Development of Prediction Model for Endocrine Disorders in the Korean Elderly Using CART Algorithm

• Published in: International journal of advanced computer science & applications Vol. 6; no. 9
• Main Author: Byeon, Haewon
• Format: Journal Article
• Language: English
• Published: West Yorkshire Science and Information (SAI) Organization Limited 2015
• Subjects: Algorithms; Dependent variables; Disorders; Education; Family income; Income; Older people; Prediction models; Regression analysis; Risk analysis
• ISSN: 2158-107X; 2156-5570; 2156-5570
• DOI: 10.14569/IJACSA.2015.060916

The aim of the present cross-sectional study was to analyze the factors that affect endocrine disorders in the Korean elderly. The data were taken from the A Study of the Seoul Welfare Panel Study 2010. The subjects were 2111 people (879 males, 1,232 females) aged 60 and older living in the community. The dependent variable was defined as the prevalence of endocrine disorders. The explanatory variables were gender, level of education, household income, employment status, marital status, drinking, smoking, BMI, subjective health status, physical activity, experience of stress, and depression. In the Classification and Regression Tree (CART) algorithm analysis, subjective health status, BMI, education level, and household income were significantly associated with endocrine disorders in the Korean elderly. The most preferentially involved predictor was subjective health status. The development of guidelines and health education to prevent endocrine disorders is required for taking multiple risk factors into account.