Receiver Operating Characteristic (ROC) 영역 95%로 나타났다. 본 알고리즘을 바탕으로 개인적 가치의 상대적 중요도를 분석한 결과, 건강한 식단, 건강 매체, 여가활동, 건강 제품 및 머신러닝에 주의를 기울일수록, 해당 가치에 따라 중고령 노인은 건강한 라이프스타일을 실천하는 그룹에 속할 가능성이 큰 것으로 나타났다. 결론 : 본 연구는 중고령 노인의 사회적 관계망을 포함한 건강한 라이프스타일을 유도하기 위해, 건강 식단, 매체, 여가, 제품 및 습관에 대한 가치 향상을 중점적으로 다루는 종합적인 프로그램 및 서비스의 필요성을 시사한다.

Keywords

Acknowledgement

이 논문 또는 저서는 2021년 대한민국 교육부와 한국연구재단의 지원을 받아 수행된 연구임(NRF-2021S1A3A2A02096338)

References

  1. Ahmed, M., Zehou, S., Raza, S. A., Qureshi, M. A., & Yousufi, S. Q. (2020). Impact of CSR and environmental triggers on employee green behavior: The mediating effect of employee well-being. Corporate Social Responsibility and Environmental Management, 27(5), 2225-2239. https://doi.org/10.1002/csr.1960
  2. Akaike, H. (1974). A new look at the statistical model identification. Institute of Electrical and Electronics Engineers Transactions on Automatic Control, 19(6), 716-723. http://doi.org/10.1109/TAC.1974.1100705
  3. Arlot, S., & Celisse, A. (2010). A survey of cross-validation procedures for model selection. Statistics Surveys, 4, 40-79. https://doi.org/10.1214/09-SS054
  4. Bethlehem, J. (2010). Selection bias in web surveys. International Statistical Review, 78(2), 161-188. https://doi.org/10.1111/j.1751-5823.2010.00112.x
  5. Bisong, E. (2019). Building machine learning and deep learning models on google cloud platform: A comprehensive guide for beginners. Apress, 243-250. https://doi.org/10.1007/978-1-4842-4470-8
  6. Cawley, G. C., & Talbot, N. L. C. (2010). On over-fitting in model selection and subsequent selection bias in performance evaluation. Journal of Machine Learning Research, 11, 2079-2107.
  7. Charbuty, B., & Abdulazeez, A. (2021). Classification based on decision tree algorithm for machine learning. Journal of Applied Science and Technology Trends, 2(1), 20-28. https://doi.org/10.38094/jastt20165
  8. Chen, Y., & Feeley, T. H. (2014). Social support, social strain, loneliness, and well-being among older adults: An analysis of the health and retirement study. Journal of Social and Personal Relationships, 31(2), 141-161. https://doi.org/10.1177/0265407513488728
  9. Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273-297. https://doi.org/10.1007/BF00994018
  10. Cudjoe, T. K. M., Roth, D. L., Szanton, S. L., Wolff, J. L., Boyd, C. M., & Thorpe, R. J. (2020). The epidemiology of social isolation: National health and aging trends study. Journals of Gerontology: Series B, 75(1), 107-113. https://doi.org/10.1093/geronb/gby037
  11. Depp, C. A., & Jeste, D. V. (2006). Definitions and predictors of successful aging: A comprehensive review of larger quantitative studies. American Journal of Geriatric Psychiatry, 14(1), 6-20. https://doi.org/10.1097/01.JGP.0000192501.03069.bc
  12. Divine, R. L., & Lepisto, L. (2005). Analysis of the healthy lifestyle consumer. Journal of Consumer Marketing, 22(5), 275-283. https://doi.org/10.1108/07363760510611707
  13. Feurer, M., & Hutter, F. (2019). Hyperparameter optimization. In F. Hutter, L. Kotthoff, & J. Vanschoren (Eds.), Automated machine learning: Methods, systems, challenges (pp. 3-33). Springer. https://doi.org/10.1007/978-3-030-05318-5_1
  14. Ford, E. S., Bergmann, M. M., Boeing, H., Li, C., & Capewell, S. (2012). Healthy lifestyle behaviors and all-cause mortality among adults in the United States. Preventive Medicine, 55(1), 23-27. https://doi.org/10.1016/j.ypmed.2012.04.016
  15. Glas, A. S., Lijmer, J. G., Prins, M. H., Bonsel, G. J., & Bossuyt, P. M. (2003). The diagnostic odds ratio: A single indicator of test performance. Journal of Clinical Epidemiology, 56(11), 1129-1135. https://doi.org/10.1016/S0895-4356(03)00177-X
  16. Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-basfed studies? A comparative analysis of six preconceptions about internet questionnaires. American Psychologist, 59(2), 93-104. https://doi.org/10.1037/0003-066X.59.2.93
  17. Han, Y. (2023). A study of the lifestyle effects on social health of Korean older adults: Focused on structural equation modeling (Master's thesis). Yonsei University.
