Practical and statistical issues in missing data for longitudinal patient-reported outcomes

Melanie L Bell, Diane L. Fairclough

Research output: Contribution to journalArticle

95 Citations (Scopus)

Abstract

Patient-reported outcomes are increasingly used in health research, including randomized controlled trials and observational studies. However, the validity of results in longitudinal studies can crucially hinge on the handling of missing data. This paper considers the issues of missing data at each stage of research. Practical strategies for minimizing missingness through careful study design and conduct are given. Statistical approaches that are commonly used, but should be avoided, are discussed, including how these methods can yield biased and misleading results. Methods that are valid for data which are missing at random are outlined, including maximum likelihood methods, multiple imputation and extensions to generalized estimating equations: weighted generalized estimating equations, generalized estimating equations with multiple imputation, and doubly robust generalized estimating equations. Finally, we discuss the importance of sensitivity analyses, including the role of missing not at random models, such as pattern mixture, selection, and shared parameter models. We demonstrate many of these concepts with data from a randomized controlled clinical trial on renal cancer patients, and show that the results are dependent on missingness assumptions and the statistical approach.

Original languageEnglish (US)
Pages (from-to)440-459
Number of pages20
JournalStatistical Methods in Medical Research
Volume23
Issue number5
DOIs
StatePublished - Oct 1 2014
Externally publishedYes

Fingerprint

Generalized Estimating Equations
Missing Data
Missing at Random
Multiple Imputation
Randomized Controlled Trials
Kidney Neoplasms
Weighted Estimating Equations
Randomized Controlled Trial
Research
Reproducibility of Results
Observational Study
Observational Studies
Longitudinal Studies
Longitudinal Study
Maximum Likelihood Method
Clinical Trials
Biased
Cancer
Health
Valid

Keywords

  • cancer
  • generalized estimating equations
  • maximum likelihood estimation
  • Missing data
  • multiple imputation
  • patient reported outcomes
  • quality of life

ASJC Scopus subject areas

  • Epidemiology
  • Health Information Management
  • Statistics and Probability
  • Medicine(all)

Cite this

Practical and statistical issues in missing data for longitudinal patient-reported outcomes. / Bell, Melanie L; Fairclough, Diane L.

In: Statistical Methods in Medical Research, Vol. 23, No. 5, 01.10.2014, p. 440-459.

Research output: Contribution to journalArticle

@article{19e2b4c9148049c8be40860944f4309f,
title = "Practical and statistical issues in missing data for longitudinal patient-reported outcomes",
abstract = "Patient-reported outcomes are increasingly used in health research, including randomized controlled trials and observational studies. However, the validity of results in longitudinal studies can crucially hinge on the handling of missing data. This paper considers the issues of missing data at each stage of research. Practical strategies for minimizing missingness through careful study design and conduct are given. Statistical approaches that are commonly used, but should be avoided, are discussed, including how these methods can yield biased and misleading results. Methods that are valid for data which are missing at random are outlined, including maximum likelihood methods, multiple imputation and extensions to generalized estimating equations: weighted generalized estimating equations, generalized estimating equations with multiple imputation, and doubly robust generalized estimating equations. Finally, we discuss the importance of sensitivity analyses, including the role of missing not at random models, such as pattern mixture, selection, and shared parameter models. We demonstrate many of these concepts with data from a randomized controlled clinical trial on renal cancer patients, and show that the results are dependent on missingness assumptions and the statistical approach.",
keywords = "cancer, generalized estimating equations, maximum likelihood estimation, Missing data, multiple imputation, patient reported outcomes, quality of life",
author = "Bell, {Melanie L} and Fairclough, {Diane L.}",
year = "2014",
month = "10",
day = "1",
doi = "10.1177/0962280213476378",
language = "English (US)",
volume = "23",
pages = "440--459",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "5",

}

TY - JOUR

T1 - Practical and statistical issues in missing data for longitudinal patient-reported outcomes

AU - Bell, Melanie L

AU - Fairclough, Diane L.

PY - 2014/10/1

Y1 - 2014/10/1

N2 - Patient-reported outcomes are increasingly used in health research, including randomized controlled trials and observational studies. However, the validity of results in longitudinal studies can crucially hinge on the handling of missing data. This paper considers the issues of missing data at each stage of research. Practical strategies for minimizing missingness through careful study design and conduct are given. Statistical approaches that are commonly used, but should be avoided, are discussed, including how these methods can yield biased and misleading results. Methods that are valid for data which are missing at random are outlined, including maximum likelihood methods, multiple imputation and extensions to generalized estimating equations: weighted generalized estimating equations, generalized estimating equations with multiple imputation, and doubly robust generalized estimating equations. Finally, we discuss the importance of sensitivity analyses, including the role of missing not at random models, such as pattern mixture, selection, and shared parameter models. We demonstrate many of these concepts with data from a randomized controlled clinical trial on renal cancer patients, and show that the results are dependent on missingness assumptions and the statistical approach.

AB - Patient-reported outcomes are increasingly used in health research, including randomized controlled trials and observational studies. However, the validity of results in longitudinal studies can crucially hinge on the handling of missing data. This paper considers the issues of missing data at each stage of research. Practical strategies for minimizing missingness through careful study design and conduct are given. Statistical approaches that are commonly used, but should be avoided, are discussed, including how these methods can yield biased and misleading results. Methods that are valid for data which are missing at random are outlined, including maximum likelihood methods, multiple imputation and extensions to generalized estimating equations: weighted generalized estimating equations, generalized estimating equations with multiple imputation, and doubly robust generalized estimating equations. Finally, we discuss the importance of sensitivity analyses, including the role of missing not at random models, such as pattern mixture, selection, and shared parameter models. We demonstrate many of these concepts with data from a randomized controlled clinical trial on renal cancer patients, and show that the results are dependent on missingness assumptions and the statistical approach.

KW - cancer

KW - generalized estimating equations

KW - maximum likelihood estimation

KW - Missing data

KW - multiple imputation

KW - patient reported outcomes

KW - quality of life

UR - http://www.scopus.com/inward/record.url?scp=84907499920&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907499920&partnerID=8YFLogxK

U2 - 10.1177/0962280213476378

DO - 10.1177/0962280213476378

M3 - Article

C2 - 23427225

AN - SCOPUS:84907499920

VL - 23

SP - 440

EP - 459

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 5

ER -