OBJECTIVE: To assess the impact, in terms of statistical power and bias of treatment effect, of approaches to dealing with missing data in randomized controlled trials of rheumatoid arthritis with radiographic outcomes. METHODS: We performed a simulation study. The missingness mechanisms we investigated copied the process of withdrawal from trials due to lack of efficacy. We compared 3 methods of managing missing data: all available data (case-complete), last observation carried forward (LOCF), and multiple imputation. Data were then analyzed by classic t-test (comparing the mean absolute change between baseline and final visit) or F test (estimation of treatment effect with repeated measurements by a linear mixed-effects model). RESULTS: With a missing data rate close to 15%, the treatment effect was underestimated by 18% as estimated by a linear mixed-effects model with a multiple imputation approach to missing data. This bias was lower than that obtained with the case-complete approach (-25%) or LOCF approach (-35%). This statistical approach (combination of multiple imputation and mixed-effects analysis) was moreover associated with a power of 70% (for a 90% nominal level), whereas LOCF was associated with a power of 55% and a case-complete power of 58%. Analysis with the t-test gave qualitatively equivalent but poorer quality results, except when multiple imputation was applied. CONCLUSION: Our simulation study demonstrated multiple imputation, offering the smallest bias in treatment effect and the highest power. These results can help in planning trials, especially in choosing methods of imputation and data analysis.
OBJECTIVE: To assess the impact, in terms of statistical power and bias of treatment effect, of approaches to dealing with missing data in randomized controlled trials of rheumatoid arthritis with radiographic outcomes. METHODS: We performed a simulation study. The missingness mechanisms we investigated copied the process of withdrawal from trials due to lack of efficacy. We compared 3 methods of managing missing data: all available data (case-complete), last observation carried forward (LOCF), and multiple imputation. Data were then analyzed by classic t-test (comparing the mean absolute change between baseline and final visit) or F test (estimation of treatment effect with repeated measurements by a linear mixed-effects model). RESULTS: With a missing data rate close to 15%, the treatment effect was underestimated by 18% as estimated by a linear mixed-effects model with a multiple imputation approach to missing data. This bias was lower than that obtained with the case-complete approach (-25%) or LOCF approach (-35%). This statistical approach (combination of multiple imputation and mixed-effects analysis) was moreover associated with a power of 70% (for a 90% nominal level), whereas LOCF was associated with a power of 55% and a case-complete power of 58%. Analysis with the t-test gave qualitatively equivalent but poorer quality results, except when multiple imputation was applied. CONCLUSION: Our simulation study demonstrated multiple imputation, offering the smallest bias in treatment effect and the highest power. These results can help in planning trials, especially in choosing methods of imputation and data analysis.
Authors: Patrick E McKnight; Shelley Kasle; Scott Going; Isidro Villanueva; Michelle Cornett; Josh Farr; Jill Wright; Clara Streeter; Alex Zautra Journal: Arthritis Care Res (Hoboken) Date: 2010-01-15 Impact factor: 4.794
Authors: Benjamin W Van Voorhees; Karen Vanderplough-Booth; Joshua Fogel; Tracy Gladstone; Carl Bell; Scott Stuart; Jackie Gollan; Nathan Bradford; Rocco Domanico; Blake Fagan; Ruth Ross; Jon Larson; Natalie Watson; Dave Paunesku; Stephanie Melkonian; Sachiko Kuwabara; Tim Holper; Nicholas Shank; Donald Saner; Amy Butler; Amy Chandler; Tina Louie; Cynthia Weinstein; Shannon Collins; Melinda Baldwin; Abigail Wassel; Mark A Reinecke Journal: J Can Acad Child Adolesc Psychiatry Date: 2008-11
Authors: Benjamin W Van Voorhees; Joshua Fogel; Mark A Reinecke; Tracy Gladstone; Scott Stuart; Jackie Gollan; Nathan Bradford; Rocco Domanico; Blake Fagan; Ruth Ross; Jon Larson; Natalie Watson; Dave Paunesku; Stephanie Melkonian; Sachiko Kuwabara; Tim Holper; Nicholas Shank; Donald Saner; Amy Butler; Amy Chandler; Tina Louie; Cynthia Weinstein; Shannon Collins; Melinda Baldwin; Abigail Wassel; Karin Vanderplough-Booth; Jennifer Humensky; Carl Bell Journal: J Dev Behav Pediatr Date: 2009-02 Impact factor: 2.225
Authors: N Jeanne Conley; Patricia B Pavlinac; Brandon L Guthrie; Romel D Mackelprang; Anthony N Muiru; Robert Y Choi; Rose Bosire; Ann Gatuguta; Carey Farquhar Journal: PLoS One Date: 2012-08-21 Impact factor: 3.240
Authors: J Haxby Abbott; M Clare Robertson; Joanne E McKenzie; G David Baxter; Jean-Claude Theis; A John Campbell Journal: Trials Date: 2009-02-08 Impact factor: 2.279
Authors: Eveline Nüesch; Sven Trelle; Stephan Reichenbach; Anne W S Rutjes; Elizabeth Bürgi; Martin Scherer; Douglas G Altman; Peter Jüni Journal: BMJ Date: 2009-09-07
Authors: Arthur Kavanaugh; Roy M Fleischmann; Paul Emery; Hartmut Kupper; Laura Redden; Benoit Guerette; Sourav Santra; Josef S Smolen Journal: Ann Rheum Dis Date: 2012-05-05 Impact factor: 19.103
Authors: Cheryl Barnabe; Ye Sun; Gilles Boire; Carol A Hitchon; Boulos Haraoui; J Carter Thorne; Diane Tin; Désirée van der Heijde; Jeffrey R Curtis; Shahin Jamal; Janet E Pope; Edward C Keystone; Susan Bartlett; Vivian P Bykerk Journal: PLoS One Date: 2015-08-24 Impact factor: 3.240