  18. Hastie, T., Tibshirani, R., & Friedman, J. H. (2009). The elements of statistical learning: Data mining, inference, and prediction. Springer. https://doi.org/10.1007/978-0-387-84858-7
  19. Hruba, R., & Sadilek, T. (2022). Social relations to food, the meaning of the farmers' market and healthy lifestyle: The case of Czechia and Ukraine. Journal of International Food and Agribusiness Marketing, 35(4), 1-19. https://doi.org/10.1080/08974438.2022.2056558
  20. Hu, F. B., Manson, J. E., Stampfer, M. J., Colditz, G., Liu, S., Solomon, C. G., Walter, C., & Willett, W. C. (2001). Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. New England Journal of Medicine, 345(11), 790-797. https://doi.org/10.1056/NEJMoa010492
  21. 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
  22. Jung, T., & Wickrama, K. A. (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and Personality Psychology Compass, 2(1), 302-317. https://doi.org/10.1111/j.1751-9004.2007.00054.x
  23. Knoops, K. T., de Groot, L. C., Kromhout, D., Perrin, A. E., Moreiras-Varela, O., Menotti, A., & Van Staveren, W. A. (2004). Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: The HALE project. Journal of the American Medical Association, 292(12), 1433-1439. https://doi.org/10.1001/jama.292.12.1433
  24. Kohavi, R., & John, G. H. (1997). Wrappers for feature subset selection. Artificial Intelligence, 97(1-2), 273-324. https://doi.org/10.1016/S0004-3702(97)00043-X
  25. Kotsiantis, S. (2011). Feature selection for machine learning classification problems: A recent overview. Artificial Intelligence Review, 42(1), 157-176. https://doi.org/10.1007/s10462-011-9230-1
  26. Lim, S. J., & Park, J. H. (2020). The study of the convergent factors of the lifestyle on the cognitive decline among elderly. Journal of the Korea Convergence Society, 11(8), 229-236. https://doi.org/10.15207/JKCS.2020.11.8.229
  27. Lim, Y. M., Park, K. H., & Park, J. H. (2023). Development of lifestyle evaluation tools based on the values of the elderly: A delphi study. Sage Open, 13(3), 1-9. https://doi.org/10.1177/21582440231194802
  28. Livingston, F. (2005). Implementation of Breiman's random forest machine learning algorithm. ECE591Q Machine Learning Journal Paper, 1-13.
  29. Lo, Y., Mendell, N. R., & Rubin, D. B. (2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767-778. https://doi.org/10.1093/biomet/88.3.767
  30. Loef, M., & Walach, H. (2012). The combined effects of healthy lifestyle behaviors on all cause mortality: A systematic review and meta-analysis. Preventive Medicine, 55(3), 163-170. https://doi.org/10.1016/j.ypmed.2012.06.017
  31. 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
  32. Michel, J. P., & Sadana, R. (2017). "Healthy aging" concepts and measures. Journal of the American Medical Directors Association, 18(6), 460-464. https://doi.org/10.1016/j.jamda.2017.03.008
  33. Moraes, M. B. D., Avgerinou, C., Fukushima, F. B., & Vidal, E. I. (2021). Nutritional interventions for the management of frailty in older adults: Systematic review and meta-analysis of randomized clinical trials. Nutrition Reviews, 79(8), 889-913. https://doi.org/10.1093/nutrit/nuaa101
  34. Muthen, B. (2004). Latent variable analysis. The Sage Handbook of Quantitative Methodology for the Social Sciences, 345(368), 106-109. https://doi.org/10.4135/9781412986311
  35. Navada, A., Ansari, A. N., Patil, S., & Sonkamble, B. A. (2011). Overview of use of decision tree algorithms in machine learning. Proceedings of the 2011 Institute of Electrical and Electronics Engineers Control and System Graduate Research Colloquium, 37-42. https://doi.org/10.1109/ICSGRC.2011.5991826
  36. Nylund, K. L., Asparouhov, T., & Muthen, B. O. (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A monte carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 535-569. http://doi.org/10.1080/10705510701575396
  37. Park, J. H., Park, H. Y., Hong, I., Han, D. S., Lim, Y. M., Kim, A. R., Nam, S., Park, K. H., Lim, S., Bae, S., & Jin, Y. (2023). Conceptual model of establishing lifestyle (Lifestyle-DEPER [Decision, execution, personal factor, environment, resources]) and lifestyle intervention strategies. Therapeutic Science for Rehabilitation, 12(4), 9-22. https://doi.org/10.22683/tsnr.2023.12.4.009
