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

Nancy J. Preston*, Peter Fayers, Stephen J. Walters, Mark Pilling, Gunn E. Grande, Vicky Short, Eleanor Owen-Jones, Catherine J. Evans, Hamid Benalia, Irene J. Higginson, Chris J. Todd, MORECare

*Corresponding author for this work

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

Background: Statistical analysis in palliative and end-of-life care research can be problematic due to high levels of missing data, attrition and response shift as disease progresses.

Aim: To develop recommendations about managing missing data, attrition and response shift in palliative and end-of-life care research data.

Design: We used the MORECare Transparent Expert Consultation approach to conduct a consultation workshop with experts in statistical methods in palliative and end-of-life care research. Following presentations and discussion, nominal group techniques were used to produce recommendations about attrition, missing data and response shift. These were rated online by experts and analysed using descriptive statistics for consensus and importance.

Results: In total, 20 participants attended the workshop and 19 recommendations were subsequently ranked. There was broad agreement across recommendations. The top five recommendations were as follows:

A taxonomy should be devised to define types of attrition. Types and amount of missing data should be reported with details of imputation methods. The pattern of missing data should be investigated to inform the imputation approach. A statistical analysis plan should be pre-specified in the protocol. High rates of attrition should be assumed when planning studies and specifying analyses.

The leading recommendation for response shift was for more research.

Conclusions: When designing studies in palliative and end-of-life care, it is recommended that high rates of attrition should not be seen as indicative of poor design and that a clear statistical analysis plan is in place to account for missing data and attrition.

Original languageEnglish
Pages (from-to)899-907
Number of pages9
JournalPalliative Medicine
Volume27
Issue number10
Early online date7 May 2013
DOIs
Publication statusPublished - Dec 2013

Keywords

  • statistics
  • research design
  • palliative care
  • consensus
  • cancer-patients
  • lung-cancer
  • trial

Cite this

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. / Preston, Nancy J.; Fayers, Peter; Walters, Stephen J.; Pilling, Mark; Grande, Gunn E.; Short, Vicky; Owen-Jones, Eleanor; Evans, Catherine J.; Benalia, Hamid; Higginson, Irene J.; Todd, Chris J.; MORECare.

In: Palliative Medicine, Vol. 27, No. 10, 12.2013, p. 899-907.

Research output: Contribution to journalArticle

Preston, NJ, Fayers, P, Walters, SJ, Pilling, M, Grande, GE, Short, V, Owen-Jones, E, Evans, CJ, Benalia, H, Higginson, IJ, Todd, CJ & MORECare 2013, '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', Palliative Medicine, vol. 27, no. 10, pp. 899-907. https://doi.org/10.1177/0269216313486952
Preston, Nancy J. ; Fayers, Peter ; Walters, Stephen J. ; Pilling, Mark ; Grande, Gunn E. ; Short, Vicky ; Owen-Jones, Eleanor ; Evans, Catherine J. ; Benalia, Hamid ; Higginson, Irene J. ; Todd, Chris J. ; MORECare. / 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. In: Palliative Medicine. 2013 ; Vol. 27, No. 10. pp. 899-907.
@article{98d97255a5394422bec8b4cc4c5746c5,
title = "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",
abstract = "Background: Statistical analysis in palliative and end-of-life care research can be problematic due to high levels of missing data, attrition and response shift as disease progresses.Aim: To develop recommendations about managing missing data, attrition and response shift in palliative and end-of-life care research data.Design: We used the MORECare Transparent Expert Consultation approach to conduct a consultation workshop with experts in statistical methods in palliative and end-of-life care research. Following presentations and discussion, nominal group techniques were used to produce recommendations about attrition, missing data and response shift. These were rated online by experts and analysed using descriptive statistics for consensus and importance.Results: In total, 20 participants attended the workshop and 19 recommendations were subsequently ranked. There was broad agreement across recommendations. The top five recommendations were as follows:A taxonomy should be devised to define types of attrition. Types and amount of missing data should be reported with details of imputation methods. The pattern of missing data should be investigated to inform the imputation approach. A statistical analysis plan should be pre-specified in the protocol. High rates of attrition should be assumed when planning studies and specifying analyses.The leading recommendation for response shift was for more research.Conclusions: When designing studies in palliative and end-of-life care, it is recommended that high rates of attrition should not be seen as indicative of poor design and that a clear statistical analysis plan is in place to account for missing data and attrition.",
keywords = "statistics, research design, palliative care, consensus, cancer-patients, lung-cancer, trial",
author = "Preston, {Nancy J.} and Peter Fayers and Walters, {Stephen J.} and Mark Pilling and Grande, {Gunn E.} and Vicky Short and Eleanor Owen-Jones and Evans, {Catherine J.} and Hamid Benalia and Higginson, {Irene J.} and Todd, {Chris J.} and MORECare",
year = "2013",
month = "12",
doi = "10.1177/0269216313486952",
language = "English",
volume = "27",
pages = "899--907",
journal = "Palliative Medicine",
issn = "0269-2163",
publisher = "SAGE Publications Ltd",
number = "10",

