A Model of Fall Correlates in Community-Dwelling Older Adults

Purpose: The aims of this study were to investigate the incidences of falling and test a model that explains the relationships between the correlates of falling among the community-dwelling elderly by using a cross-sectional survey. Method: Purposive samples of older adults were living in 42 communities in Taiwan were selected to participate in the fall prevention program, A total of 904 subjects aged 65 years and above. Based upon PRECEDE model and used structured questionnaires, including personal profiles and scales to obtain data on the knowledge, health beliefs, self-efficacy of fall prevention, environmental support, and physical function. logistic regression, and structural equation modeling analysis were performed as the main statistical methods.Results: 16.92% of the subjects had suffered from falls in the past year. Logistic regression identified the following significant predictors of falls were: knowledge of fall prevention, self-efficacy of fall prevention, supportive environment, reaction time, 20-meters walking test, living alone, use of assistive devices and chronic disease. Discriminant analysis showed that classified the percentage of the overall prediction was 82.9%. the results indicated a good model fit, and parameter estimates and the direction of the relationship were consistent with the theory. The significant path coefficients with “reinforcing factor,” followed by “prediposing factor,” and finally “enabling factor.” Conclusion: The above model had a multifactor structure. This study is based upon PRECEDE model framework and uses SEM to explain the fall model. The results indicated a good model fit. Thus, future programs and policies can apply this model to prevent falls.