Incomplete quality of life data in randomized trials

Missing items

Peter M. Fayers* (Corresponding Author), Desmond Curran, David Machin

*Corresponding author for this work

Research output: Contribution to journalArticle

95 Citations (Scopus)

Abstract

Missing data has been a problem in many quality of life studies. This paper focuses upon the issues involved in handling forms which contain one or more missing items, and reviews the alternative procedures. One of the most widely practised approaches is imputation using the mean of all observed items in the same subscale. This, together with the related estimation of the subscale score, is based upon traditional psychometric approaches to scale design and analysis. We show that it may be an inappropriate method for many of the items in quality of life questionnaires, and would result in biased or misleading estimates. We provide examples of items and subscales which violate the psychometric foundations that underpin simple mean imputation. A checklist is proposed for examining the adequacy of simple imputation, and some alternative procedures are indicated.

Original languageEnglish
Pages (from-to)679-696
Number of pages18
JournalStatistics in Medicine
Volume17
Issue number5-7
DOIs
Publication statusPublished - 15 Mar 1998

Fingerprint

Randomized Trial
Quality of Life
Imputation
Psychometrics
Alternatives
Violate
Checklist
Missing Data
Questionnaire
Biased
Estimate

ASJC Scopus subject areas

  • Epidemiology

Cite this

Incomplete quality of life data in randomized trials : Missing items. / Fayers, Peter M. (Corresponding Author); Curran, Desmond; Machin, David.

In: Statistics in Medicine, Vol. 17, No. 5-7, 15.03.1998, p. 679-696.

Research output: Contribution to journalArticle

Fayers, Peter M. ; Curran, Desmond ; Machin, David. / Incomplete quality of life data in randomized trials : Missing items. In: Statistics in Medicine. 1998 ; Vol. 17, No. 5-7. pp. 679-696.
@article{8074a44d93d34478948d60d51bb1423f,
title = "Incomplete quality of life data in randomized trials: Missing items",
abstract = "Missing data has been a problem in many quality of life studies. This paper focuses upon the issues involved in handling forms which contain one or more missing items, and reviews the alternative procedures. One of the most widely practised approaches is imputation using the mean of all observed items in the same subscale. This, together with the related estimation of the subscale score, is based upon traditional psychometric approaches to scale design and analysis. We show that it may be an inappropriate method for many of the items in quality of life questionnaires, and would result in biased or misleading estimates. We provide examples of items and subscales which violate the psychometric foundations that underpin simple mean imputation. A checklist is proposed for examining the adequacy of simple imputation, and some alternative procedures are indicated.",
author = "Fayers, {Peter M.} and Desmond Curran and David Machin",
note = "ACKNOWLEDGEMENTS We wish to thank the MRC Cancer Therapy Committee and its Working Parties for access to data from MRC trials CR04, LU16, and the TE17 study. Copies of MRC protocols are available upon request to MRC Cancer Trials Office, 5 Shaftesbury Road, Cambridge CB2 2BW, U.K.",
year = "1998",
month = "3",
day = "15",
doi = "10.1002/(SICI)1097-0258(19980315/15)17:5/7<679::AID-SIM814>3.0.CO;2-X",
language = "English",
volume = "17",
pages = "679--696",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "5-7",

}

TY - JOUR

T1 - Incomplete quality of life data in randomized trials

T2 - Missing items

AU - Fayers, Peter M.

AU - Curran, Desmond

AU - Machin, David

N1 - ACKNOWLEDGEMENTS We wish to thank the MRC Cancer Therapy Committee and its Working Parties for access to data from MRC trials CR04, LU16, and the TE17 study. Copies of MRC protocols are available upon request to MRC Cancer Trials Office, 5 Shaftesbury Road, Cambridge CB2 2BW, U.K.

PY - 1998/3/15

Y1 - 1998/3/15

N2 - Missing data has been a problem in many quality of life studies. This paper focuses upon the issues involved in handling forms which contain one or more missing items, and reviews the alternative procedures. One of the most widely practised approaches is imputation using the mean of all observed items in the same subscale. This, together with the related estimation of the subscale score, is based upon traditional psychometric approaches to scale design and analysis. We show that it may be an inappropriate method for many of the items in quality of life questionnaires, and would result in biased or misleading estimates. We provide examples of items and subscales which violate the psychometric foundations that underpin simple mean imputation. A checklist is proposed for examining the adequacy of simple imputation, and some alternative procedures are indicated.

AB - Missing data has been a problem in many quality of life studies. This paper focuses upon the issues involved in handling forms which contain one or more missing items, and reviews the alternative procedures. One of the most widely practised approaches is imputation using the mean of all observed items in the same subscale. This, together with the related estimation of the subscale score, is based upon traditional psychometric approaches to scale design and analysis. We show that it may be an inappropriate method for many of the items in quality of life questionnaires, and would result in biased or misleading estimates. We provide examples of items and subscales which violate the psychometric foundations that underpin simple mean imputation. A checklist is proposed for examining the adequacy of simple imputation, and some alternative procedures are indicated.

UR - http://www.scopus.com/inward/record.url?scp=2642617783&partnerID=8YFLogxK

U2 - 10.1002/(SICI)1097-0258(19980315/15)17:5/7<679::AID-SIM814>3.0.CO;2-X

DO - 10.1002/(SICI)1097-0258(19980315/15)17:5/7<679::AID-SIM814>3.0.CO;2-X

M3 - Article

VL - 17

SP - 679

EP - 696

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 5-7

ER -