Missing not at random in end of life care studies: multiple imputation and sensitivity analysis on data from the ACTION study

Giulia Carreras, Guido Miccinesi, Andrew Wilcock, Nancy Preston, Daan Nieboer, Luc Deliens, Mogensm Groenvold, Urska Lunder, Agnes van der Heide, Michela Baccini, ACTION consortium, Agnes van der Heide, Ida J Korfage, Judith A C Rietjens, Lea J Jabbarian, Suzanne Polinder, Hans van Delden, Marijke Kars, Marieke Zwakman, Luc Deliens, Mariëtte N Verkissen, Kim Eecloo, Kristof Faes, Kristian Pollock, Jane Seymour, Glenys Caswell, Andrew Wilcock, Louise Bramley, Sheila Payne, Nancy Preston, Lesley Dunleavy, Eleanor Sowerby, Guido Miccinesi, Francesco Bulli, Francesca Ingravallo, Giulia Carreras, Alessandro Toccafondi, Giuseppe Gorini, Urška Lunder, Branka Červ, Anja Simonič, Alenka Mimić, Hana Kodba-Čeh, Polona Ozbič, Mogens Groenvold, Caroline Arnfeldt, Anna Thit Johnsen, Giulia Carreras, Guido Miccinesi, Andrew Wilcock, Nancy Preston, Daan Nieboer, Luc Deliens, Mogensm Groenvold, Urska Lunder, Agnes van der Heide, Michela Baccini, ACTION consortium, Agnes van der Heide, Ida J Korfage, Judith A C Rietjens, Lea J Jabbarian, Suzanne Polinder, Hans van Delden, Marijke Kars, Marieke Zwakman, Luc Deliens, Mariëtte N Verkissen, Kim Eecloo, Kristof Faes, Kristian Pollock, Jane Seymour, Glenys Caswell, Andrew Wilcock, Louise Bramley, Sheila Payne, Nancy Preston, Lesley Dunleavy, Eleanor Sowerby, Guido Miccinesi, Francesco Bulli, Francesca Ingravallo, Giulia Carreras, Alessandro Toccafondi, Giuseppe Gorini, Urška Lunder, Branka Červ, Anja Simonič, Alenka Mimić, Hana Kodba-Čeh, Polona Ozbič, Mogens Groenvold, Caroline Arnfeldt, Anna Thit Johnsen

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.

Keywords: Advance care planning; MAR; MNAR; Missing data; Oncology; Quality of life.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1
Fig. 1
Correlations (90% confidence intervals) between Quality of life and secondary endpoints (Patient involvement; Overall quality of care; Active coping; Denial) calculated after MI under the MAR assumption and under different MNAR models (kWHO; k1; k2; k3; k4)

