Giulia Carreras1, Guido Miccinesi2, Andrew Wilcock3, Nancy Preston4, Daan Nieboer5, Luc Deliens6, Mogensm Groenvold7, Urska Lunder8, Agnes van der Heide5, Michela Baccini9. 1. Oncological Network, Prevention and Research Institute (ISPRO), Florence, Italy. g.carreras@ispro.toscana.it. 2. Oncological Network, Prevention and Research Institute (ISPRO), Florence, Italy. 3. Department of Clinical Oncology, University of Nottingham, Nottingham, UK. 4. Lancaster University, International Observatory on end of life care, Lancaster, UK. 5. Department of Public Health, Erasmus University, Rotterdam, Netherlands. 6. Vrije Universiteit Brussel & Ghent University, Brussels, Belgium. 7. Department of Public Health, Copenhagen University, Copenhagen, Denmark. 8. University Clinic for Respiratory and Allergic Diseases, Golnik, Slovenia. 9. Department of Statistics, Computer Science, Applications 'G. Parenti' (DISIA), University of Florence, Florence, Italy.
Abstract
BACKGROUND: Missing data are common in end-of-life care studies, but there is still relatively little exploration of which is the best method to deal with them, and, in particular, if the missing at random (MAR) assumption is valid or missing not at random (MNAR) mechanisms should be assumed. In this paper we investigated this issue through a sensitivity analysis within the ACTION study, a multicenter cluster randomized controlled trial testing advance care planning in patients with advanced lung or colorectal cancer. METHODS: Multiple imputation procedures under MAR and MNAR assumptions were implemented. Possible violation of the MAR assumption was addressed with reference to variables measuring quality of life and symptoms. The MNAR model assumed that patients with worse health were more likely to have missing questionnaires, making a distinction between single missing items, which were assumed to satisfy the MAR assumption, and missing values due to completely missing questionnaire for which a MNAR mechanism was hypothesized. We explored the sensitivity to possible departures from MAR on gender differences between key indicators and on simple correlations. RESULTS: Up to 39% of follow-up data were missing. Results under MAR reflected that missingness was related to poorer health status. Correlations between variables, although very small, changed according to the imputation method, as well as the differences in scores by gender, indicating a certain sensitivity of the results to the violation of the MAR assumption. CONCLUSIONS: The findings confirmed the importance of undertaking this kind of analysis in end-of-life care studies.
RCT Entities:
BACKGROUND: Missing data are common in end-of-life care studies, but there is still relatively little exploration of which is the best method to deal with them, and, in particular, if the missing at random (MAR) assumption is valid or missing not at random (MNAR) mechanisms should be assumed. In this paper we investigated this issue through a sensitivity analysis within the ACTION study, a multicenter cluster randomized controlled trial testing advance care planning in patients with advanced lung or colorectal cancer. METHODS: Multiple imputation procedures under MAR and MNAR assumptions were implemented. Possible violation of the MAR assumption was addressed with reference to variables measuring quality of life and symptoms. The MNAR model assumed that patients with worse health were more likely to have missing questionnaires, making a distinction between single missing items, which were assumed to satisfy the MAR assumption, and missing values due to completely missing questionnaire for which a MNAR mechanism was hypothesized. We explored the sensitivity to possible departures from MAR on gender differences between key indicators and on simple correlations. RESULTS: Up to 39% of follow-up data were missing. Results under MAR reflected that missingness was related to poorer health status. Correlations between variables, although very small, changed according to the imputation method, as well as the differences in scores by gender, indicating a certain sensitivity of the results to the violation of the MAR assumption. CONCLUSIONS: The findings confirmed the importance of undertaking this kind of analysis in end-of-life care studies.
Entities:
Keywords:
Advance care planning; MAR; MNAR; Missing data; Oncology; Quality of life
Authors: Mogens Groenvold; Morten Aa Petersen; Neil K Aaronson; Juan I Arraras; Jane M Blazeby; Andrew Bottomley; Peter M Fayers; Alexander de Graeff; Eva Hammerlid; Stein Kaasa; Mirjam A G Sprangers; Jakob B Bjorner Journal: Eur J Cancer Date: 2005-09-12 Impact factor: 9.162
Authors: Jonathan A C Sterne; Ian R White; John B Carlin; Michael Spratt; Patrick Royston; Michael G Kenward; Angela M Wood; James R Carpenter Journal: BMJ Date: 2009-06-29
Authors: Wendy J Post; Ciska Buijs; Ronald P Stolk; Elisabeth G E de Vries; Saskia le Cessie Journal: Qual Life Res Date: 2009-12-30 Impact factor: 4.147
Authors: Thomas R Sullivan; Ian R White; Amy B Salter; Philip Ryan; Katherine J Lee Journal: Stat Methods Med Res Date: 2016-12-19 Impact factor: 3.021
Authors: Anoop D Shah; Jonathan W Bartlett; James Carpenter; Owen Nicholas; Harry Hemingway Journal: Am J Epidemiol Date: 2014-01-12 Impact factor: 4.897
Authors: Daniel Mark Tompsett; Finbarr Leacy; Margarita Moreno-Betancur; Jon Heron; Ian R White Journal: Stat Med Date: 2018-04-02 Impact factor: 2.373
Authors: Sun Sun; Nan Luo; Erik Stenberg; Lars Lindholm; Klas-Göran Sahlén; Karl A Franklin; Yang Cao Journal: Int J Environ Res Public Health Date: 2022-08-30 Impact factor: 4.614
Authors: Andreas Staudt; Jennis Freyer-Adam; Till Ittermann; Christian Meyer; Gallus Bischof; Ulrich John; Sophie Baumann Journal: BMC Med Res Methodol Date: 2022-09-24 Impact factor: 4.612