Takaaki Ikeda1,2, Upul Cooray3, Masanori Hariyama4, Jun Aida5,6, Katsunori Kondo7,8, Masayasu Murakami9, Ken Osaka3. 1. Department of Health Policy Science, Graduate School of Medical Science, Yamagata University, Yamagata, Yamagata, Japan. tikeda@med.id.yamagata-u.ac.jp. 2. Department of International and Community Oral Health, Graduate School of Dentistry, Tohoku University, Sendai, Miyagi, Japan. tikeda@med.id.yamagata-u.ac.jp. 3. Department of International and Community Oral Health, Graduate School of Dentistry, Tohoku University, Sendai, Miyagi, Japan. 4. Intelligent Integrated Systems Laboratory, Graduate School of Information Sciences, Tohoku University, Sendai, Miyagi, Japan. 5. Department of Oral Health Promotion, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan. 6. Division for Regional Community Development, Liaison Center for Innovative Dentistry, Graduate School of Dentistry, Tohoku University, Sendai, Miyagi, Japan. 7. Department of Social Preventive Medical Sciences, Center for Preventive Medical Sciences, Chiba University, Chiba, Chiba, Japan. 8. Department of Gerontological Evaluation, Center for Gerontology and Social Science, National Center for Geriatrics and Gerontology, Obu, Aichi, Japan. 9. Department of Health Policy Science, Graduate School of Medical Science, Yamagata University, Yamagata, Yamagata, Japan.
Abstract
BACKGROUND: Adverse health effects resulting from falls are a major public health concern. Although studies have identified risk factors for falls, none have examined long-term prediction of fall risk. Furthermore, recent evidence suggests that there are additional risk factors, such as psychosocial factors. OBJECTIVE: In this 3-year longitudinal study, we evaluated a predictive model for risk of fall among community-dwelling older adults using machine learning methods. DESIGN: A 3-year follow-up prospective longitudinal study (from 2010 to 2013). SETTING: Twenty-four municipalities in nine of the 47 prefectures (provinces) of Japan. PARTICIPANTS: Community-dwelling individuals aged ≥65 years who were functionally independent at baseline (n = 61,883). METHODS: The baseline survey was conducted from August 2010 to January 2012, and the follow-up survey was conducted from October to December 2013. Both surveys were conducted involving self-reported questionnaires. The measured outcome at the follow-up survey was self-reported multiple falls during the previous year. The 142 variables included in the baseline survey were regarded as candidate predictors. The random-forest-based Boruta algorithm was used to select predictors, and the eXtreme Gradient Boosting algorithm with 10 repetitions of nested k-fold cross-validation was used for modeling and model evaluation. Furthermore, we used shapley additive explanations to gain insight into the behavior of the prediction model. KEY RESULTS: Fourteen out of 142 candidate features were selected as predictors. Among these predictors, experience of falling as of the baseline survey was the most important feature, followed by self-rated health and age. Moreover, sense of coherence was newly identified as a risk factor for falls. CONCLUSIONS: This study suggests that machine learning tools can be adapted to explore new associative factors, make accurate predictions, and provide actionable insights for fall prevention strategies.
BACKGROUND: Adverse health effects resulting from falls are a major public health concern. Although studies have identified risk factors for falls, none have examined long-term prediction of fall risk. Furthermore, recent evidence suggests that there are additional risk factors, such as psychosocial factors. OBJECTIVE: In this 3-year longitudinal study, we evaluated a predictive model for risk of fall among community-dwelling older adults using machine learning methods. DESIGN: A 3-year follow-up prospective longitudinal study (from 2010 to 2013). SETTING: Twenty-four municipalities in nine of the 47 prefectures (provinces) of Japan. PARTICIPANTS: Community-dwelling individuals aged ≥65 years who were functionally independent at baseline (n = 61,883). METHODS: The baseline survey was conducted from August 2010 to January 2012, and the follow-up survey was conducted from October to December 2013. Both surveys were conducted involving self-reported questionnaires. The measured outcome at the follow-up survey was self-reported multiple falls during the previous year. The 142 variables included in the baseline survey were regarded as candidate predictors. The random-forest-based Boruta algorithm was used to select predictors, and the eXtreme Gradient Boosting algorithm with 10 repetitions of nested k-fold cross-validation was used for modeling and model evaluation. Furthermore, we used shapley additive explanations to gain insight into the behavior of the prediction model. KEY RESULTS: Fourteen out of 142 candidate features were selected as predictors. Among these predictors, experience of falling as of the baseline survey was the most important feature, followed by self-rated health and age. Moreover, sense of coherence was newly identified as a risk factor for falls. CONCLUSIONS: This study suggests that machine learning tools can be adapted to explore new associative factors, make accurate predictions, and provide actionable insights for fall prevention strategies.
Authors: Francesco Landi; Andrea Russo; Rosa Liperoti; Matteo Tosato; Christian Barillaro; Marco Pahor; Roberto Bernabei; Graziano Onder Journal: J Am Med Dir Assoc Date: 2010-04-03 Impact factor: 4.669
Authors: Silvia G R Neri; Juliana S Oliveira; Amabile B Dario; Ricardo M Lima; Anne Tiedemann Journal: J Gerontol A Biol Sci Med Sci Date: 2020-04-17 Impact factor: 6.053
Authors: Annie M Racine; Douglas Tommet; Madeline L D'Aquila; Tamara G Fong; Yun Gou; Patricia A Tabloski; Eran D Metzger; Tammy T Hshieh; Eva M Schmitt; Sarinnapha M Vasunilashorn; Lisa Kunze; Kamen Vlassakov; Ayesha Abdeen; Jeffrey Lange; Brandon Earp; Bradford C Dickerson; Edward R Marcantonio; Jon Steingrimsson; Thomas G Travison; Sharon K Inouye; Richard N Jones Journal: J Gen Intern Med Date: 2020-10-19 Impact factor: 5.128