Jye Wang1, Wender Lin2, Ling-Hui Chang3. 1. Department of Health Care Administration, Chang Jung Christian University, No.1 Changda Rd., Gueiren District, Tainan City 71101, Taiwan, ROC. Electronic address: jyewang@mail.cjcu.edu.tw. 2. Department of Health Care Administration, Chang Jung Christian University, No.1 Changda Rd., Gueiren District, Tainan City 71101, Taiwan, ROC. Electronic address: vincelin@mail.cjcu.edu.tw. 3. Department of Occupational Therapy & Institute of Allied Health Sciences, College of Medicine, National Cheng Kung University, No.1 University Road, Tainan City 701, Taiwan, ROC. Electronic address: lhchang@mail.ncku.edu.tw.
Abstract
BACKGROUND: The Vulnerable Elders Survey-13 (VES-13) has been used as a screening tool to identify vulnerable community-dwelling older persons for more in-depth assessment and targeted interventions. Although many studies supported its use in different populations, few have addressed Asian populations. The optimal scaling system for the VES-13 in predicting health outcomes also has not been adequately tested. This study (1) assesses the applicability of the VES-13 to predict the mortality of community-dwelling older persons in Taiwan, (2) identifies the best scaling system for the VES-13 in predicting mortality using generalized additive models (GAMs), and (3) determines whether including covariates, such as socio-demographic factors and common geriatric syndromes, improves model fitting. MATERIALS AND METHODS: This retrospective longitudinal cohort study analyzed the data of 2184 community-dwelling persons 65 years old or older from the 2003 wave of the national-wide Taiwan Longitudinal Study on Aging. Cox proportional hazards models and Generalized Additive Models (GAMs) were used. RESULTS: The VES-13 significantly predicted the mortality of Taiwan's community-dwelling elders. A one-point increase in the VES-13 score raised the risk of death by 26% (hazard ratio, 1.26; 95% confidence interval, 1.21-1.32). The hazard ratio of death increased linearly with each additional VES-13 score point, suggesting that using a continuous scale is appropriate. Inclusion of socio-demographic factors and geriatric syndromes improved the model-fitting. CONCLUSIONS: The VES-13 is appropriate for an Asian population. VES-13 scores linearly predict the mortality of this population. Adjusting the weighting of the physical activity items may improve the performance of the VES-13.
BACKGROUND: The Vulnerable Elders Survey-13 (VES-13) has been used as a screening tool to identify vulnerable community-dwelling older persons for more in-depth assessment and targeted interventions. Although many studies supported its use in different populations, few have addressed Asian populations. The optimal scaling system for the VES-13 in predicting health outcomes also has not been adequately tested. This study (1) assesses the applicability of the VES-13 to predict the mortality of community-dwelling older persons in Taiwan, (2) identifies the best scaling system for the VES-13 in predicting mortality using generalized additive models (GAMs), and (3) determines whether including covariates, such as socio-demographic factors and common geriatric syndromes, improves model fitting. MATERIALS AND METHODS: This retrospective longitudinal cohort study analyzed the data of 2184 community-dwelling persons 65 years old or older from the 2003 wave of the national-wide Taiwan Longitudinal Study on Aging. Cox proportional hazards models and Generalized Additive Models (GAMs) were used. RESULTS: The VES-13 significantly predicted the mortality of Taiwan's community-dwelling elders. A one-point increase in the VES-13 score raised the risk of death by 26% (hazard ratio, 1.26; 95% confidence interval, 1.21-1.32). The hazard ratio of death increased linearly with each additional VES-13 score point, suggesting that using a continuous scale is appropriate. Inclusion of socio-demographic factors and geriatric syndromes improved the model-fitting. CONCLUSIONS: The VES-13 is appropriate for an Asian population. VES-13 scores linearly predict the mortality of this population. Adjusting the weighting of the physical activity items may improve the performance of the VES-13.