}

TY - JOUR

T1 - Recommendations for managing missing data, attrition and response shift in palliative and end-of-life care research

T2 - Part of the MORECare research method guidance on statistical issues

AU - Preston, Nancy J.

AU - Fayers, Peter

AU - Walters, Stephen J.

AU - Pilling, Mark

AU - Grande, Gunn E.

AU - Short, Vicky

AU - Owen-Jones, Eleanor

AU - Evans, Catherine J.

AU - Benalia, Hamid

AU - Higginson, Irene J.

AU - Todd, Chris J.

AU - MORECare

PY - 2013/12

Y1 - 2013/12

N2 - Background: Statistical analysis in palliative and end-of-life care research can be problematic due to high levels of missing data, attrition and response shift as disease progresses.Aim: To develop recommendations about managing missing data, attrition and response shift in palliative and end-of-life care research data.Design: We used the MORECare Transparent Expert Consultation approach to conduct a consultation workshop with experts in statistical methods in palliative and end-of-life care research. Following presentations and discussion, nominal group techniques were used to produce recommendations about attrition, missing data and response shift. These were rated online by experts and analysed using descriptive statistics for consensus and importance.Results: In total, 20 participants attended the workshop and 19 recommendations were subsequently ranked. There was broad agreement across recommendations. The top five recommendations were as follows:A taxonomy should be devised to define types of attrition. Types and amount of missing data should be reported with details of imputation methods. The pattern of missing data should be investigated to inform the imputation approach. A statistical analysis plan should be pre-specified in the protocol. High rates of attrition should be assumed when planning studies and specifying analyses.The leading recommendation for response shift was for more research.Conclusions: When designing studies in palliative and end-of-life care, it is recommended that high rates of attrition should not be seen as indicative of poor design and that a clear statistical analysis plan is in place to account for missing data and attrition.

AB - Background: Statistical analysis in palliative and end-of-life care research can be problematic due to high levels of missing data, attrition and response shift as disease progresses.Aim: To develop recommendations about managing missing data, attrition and response shift in palliative and end-of-life care research data.Design: We used the MORECare Transparent Expert Consultation approach to conduct a consultation workshop with experts in statistical methods in palliative and end-of-life care research. Following presentations and discussion, nominal group techniques were used to produce recommendations about attrition, missing data and response shift. These were rated online by experts and analysed using descriptive statistics for consensus and importance.Results: In total, 20 participants attended the workshop and 19 recommendations were subsequently ranked. There was broad agreement across recommendations. The top five recommendations were as follows:A taxonomy should be devised to define types of attrition. Types and amount of missing data should be reported with details of imputation methods. The pattern of missing data should be investigated to inform the imputation approach. A statistical analysis plan should be pre-specified in the protocol. High rates of attrition should be assumed when planning studies and specifying analyses.The leading recommendation for response shift was for more research.Conclusions: When designing studies in palliative and end-of-life care, it is recommended that high rates of attrition should not be seen as indicative of poor design and that a clear statistical analysis plan is in place to account for missing data and attrition.

KW - statistics

KW - research design

KW - palliative care

KW - consensus

KW - cancer-patients

KW - lung-cancer

KW - trial

U2 - 10.1177/0269216313486952

DO - 10.1177/0269216313486952

M3 - Article

VL - 27

SP - 899

EP - 907

JO - Palliative Medicine

JF - Palliative Medicine

SN - 0269-2163

IS - 10

ER -