Yan Ma1, Madhu Mazumdar, Stavros G Memtsoudis. 1. Research Division, Hospital for Special Surgery, Weill Medical College of Cornell University, New York, NY 10021, USA. may@hss.edu
Abstract
BACKGROUND AND OBJECTIVES: Research in the field of anesthesiology relies heavily on longitudinal designs for answering questions about long-term efficacy and safety of various anesthetic and pain regimens. Yet, anesthesiology research is lagging in the use of advanced statistical methods for analyzing longitudinal data. The goal of this article was to increase awareness of the advantages of modern statistical methods and promote their use in anesthesia research. METHODS: Here we introduce 2 modern and advanced statistical methods for analyzing longitudinal data: the generalized estimating equations (GEE) and mixed-effects models (MEM). These methods were compared with the conventional repeated-measures analysis of variance (RM-ANOVA) through a clinical example with 2 types of end points (continuous and binary). In addition, we compared GEE and MEM to RM-ANOVA through a simulation study with varying sample sizes, varying number of repeated measures, and scenarios with and without missing data. RESULTS: In the clinical study, the 3 methods are found to be similar in terms of statistical estimation, whereas the parameter interpretations are somewhat different. The simulation study shows that the methods of GEE and MEM are more efficient in that they are able to achieve higher power with smaller sample size or lower number of repeated measurements in both complete and missing data scenarios. CONCLUSIONS: Based on their advantages over RM-ANOVA, GEE and MEM should be strongly considered for the analysis of longitudinal data. In particular, GEE should be used to explore overall average effects, and MEM should be used when subject-specific effects (in addition to overall average effects) are of primary interest.
BACKGROUND AND OBJECTIVES: Research in the field of anesthesiology relies heavily on longitudinal designs for answering questions about long-term efficacy and safety of various anesthetic and pain regimens. Yet, anesthesiology research is lagging in the use of advanced statistical methods for analyzing longitudinal data. The goal of this article was to increase awareness of the advantages of modern statistical methods and promote their use in anesthesia research. METHODS: Here we introduce 2 modern and advanced statistical methods for analyzing longitudinal data: the generalized estimating equations (GEE) and mixed-effects models (MEM). These methods were compared with the conventional repeated-measures analysis of variance (RM-ANOVA) through a clinical example with 2 types of end points (continuous and binary). In addition, we compared GEE and MEM to RM-ANOVA through a simulation study with varying sample sizes, varying number of repeated measures, and scenarios with and without missing data. RESULTS: In the clinical study, the 3 methods are found to be similar in terms of statistical estimation, whereas the parameter interpretations are somewhat different. The simulation study shows that the methods of GEE and MEM are more efficient in that they are able to achieve higher power with smaller sample size or lower number of repeated measurements in both complete and missing data scenarios. CONCLUSIONS: Based on their advantages over RM-ANOVA, GEE and MEM should be strongly considered for the analysis of longitudinal data. In particular, GEE should be used to explore overall average effects, and MEM should be used when subject-specific effects (in addition to overall average effects) are of primary interest.
Authors: Steven H Renes; Geert J van Geffen; Harald C Rettig; Mathieu J Gielen; Gert J Scheffer Journal: Reg Anesth Pain Med Date: 2010 Nov-Dec Impact factor: 6.288
Authors: Simon J Parrington; Dermot O'Donnell; Vincent W S Chan; Danielle Brown-Shreves; Rajeev Subramanyam; Melody Qu; Richard Brull Journal: Reg Anesth Pain Med Date: 2010 Sep-Oct Impact factor: 6.288
Authors: Stavros G Memtsoudis; Eduardo A Salvati; George Go; Yan Ma; Nigel E Sharrock Journal: Reg Anesth Pain Med Date: 2010 Sep-Oct Impact factor: 6.288
Authors: J T YaDeau; Y Lin; D J Mayman; E A Goytizolo; M M Alexiades; D E Padgett; R L Kahn; K M Jules-Elysee; A S Ranawat; D D Bhagat; K G Fields; A K Goon; J Curren; G H Westrich Journal: Br J Anaesth Date: 2015-08 Impact factor: 9.166
Authors: Jacques T YaDeau; Leonardo Paroli; Kara G Fields; Richard L Kahn; Vincent R LaSala; Kethy M Jules-Elysee; David H Kim; Stephen C Haskins; Jacob Hedden; Amanda Goon; Matthew M Roberts; David S Levine Journal: Reg Anesth Pain Med Date: 2015 Jul-Aug Impact factor: 6.288
Authors: Kethy M Jules-Elysee; Amanda K Goon; Geoffrey H Westrich; Douglas E Padgett; David J Mayman; Amar S Ranawat; Chitranjan S Ranawat; Yi Lin; Richard L Kahn; Devan D Bhagat; Enrique A Goytizolo; Yan Ma; Shane C Reid; Jodie Curren; Jacques T YaDeau Journal: J Bone Joint Surg Am Date: 2015-05-20 Impact factor: 5.284
Authors: S Carnio; D Galetta; V Scotti; D L Cortinovis; A Antonuzzo; S Pisconti; A Rossi; O Martelli; F L Cecere; A Lunghi; A Del Conte; E S Montagna; J Topulli; D Pelizzoni; S G Rapetti; M Gianetta; M V Pacchiana; V Pegoraro; N Cataldo; E Bria; S Novello Journal: Support Care Cancer Date: 2017-12-21 Impact factor: 3.603
Authors: Jose A Cortes-Briones; John D Cahill; Patrick D Skosnik; Daniel H Mathalon; Ashley Williams; R Andrew Sewell; Brian J Roach; Judith M Ford; Mohini Ranganathan; Deepak Cyril D'Souza Journal: Biol Psychiatry Date: 2015-03-30 Impact factor: 13.382
Authors: Anize D von Frankenberg; Anna Marina; Xiaoling Song; Holly S Callahan; Mario Kratz; Kristina M Utzschneider Journal: Eur J Nutr Date: 2015-11-28 Impact factor: 5.614
Authors: Abdullah S Terkawi; Siny Tsang; Ali Kazemi; Steve Morton; Roy Luo; Daniel T Sanders; Lindsay A Regali; Heather Columbano; Nicole Y Kurtzeborn; Marcel E Durieux Journal: Reg Anesth Pain Med Date: 2016 Jan-Feb Impact factor: 6.288