References

    1. Palmer JL. Analysis of missing data in palliative care studies. J Pain Symptom Manag. 2004;28:612–618. doi: 10.1016/j.jpainsymman.2004.02.026.
    1. Hussain JA, White IR, Langan D, Johnson MJ, Currow DC, Torgerson DJ, et al. Missing data in randomized controlled trials testing palliative interventions pose a significant risk of bias and loss of power: a systematic review and meta-analyses. J Clin Epidemiol. 2015;74:57–65. doi: 10.1016/j.jclinepi.2015.12.003.
    1. Hussain JA, Bland M, Langan D, Johnson MJ, Currow DC, White IR. Quality of missing data reporting and handling in palliative care trials demonstrates that further development of the CONSORT missing data reporting guidance is required: a systematic review. J Clin Epidemiol. 2017;88:81–91. doi: 10.1016/j.jclinepi.2017.05.009.
    1. Rubin DB. Multiple imputation for nonresponse in surveys. New York: Wiley; 1987.
    1. Sullivan TR, Yelland LN, Lee KJ, Ryan P, Salter AB. Treatment of missing data in follow-up studies of randomised controlled trials: a systematic review of the literature. Clin Trials. 2017;14:387–395. doi: 10.1177/1740774517703319.
    1. Bell ML, Fiero M, Horton NJ, Hsu CH. Handling missing data in RCTs; a review of the top medical journals. BMC Med Res Methodol. 2014;14:118. doi: 10.1186/1471-2288-14-118.
    1. Sterne JA, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, Wood AM, Carpenter JR. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ. 2009;338:b2393. doi: 10.1136/bmj.b2393.
    1. Little RJA, Rubin DB. Statistical analysis with missing data. New Jersey: Wiley; 2002.
    1. Raghunathan TE, Lepkowski JM, van Hoewyk J, Solenberger P. A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol. 2001;27:85–95.
    1. Van Buuren S. Flexible imputation of missing data. Boca Raton: Chapman & Hall; 2018.
    1. Jones M, Mishra GD, Dobson A. Analytical results in longitudinal studies depended on target of inference and assumed mechanism of attrition. J Clin Epidemiol. 2015;68:1165–1175. doi: 10.1016/j.jclinepi.2015.03.011.
    1. Hu B, Li L, Greene T. Joint multiple imputation for longitudinal outcomes and clinical events that truncate longitudinal follow-up. Stat Med. 2016;35:2991–3006. doi: 10.1002/sim.6590.
    1. Moreno-Betancur M, Chavance M. Sensitivity analysis of incomplete longitudinal data departing from the missing at random assumption: methodology and application in a clinical trial with drop-outs. Stat Methods Med Res. 2016;25:1471–1489. doi: 10.1177/0962280213490014.
    1. Preston NJ, Fayers P, Walters SJ, Pilling M, Grande GE, Short V, et al. Recommendations for managing missing data, attrition and response shift in palliative and end-of-life care research: part of the MORECare research method guidance on statistical issues. Palliat Med. 2013;27:899e907. doi: 10.1177/0269216313486952.
    1. Post WJ, Buijs C, Stolk RP, de Vries EG, le Cessie S. The analysis of longitudinal quality of life measures with informative drop-out: a pattern mixture approach. Qual Life Res. 2010;19:137–148. doi: 10.1007/s11136-009-9564-1.
    1. Fielding S, Fayers PM, McDonald A, McPherson G, Campbell MK, RECORD Study Group Simple imputation methods were inadequate for missing not at random (MNAR) quality of life data. Health Qual Life Outcomes. 2008;6:57. doi: 10.1186/1477-7525-6-57.
    1. Pauler DK, McCoy S, Moinpour C. Pattern mixture models for longitudinal quality of life studies in advanced stage disease. Stat Med. 2003;22:795–809. doi: 10.1002/sim.1397.
    1. Diehr P, Johnson LL. Accounting for missing data in end-of-life research. J Palliat Med. 2005;8(Suppl 1):S50–S57. doi: 10.1089/jpm.2005.8.s-50.
    1. Rietjens JA, Korfage IJ, Dunleavy L, Preston NJ, Jabbarian LJ, Christensen CA, et al. Advance care planning- a multi-centre cluster randomised clinical trial: the research protocol of the ACTION study. BMC Cancer. 2016;8(16):264. doi: 10.1186/s12885-016-2298-x.
    1. Ratitch B, O'Kelly M, Tosiello R. Missing data in clinical trials: from clinical assumptions to statistical analysis using pattern mixture models. Pharm Stat. 2013;12:337–347. doi: 10.1002/pst.1549.
    1. Groenvold M, Petersen MA, Aaronson NK, Arraras JI, Blazeby JM, Bottomley A, Fayers PM, de Graeff A, Hammerlid E, Kaasa S, Sprangers MA, Bjorner JB, EORTC Quality of Life Group The development of the EORTC QLQ-C15-PAL: a shortened questionnaire for cancer patients in palliative care. Eur J Cancer. 2006;42:55–64. doi: 10.1016/j.ejca.2005.06.022.
    1. Fayers PM, Aaronson NK, Bjordal K, Groenvold M, Curran D, Bottomley A, on behalf of the EORTC Quality of Life Group . EORTC QLQ-C30 scoring manual. 3. Brussels: EORTC; 2001.
    1. Arora NK, Weaver KE, Clayman ML, Oakley-Girvan I, Potosky AL. Physicians’ decision-making style and psychosocial outcomes among cancer survivors. Patient Educ Couns. 2009;77:404–412. doi: 10.1016/j.pec.2009.10.004.
    1. Brédart A, Bottomley A, Blazeby JM, Conroy T, Coens C, D'Haese S, et al. An international prospective study of the EORTC cancer in-patient satisfaction with care measure (EORTC IN-PATSAT32) Eur J Cancer. 2005;41:2120–2131. doi: 10.1016/j.ejca.2005.04.041.
    1. Carver CS, Scheier MF, Weintraub JK. Assessing coping strategies: a theoretically based approach. J Pers Soc Psychol. 1989;56:267–283. doi: 10.1037/0022-3514.56.2.267.