  38. Park, K. H., Yang, M. A., Won, K. A., & Park, J. H. (2021). Predicting healthy lifestyle patterns in older community dwelling adults: A latent profile analysis. Therapeutic Science for Rehabilitation, 10(2), 75-93. https://doi.org/10.22683/tsnr.2022.11.2.075
  39. Park, K. H., Yoo, E. Y., Kim, J., Hong, I., Lee, J. S., & Park, J. H. (2021). Applying latent profile analysis to identify lifestyle profiles and their association with loneliness and quality of life among community-dwelling middle-and older-aged adults in South Korea. International Journal of Environmental Research and Public Health, 18(23), 12374. https://doi.org/10.3390/ijerph182312374
  40. Pavlyshenko, B. (2016). Machine learning, linear and bayesian models for logistic regression in failure detection problems. Proceedings of the 2016 Institute of Electrical and Electronics Engineers International Conference on Big Data, 2046-2050. http://doi.org/10.1109/bigdata.2016.7840828
  41. Plummer, J. T. (1974). The concept and application of life style segmentation: The combination of two useful concepts provides a unique and important view of the market. Journal of Marketing, 38(1), 33-37. https://doi.org/10.2307/1250164
  42. Polidori, M. C., Nelles, G., & Pientka, L. (2010). Prevention of dementia: Focus on lifestyle. International Journal of Alzheimer's Disease, 2010, 1-9. https://doi.org/10.4061/2010/393579.
  43. Safari, S., Baratloo, A., Elfil, M., & Negida, A. (2016). Evidence based emergency medicine: Part 5 receiver operating curve and area under the curve. Emergency, 4(2), 111-113. https://doi.org/10.22037/emergency.v4i2.11903
  44. Salam, M. A., Azar, A. T., Elgendy, M. S., & Fouad, K. M. (2021). The effect of different dimensionality reduction techniques on machine learning overfitting problem. International Journal of Advanced Computer Science and Applications, 12(4), 641-655. https://doi.org/10.14569/IJACSA.2021.0120480
  45. Schonlau, M., & Zou, R. Y. (2020). The random forest algorithm for statistical learning. Stata Journal, 20(1), 3-29. https://doi.org/10.1177/1536867X20909688
  46. Schwarz, G. (1978). Estimating the dimension of a model. Annals of Statistics, 461-464. https://doi.org/10.1214/aos/1176344136
  47. Somvanshi, M., Chavan, P., Tambade, S., & Shinde, S. V. (2016). A review of machine learning techniques using decision tree and support vector machine. Proceedings of the 2016 International Conference on Computing Communication Control and Automation, 1-7. https://doi.org/10.1109/iccubea.2016.7860040
  48. Spring, B., Moller, A. C., & Coons, M. J. (2012). Multiple health behaviours: Overview and implications. Journal of Public Health, 34(suppl_1), i3-i10. https://doi.org/10.1093/pubmed/fdr111
  49. Sun, Q., Townsend, M. K., Okereke, O. I., Franco, O. H., Hu, F. B., & Grodstein, F. (2010). Physical activity at midlife in relation to successful survival in women at age 70 years or older. Archives of Internal Medicine, 170(2), 194-201. https://doi.org/10.1001/archinternmed.2009.503
  50. Uemura, K., Yamada, M., & Okamoto, H. (2021). The effectiveness of an active learning program in promoting a healthy lifestyle among older adults with low health literacy: A randomized controlled trial. Gerontology, 67(1), 25-35. https://doi.org/10.1159/000511357
  51. Wang, L. (Ed.). (2005). Support vector machines: Theory and applications (STUDFUZZ, Vol. 177). Springer.
  52. Wang, R., Chaudhari, P., & Davatzikos, C. (2023). Reply to Holm et al.: Careful training is a good practice. Proceedings of the National Academy of Sciences, 120(23), e2305855120. https://doi.org/10.1073/pnas.2305855120
  53. Yang, M. A., Lim, S., Han, D., & Park, J. (2020). A policy proposal for improving the quality of life for the middle-aged people: Focusing on lifestyle. Journal of the Korean Gerontology Society, 40(3), 531-541. https://doi.org/10.31888/JKGS.2020.40.3.531
  54. Zhou, Z. H. (2021). Machine learning. Springer Nature.