The purpose of this study was to test the hypothetical model designed to explain fall prevention behavior of community residing elderly with osteoarthritis. This hypothetical model was based on the health belief model and literature review. The exogenous variables of the hypothetical model were perceived fall risk and fall knowledge. The endogenous variables of the hypothetical model were perceived benefit, perceived barrier, self-efficacy of fall prevention and fall prevention behavior.
The data were collected from July 10 to August 30, 2012. through individual interviews with the structural questionnaire on 208 elderly with osteoarthritis living in the community. Among 208 copies of questionnaire, 200 copies without any missing data were used for testing the hypothetical model
The collected data were analyzed using SPSSWIN 20,0 and AMOS 20.0 program for descriptive statistical analysis and covariance structural analysis.
The results were as follows.
1) The hypothetical model showed a good fit(()=14(<.044), Q(/ df)=2.064, GFI=0.980， TLI=0.918, CFI= 0.973, PNFI=0.317, RMSEA=0.073 to the empirical data.
2) The 13 paths established in this research model and 9 paths were statistically significant. 5 paths(perceived fall risk → perceived benefit, perceived fall risk → perceived barrier, perceived fall risk→fall prevention behavior, perceived benefit → efficacy of fall prevention, efficacy of fall prevention → fall prevention behavior) had a significant direct effect. and 1 paths (perceived fall risk → efficacy of fall prevention) had significant total effects. Fall knowledge had a direct, indirect, total effect on efficacy of fall prevention. Fall knowledge and perceived benefit had a indirect, total effect on fall prevention behavior.
3) Factors that influence the fall prevention behavior were identified, as perceived fall risk, fall knowledge, perceived benefit and efficacy of fall prevention.
(1) Significant variable to perceived benefit was perceived fall risk and this variable explained 5% of variance of perceived benefit. (2) Significant variable to perceived barrier was perceived fall risk and this variable had 26% explantion for perceived barrier. (3) Significant variables to efficacy of fall prevention were perceived fall risk and fall knowledge and perceived benefit and this variables had 13% explantion for efficacy of fall prevention.
(4) Significant variables to fall prevention behavior were perceived fall risk and fall knowledge and perceived benefit and efficacy of fall prevention this variables had 33% explantion for fall prevention behavior.
4) Hypothetical Model with Control Variables showed a good fit(()=31.54(.035), Q(/df)=1.66, GFI(0.967), TLI(0.917), CFI(0.956), PNFI(0.476), RMSEA(0.058)) to the empirical data.
5) After hypothetical model with Control Variables, factors that influence the fall prevention behavior were identified perceived fall risk and fall knowledge, perceived benefit and efficacy of fall prevention and these variables together accounted for 36% of variance of fall prevention behavior.
In conclusion, the hypothetical model of this study was confirmed to be proper in explaning and predicting of fall prevention behavior of elderly with osteoarthritis living in the community. These findings suggest the need to develop an effective nursing intervention for fall prevention behaviors of communiry residing elderly with osteoarthritis by enhancing self-efficacy of fall prevention behaviors and controlling their fear of falling.