Élodie de Bock1, Jean-Benoit Hardouin, Myriam Blanchin, Tanguy Le Neel, Gildas Kubis, Véronique Sébille. 1. EA4275-SPHERE, 'Biostatistics, Pharmacoepidemiology and Subjective Measures in Health Sciences', Faculté de Pharmacie, Université de Nantes, Pres Université Nantes Angers Le Mans, 1, rue Gaston Veil, 44035, Nantes Cedex 01, France, elodie.debock@univ-nantes.fr.
Abstract
PURPOSE: The purpose of this study was to identify the most adequate strategy for group comparison of longitudinal patient-reported outcomes in the presence of possibly informative intermittent missing data. Models coming from classical test theory (CTT) and item response theory (IRT) were compared. METHODS: Two groups of patients' responses to dichotomous items with three times of assessment were simulated. Different cases were considered: presence or absence of a group effect and/or a time effect, a total of 100 or 200 patients, 4 or 7 items and two different values for the correlation coefficient of the latent trait between two consecutive times (0.4 or 0.9). Cases including informative and non-informative intermittent missing data were compared at different rates (15, 30 %). These simulated data were analyzed with CTT using score and mixed model (SM) and with IRT using longitudinal Rasch mixed model (LRM). The type I error, the power and the bias of the group effect estimations were compared between the two methods. RESULTS: This study showed that LRM performs better than SM. When the rate of missing data rose to 30 %, estimations were biased with SM mainly for informative missing data. Otherwise, LRM and SM methods were comparable concerning biases. However, regardless of the rate of intermittent missing data, power of LRM was higher compared to power of SM. CONCLUSIONS: In conclusion, LRM should be favored when the rate of missing data is higher than 15 %. For other cases, SM and LRM provide similar results.
PURPOSE: The purpose of this study was to identify the most adequate strategy for group comparison of longitudinal patient-reported outcomes in the presence of possibly informative intermittent missing data. Models coming from classical test theory (CTT) and item response theory (IRT) were compared. METHODS: Two groups of patients' responses to dichotomous items with three times of assessment were simulated. Different cases were considered: presence or absence of a group effect and/or a time effect, a total of 100 or 200 patients, 4 or 7 items and two different values for the correlation coefficient of the latent trait between two consecutive times (0.4 or 0.9). Cases including informative and non-informative intermittent missing data were compared at different rates (15, 30 %). These simulated data were analyzed with CTT using score and mixed model (SM) and with IRT using longitudinal Rasch mixed model (LRM). The type I error, the power and the bias of the group effect estimations were compared between the two methods. RESULTS: This study showed that LRM performs better than SM. When the rate of missing data rose to 30 %, estimations were biased with SM mainly for informative missing data. Otherwise, LRM and SM methods were comparable concerning biases. However, regardless of the rate of intermittent missing data, power of LRM was higher compared to power of SM. CONCLUSIONS: In conclusion, LRM should be favored when the rate of missing data is higher than 15 %. For other cases, SM and LRM provide similar results.
Authors: David Cella; Jennifer L Beaumont; Kimberly A Webster; Jin-Shei Lai; Linda Elting Journal: Support Care Cancer Date: 2006-08-30 Impact factor: 3.603
Authors: J B Bjorner; M Aa Petersen; M Groenvold; N Aaronson; M Ahlner-Elmqvist; J I Arraras; A Brédart; P Fayers; M Jordhoy; M Sprangers; M Watson; T Young Journal: Qual Life Res Date: 2004-12 Impact factor: 4.147
Authors: David W Kissane; Snehal G Patel; Raymond E Baser; Rachel Bell; Maria Farberov; Jamie S Ostroff; Yuelin Li; Bhuvanesh Singh; Dennis H Kraus; Jatin P Shah Journal: Head Neck Date: 2012-02-24 Impact factor: 3.147
Authors: J Bernhard; D F Cella; A S Coates; L Fallowfield; P A Ganz; C M Moinpour; P Mosconi; D Osoba; J Simes; C Hürny Journal: Stat Med Date: 1998 Mar 15-Apr 15 Impact factor: 2.373
Authors: Aynslie M Hinds; Tolulope T Sajobi; Véronique Sebille; Richard Sawatzky; Lisa M Lix Journal: Qual Life Res Date: 2018-04-20 Impact factor: 4.147
Authors: Alice Guilleux; Myriam Blanchin; Antoine Vanier; Francis Guillemin; Bruno Falissard; Carolyn E Schwartz; Jean-Benoit Hardouin; Véronique Sébille Journal: Qual Life Res Date: 2014-12-05 Impact factor: 4.147
Authors: Amélie Anota; Antoine Barbieri; Marion Savina; Alhousseiny Pam; Sophie Gourgou-Bourgade; Franck Bonnetain; Caroline Bascoul-Mollevi Journal: Health Qual Life Outcomes Date: 2014-12-31 Impact factor: 3.186