OBJECTIVE: To assess the added value of multiple imputation (MI) of missing repeated outcomes measures in longitudinal data sets analyzed with linear mixed-effects (LME) models. STUDY DESIGN AND SETTING: Data were used from a trial on the effects of Rosuvastatin on rate of change in carotid intima-media thickness (CIMT). The reference treatment effect was derived from a complete data set. Scenarios and proportions of missing values in CIMT measurements were applied and LME analyses were used before and after MI. The added value of MI, in terms of bias and precision, was assessed using the mean-squared error (MSE) of the treatment effects and coverage of the 95% confidence interval. RESULTS: The reference treatment effect was -0.0177 mm/y. The MSEs for LME analysis without and with MI were similar in scenarios with up to 40% missing values. Coverage was large in all scenarios and was similar for LME with and without MI. CONCLUSION: Our study empirically shows that MI of missing end point data before LME analyses does not increase precision in the estimated rate of change in the end point. Hence, MI had no added value in this setting and standard LME modeling remains the method of choice.
OBJECTIVE: To assess the added value of multiple imputation (MI) of missing repeated outcomes measures in longitudinal data sets analyzed with linear mixed-effects (LME) models. STUDY DESIGN AND SETTING: Data were used from a trial on the effects of Rosuvastatin on rate of change in carotid intima-media thickness (CIMT). The reference treatment effect was derived from a complete data set. Scenarios and proportions of missing values in CIMT measurements were applied and LME analyses were used before and after MI. The added value of MI, in terms of bias and precision, was assessed using the mean-squared error (MSE) of the treatment effects and coverage of the 95% confidence interval. RESULTS: The reference treatment effect was -0.0177 mm/y. The MSEs for LME analysis without and with MI were similar in scenarios with up to 40% missing values. Coverage was large in all scenarios and was similar for LME with and without MI. CONCLUSION: Our study empirically shows that MI of missing end point data before LME analyses does not increase precision in the estimated rate of change in the end point. Hence, MI had no added value in this setting and standard LME modeling remains the method of choice.
Authors: Kumaravel Rajakumar; Charity G Moore; Arshad T Khalid; Abbe N Vallejo; Mohamed A Virji; Michael F Holick; Susan L Greenspan; Silva Arslanian; Steven E Reis Journal: Am J Clin Nutr Date: 2020-04-01 Impact factor: 7.045
Authors: Ângela Jornada Ben; Johanna M van Dongen; Mohamed El Alili; Martijn W Heymans; Jos W R Twisk; Janet L MacNeil-Vroomen; Maartje de Wit; Susan E M van Dijk; Teddy Oosterhuis; Judith E Bosmans Journal: Eur J Health Econ Date: 2022-09-26
Authors: Kathleen M Carroll; Brian D Kiluk; Charla Nich; Elise E DeVito; Suzanne Decker; Donna LaPaglia; Dianne Duffey; Theresa A Babuscio; Samuel A Ball Journal: Drug Alcohol Depend Date: 2014-01-31 Impact factor: 4.492
Authors: Albert H A Mazairac; G Ardine de Wit; Muriel P C Grooteman; E Lars Penne; Neelke C van der Weerd; Claire H den Hoedt; Renée Lévesque; Marinus A van den Dorpel; Menso J Nubé; Piet M ter Wee; Michiel L Bots; Peter J Blankestijn Journal: Clin J Am Soc Nephrol Date: 2012-11-02 Impact factor: 8.237
Authors: Electra D Paskett; Ryan D Baltic; Gregory S Young; Mira L Katz; Samuel M Lesko; Kelly H Webber; Karen A Roberto; Eugene J Lengerich; Nancy E Schoenberg; Stephenie K Kennedy; Scherezade Mama; Courtney C Midkiff; Mark B Dignan Journal: Cancer Epidemiol Biomarkers Prev Date: 2018-10-18 Impact factor: 4.254
Authors: Katharina L Wiest; Victoria J Asphaug; Kathryn E Carr; Emily A Gowen; Timothy T Hartnett Journal: Int J Ther Massage Bodywork Date: 2015-03-01
Authors: François Boudreau; Gilles R Dagenais; Hein de Vries; Michel Jean Louis Walthouwer; José Côté; Ginette Turbide; Anne-Sophie Bourlaud; Paul Poirier Journal: Health Psychol Behav Med Date: 2020-12-08