M Clare Robertson1, A John Campbell, Peter Herbison. 1. Department of Medical and Surgical Sciences, University of Otago Medical School, P.O. Box 913, Dunedin, New Zealand. clare.robertson@stonebow.otago.ac.nz
Abstract
BACKGROUND: Many different and sometimes inappropriate statistical techniques have been used to analyze the results of randomized controlled trials of falls prevention programs for elderly people. This makes comparison of the efficacy of particular interventions difficult. METHODS: We used raw data from two randomized controlled trials of a home exercise program to compare the number of falls in the exercise and control groups during the trials. We developed two different survival analysis models (Andersen-Gill and marginal Cox regression) and a negative binomial regression model for each trial. These techniques a) allow for the fact that falls are frequent, recurrent events with a non-normal distribution; b) adjust for the follow-up time of individual participants; and c) allow the addition of covariates. RESULTS: In one trial, the three different statistical techniques gave surprisingly similar results for the efficacy of the intervention but, in a second trial, underlying assumptions were violated for the two Cox regression models. Negative binomial regression models were easier to use. CONCLUSION: We recommend negative binomial regression models for evaluating the efficacy of falls prevention programs.
RCT Entities:
BACKGROUND: Many different and sometimes inappropriate statistical techniques have been used to analyze the results of randomized controlled trials of falls prevention programs for elderly people. This makes comparison of the efficacy of particular interventions difficult. METHODS: We used raw data from two randomized controlled trials of a home exercise program to compare the number of falls in the exercise and control groups during the trials. We developed two different survival analysis models (Andersen-Gill and marginal Cox regression) and a negative binomial regression model for each trial. These techniques a) allow for the fact that falls are frequent, recurrent events with a non-normal distribution; b) adjust for the follow-up time of individual participants; and c) allow the addition of covariates. RESULTS: In one trial, the three different statistical techniques gave surprisingly similar results for the efficacy of the intervention but, in a second trial, underlying assumptions were violated for the two Cox regression models. Negative binomial regression models were easier to use. CONCLUSION: We recommend negative binomial regression models for evaluating the efficacy of falls prevention programs.
Authors: L E P P Teixeira; K N G Silva; A M Imoto; T J P Teixeira; A H Kayo; R Montenegro-Rodrigues; M S Peccin; V F M Trevisani Journal: Osteoporos Int Date: 2009-06-27 Impact factor: 4.507
Authors: A John Campbell; M Clare Robertson; Steven J La Grow; Ngaire M Kerse; Gordon F Sanderson; Robert J Jacobs; Dianne M Sharp; Leigh A Hale Journal: BMJ Date: 2005-09-23
Authors: Lee A Jennings; David B Reuben; Sung-Bou Kim; Emmett Keeler; Carol P Roth; David S Zingmond; Neil S Wenger; David A Ganz Journal: Am J Manag Care Date: 2015-09-01 Impact factor: 2.229
Authors: David Sparrow; Tamara R DeAngelis; Kathryn Hendron; Cathi A Thomas; Marie Saint-Hilaire; Terry Ellis Journal: J Neurol Phys Ther Date: 2016-01 Impact factor: 3.649
Authors: Catherine M Dean; Chris Rissel; Michelle Sharkey; Catherine Sherrington; Robert G Cumming; Ruth N Barker; Stephen R Lord; Sandra D O'Rourke; Catherine Kirkham Journal: BMC Neurol Date: 2009-07-22 Impact factor: 2.474
Authors: Colleen G Canning; Cathie Sherrington; Stephen R Lord; Victor S C Fung; Jacqueline C T Close; Mark D Latt; Kirsten Howard; Natalie E Allen; Sandra D O'Rourke; Susan M Murray Journal: BMC Neurol Date: 2009-01-22 Impact factor: 2.474
Authors: Frances A Batchelor; Keith D Hill; Shylie F Mackintosh; Catherine M Said; Craig H Whitehead Journal: BMC Neurol Date: 2009-03-31 Impact factor: 2.474