    1. Stanton AL, Kirk SB, Cameron CL, Danoff-Burg S. Coping through emotional approach: scale construction and validation. J Pers Soc Psychol. 2000;78:1150–1169. doi: 10.1037/0022-3514.78.6.1150.
    1. Greenland S, Senn SJ, Rothman KJ, Carlin JB, Poole C, Goodman SN, Altman DG. Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations. Eur J Epidemiol. 2016;31:337–350. doi: 10.1007/s10654-016-0149-3.
    1. Li F, Baccini M, Mealli F, Zell ER, Frangakis CE, Rubin DB. Multiple imputation by ordered monotone blocks with application to the anthrax vaccine research program. J Comput Graphical Stat. 2014;23:877–892. doi: 10.1080/10618600.2013.826583.
    1. Gelman A, Rubin DB. Inference from iterative simulation using multiple sequences. Stat Sci. 1992;7:457–511. doi: 10.1214/ss/1177011136.
    1. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30:377–399. doi: 10.1002/sim.4067.
    1. Van Buuren S, Groothuis-Oudshoorn K. MICE: Multivariate imputation by chained equations. R J Stat Softw. 2011;45:1–67.
    1. R Core Team . R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2013.
    1. Tompsett DM, Leacy F, Moreno-Betancur M, Heron J, White IR. On the use of the not-at-random fully conditional specification (NARFCS) procedure in practice. Stat Med. 2018;37:2338–2353. doi: 10.1002/sim.7643.
    1. Resseguier N, Giorgi R, Paoletti X. Sensitivity analysis when data are missing not-at-random. Epidemiology. 2011;22:282. doi: 10.1097/EDE.0b013e318209dec7.
    1. Leacy FP, Floyd S, Yates TA, White IR. Analyses of sensitivity to the missing-at-random assumption using multiple imputation with delta adjustment: application to a tuberculosis/HIV prevalence survey with incomplete HIV-status data. Am J Epidemiol. 2017;185:304–315.
    1. Keene ON, Roger JH, Hartley BF, Kenward MG. Missing data sensitivity analysis for recurrent event data using controlled imputation. Pharmaceut Stat. 2014;13:258–264. doi: 10.1002/pst.1624.
    1. Little RJA. Selection and pattern-mixture models. In: Fitzmaurice G, Davidian M, Verbeke G, Molenberghs G, editors. Longitudinal data analysis, chapter 18. Boca Raton: CRC Press; 2009. pp. 409–431.
    1. Van Buuren S, Boshuizen HC, Knook DL. Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999;18:681–694. doi: 10.1002/(SICI)1097-0258(19990330)18:6<681::AID-SIM71>;2-R.
    1. Carpenter JR, Kenward MG. Multiple imputation and its application. 1. Chichester: Wiley; 2013.
    1. Burgette LF, Reiter JP. Multiple imputation for missing data via sequential regression trees. Am J Epidemiol. 2010;172:1070–1076. doi: 10.1093/aje/kwq260.
    1. Tang F, Ishwaran H. Random forest missing data algorithms. Stat Anal Data Min. 2017;10:363–377. doi: 10.1002/sam.11348.
    1. Doove LL, Van Buuren S, Dusseldorp E. Recursive partitioning for missing data imputation in the presence of interaction effects. Comput Stat Data Anal. 2014;72:92–104. doi: 10.1016/j.csda.2013.10.025.
    1. Hong S, Lynn HS. Accuracy of random-forest-based imputation of missing data in the presence of non-normality, non-linearity, and interaction. BMC Med Res Methodol. 2020;20:199. doi: 10.1186/s12874-020-01080-1.
    1. Shah AD, Bartlett JW, Carpenter J, Nicholas O, Hemingway H. Comparison of random forest and parametric imputation models for imputing missing data using MICE: a CALIBER study. Am J Epidemiol. 2014;179:764–774. doi: 10.1093/aje/kwt312.
    1. Lu K. Number of imputations needed to stabilize estimated treatment difference in longitudinal data analysis. Stat Methods Med Res. 2017;26:674–690. doi: 10.1177/0962280214554439.
    1. Sullivan TR, White IR, Salter AB, Ryan P, Lee KJ. Should multiple imputation be the method of choice for handling missing data in randomized trials? Stat Methods Med Res. 2018:962280216683570.
    1. Speidel M, Drechsler J, Jolani S. R package hmi: a convenient tool for hierarchical multiple imputation and beyond. 2018.
    1. Wang S, Tang J, Sun T, Zheng X, Li J, Sun H, Zhou X, Zhou C, Zhang H, Cheng Z, Ma H, Sun H. Survival changes in patients with small cell lung cancer and disparities between different sexes. Socioecon Statuses Ages Sci Rep. 2017;7:1339. doi: 10.1038/s41598-017-01571-0.
    1. Simmons CP, Koinis F, Fallon MT, Fearon KC, Bowden J, Solheim TS, Gronberg BH, McMillan DC, Gioulbasanis I, Laird BJ. Prognosis in advanced lung cancer--a prospective study examining key clinicopathological factors. Lung Cancer. 2015;88:304–309. doi: 10.1016/j.lungcan.2015.03.020.
    1. Siebenhüner AR, Güller U, Warschkow R. Population-based SEER analysis of survival in colorectal cancer patients with or without resection of lung and liver metastases. BMC Cancer. 2020;20(246). 10.1186/s12885-020-6710-1.
    1. Yang F, Small DS. Using post-outcome measurement information in censoring‐by‐death problems. J Royal Stat Soc Series B. 2016;78:299–318. doi: 10.1111/rssb.12113.
    1. Tchetgen Tchetgen EJ. Identification and estimation of survivor average causal effects. Stat Med. 2014;33:3601–3780. doi: 10.1002/sim.6181.
    1. Wen L, Terrera GM, Seaman SR. Methods for handling longitudinal outcome processes truncated by dropout and death. Biostatistics. 2018;19:407–425. doi: 10.1093/biostatistics/kxx045.
    1. Yang F, Ding P. Using survival information in truncation by death problems without the monotonicity assumption. Biometrics. 2018. 10.1111/biom.12883.

Source: PubMed

Sponsorer og samarbeidspartnere

Medisinsk tilstand

Legemiddelintervensjoner

3
Abonnere