의사결정나무분석법을 이용한 노인의 낙상 위험요인들 간의 상호작용 분석

This study was conducted to analyze structured relationship according to an interaction between risk factors for falls in people aged 65 years and older and to propose the basic data required for preventing falls. To achieve this study objective, a secondary analysis was conducted based on the raw data obtained from Community Health Survey (CHS) 2017. The data was analyzed using SPSS 23.0 program for decision tree analysis, and using Chi-squared Automatic Interaction Detection (CHAID), it was developed to a tree-shaped structure. Results are as follows. First, the biggest risk factor for falls was observed to be the number of days sick in bed, followed by sex and then by self-rated health and walking in case of female elders and treatment for arthritis and experience of depression in case of male elders. The highest incidence of falls occurs among those who are sick in bed for 2-31 days, female, and take a walk 0-3 days per week and the lowest incidence of falls occur among those who are sick in bed for 0-1 days, male, and have no experience of treatment for arthritis. Second, the physical activities like walking have reduced incidence of falls. Based on the number of days sick in bed that have a profound effect on the incidence of falls, the risk factors were classified by sex. Therefore, strategies for fall prevention for female elders and male elders should be differentiated and for female elders, walking is recommended.