作者: M. C. Robertson , A. J. Campbell , P. Herbison
关键词:
摘要: BACKGROUND: Many different and sometimes inappropriate statistical techniques have been used to analyze the results of randomized controlled trials falls prevention programs for elderly people. This makes comparison efficacy particular interventions difficult. METHODS: We raw data from two a home exercise program compare number in control groups during trials. developed survival analysis models (Andersen-Gill marginal Cox regression) negative binomial regression model each trial. These a) allow fact that are frequent, recurrent events with non-normal distribution; b) adjust follow-up time individual participants; c) addition covariates. RESULTS: In one trial, three gave surprisingly similar intervention but, second underlying assumptions were violated models. Negative easier use. CONCLUSION: recommend evaluating programs.