Audrey Winter1,2, Benjamin Cuer3, Thierry Conroy4,5, Beata Juzyna6, Sophie Gourgou7,8, Caroline Mollevi7,8,3, Célia Touraine7,8,3. 1. Biometrics Unit, Cancer Institute of Montpellier, University of Montpellier, Parc Euromédecine, 208 Avenue des Apothicaires, 34090, Montpellier, France. audrey.winter@icm.unicancer.fr. 2. French National Platform Quality of Life and Cancer, Montpellier, France. audrey.winter@icm.unicancer.fr. 3. Desbrest Institute of Epidemiology and Public Health, University of Montpellier, INSERM, 641 Avenue du Doyen Gaston Giraud, Montpellier, France. 4. Department of Medical Oncology, Institut de Cancérologie de Lorraine, 6 Av. de Bourgogne, Vandoeuvre-lès-Nancy, France. 5. Teams MICS, APEMAC, Université de Lorraine, Nancy, France. 6. R&D Unicancer, Paris, France. 7. Biometrics Unit, Cancer Institute of Montpellier, University of Montpellier, Parc Euromédecine, 208 Avenue des Apothicaires, 34090, Montpellier, France. 8. French National Platform Quality of Life and Cancer, Montpellier, France.
Abstract
PURPOSE: A joint modeling approach is recommended for analysis of longitudinal health-related quality of life (HRQoL) data in the presence of potentially informative dropouts. However, the linear mixed model modeling the longitudinal HRQoL outcome in a joint model often assumes a linear trajectory over time, an oversimplification that can lead to incorrect results. Our aim was to demonstrate that a more flexible model gives more reliable and complete results without complicating their interpretation. METHODS: Five dimensions of HRQoL in patients with esophageal cancer from the randomized clinical trial PRODIGE 5/ACCORD 17 were analyzed. Joint models assuming linear or spline-based HRQoL trajectories were applied and compared in terms of interpretation of results, graphical representation, and goodness of fit. RESULTS: Spline-based models allowed arm-by-time interaction effects to be highlighted and led to a more precise and consistent representation of the HRQoL over time; this was supported by the martingale residuals and the Akaike information criterion. CONCLUSION: Linear relationships between continuous outcomes (such as HRQoL scores) and time are usually the default choice. However, the functional form turns out to be important by affecting both the validity of the model and the statistical significance. TRIAL REGISTRATION: This study is registered with ClinicalTrials.gov, number NCT00861094.
PURPOSE: A joint modeling approach is recommended for analysis of longitudinal health-related quality of life (HRQoL) data in the presence of potentially informative dropouts. However, the linear mixed model modeling the longitudinal HRQoL outcome in a joint model often assumes a linear trajectory over time, an oversimplification that can lead to incorrect results. Our aim was to demonstrate that a more flexible model gives more reliable and complete results without complicating their interpretation. METHODS: Five dimensions of HRQoL in patients with esophageal cancer from the randomized clinical trial PRODIGE 5/ACCORD 17 were analyzed. Joint models assuming linear or spline-based HRQoL trajectories were applied and compared in terms of interpretation of results, graphical representation, and goodness of fit. RESULTS: Spline-based models allowed arm-by-time interaction effects to be highlighted and led to a more precise and consistent representation of the HRQoL over time; this was supported by the martingale residuals and the Akaike information criterion. CONCLUSION: Linear relationships between continuous outcomes (such as HRQoL scores) and time are usually the default choice. However, the functional form turns out to be important by affecting both the validity of the model and the statistical significance. TRIAL REGISTRATION: This study is registered with ClinicalTrials.gov, number NCT00861094.
Authors: Divine E Ediebah; Francisca Galindo-Garre; Bernard M J Uitdehaag; Jolie Ringash; Jaap C Reijneveld; Linda Dirven; Efstathios Zikos; Corneel Coens; Martin J van den Bent; Andrew Bottomley; Martin J B Taphoorn Journal: Qual Life Res Date: 2014-10-14 Impact factor: 4.147
Authors: C Bascoul-Mollevi; S Gourgou; M-P Galais; J-L Raoul; O Bouché; J-Y Douillard; A Adenis; P-L Etienne; B Juzyna; L Bedenne; T Conroy Journal: Eur J Cancer Date: 2017-08-19 Impact factor: 9.162
Authors: Zhigang Li; H R Frost; Tor D Tosteson; Lihui Zhao; Lei Liu; Kathleen Lyons; Huaihou Chen; Bernard Cole; David Currow; Marie Bakitas Journal: Stat Med Date: 2017-08-17 Impact factor: 2.373
Authors: N K Aaronson; S Ahmedzai; B Bergman; M Bullinger; A Cull; N J Duez; A Filiberti; H Flechtner; S B Fleishman; J C de Haes Journal: J Natl Cancer Inst Date: 1993-03-03 Impact factor: 13.506
Authors: J M Blazeby; T Conroy; E Hammerlid; P Fayers; O Sezer; M Koller; J Arraras; A Bottomley; C W Vickery; P L Etienne; D Alderson Journal: Eur J Cancer Date: 2003-07 Impact factor: 9.162
Authors: Willi Sauerbrei; Aris Perperoglou; Matthias Schmid; Michal Abrahamowicz; Heiko Becher; Harald Binder; Daniela Dunkler; Frank E Harrell; Patrick Royston; Georg Heinze Journal: Diagn Progn Res Date: 2020-04-02