Bias in area under the curve for longitudinal clinical trials with missing patient reported outcome data: Summary measures versus summary statistics

Melanie L Bell, Madeleine T. King, Diane L. Fairclough

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

A common approach to the analysis of longitudinal patient reported outcomes (PROs) is the use of summary measures such as area under the time curve (AUC). However, it is not clear how missing data affects the validity of AUC analysis. This study aimed to compare the use of AUC summary measures (in individuals) with AUC summary statistics (on groups, calculated from the estimated parameters of a mixed model) when data are complete, missing at random, and missing not at random. A simulation experiment based on a two-armed randomized trial was carried out to investigate the precision and bias of AUC in longitudinal analysis where missingness, trajectory, and missingness allocation were varied. Summary measures AUC with ad hoc approaches to missing data were compared with mixed model AUC summary statistics. AUC summary statistics were consistently superior to AUC summary measures in terms of precision and bias. The bias of AUC summary statistic approach was very small, even when data were missing not at random and when differential attrition between groups existed. AUC summary measures on individuals should not be used to analyze longitudinal PRO data in the presence of missing data.

Original languageEnglish (US)
JournalSAGE Open
Volume4
Issue number2
DOIs
StatePublished - 2014

Fingerprint

statistics
trend
time
Clinical Trials
Summary
Statistics
Group
simulation
experiment

Keywords

  • Applied psychology
  • Data processing and interpretation
  • Health psychology
  • Psychology
  • Reliability and validity
  • Research methods
  • Social sciences
  • Statistical theory and tests

ASJC Scopus subject areas

  • Social Sciences(all)
  • Arts and Humanities(all)

Cite this

Bias in area under the curve for longitudinal clinical trials with missing patient reported outcome data : Summary measures versus summary statistics. / Bell, Melanie L; King, Madeleine T.; Fairclough, Diane L.

In: SAGE Open, Vol. 4, No. 2, 2014.

Research output: Contribution to journalArticle

@article{84e7b96164c74a86a65f40ca19784f09,
title = "Bias in area under the curve for longitudinal clinical trials with missing patient reported outcome data: Summary measures versus summary statistics",
abstract = "A common approach to the analysis of longitudinal patient reported outcomes (PROs) is the use of summary measures such as area under the time curve (AUC). However, it is not clear how missing data affects the validity of AUC analysis. This study aimed to compare the use of AUC summary measures (in individuals) with AUC summary statistics (on groups, calculated from the estimated parameters of a mixed model) when data are complete, missing at random, and missing not at random. A simulation experiment based on a two-armed randomized trial was carried out to investigate the precision and bias of AUC in longitudinal analysis where missingness, trajectory, and missingness allocation were varied. Summary measures AUC with ad hoc approaches to missing data were compared with mixed model AUC summary statistics. AUC summary statistics were consistently superior to AUC summary measures in terms of precision and bias. The bias of AUC summary statistic approach was very small, even when data were missing not at random and when differential attrition between groups existed. AUC summary measures on individuals should not be used to analyze longitudinal PRO data in the presence of missing data.",
keywords = "Applied psychology, Data processing and interpretation, Health psychology, Psychology, Reliability and validity, Research methods, Social sciences, Statistical theory and tests",
author = "Bell, {Melanie L} and King, {Madeleine T.} and Fairclough, {Diane L.}",
year = "2014",
doi = "10.1177/2158244014534858",
language = "English (US)",
volume = "4",
journal = "SAGE Open",
issn = "2158-2440",
publisher = "SAGE Publications Inc.",
number = "2",

}

TY - JOUR

T1 - Bias in area under the curve for longitudinal clinical trials with missing patient reported outcome data

T2 - Summary measures versus summary statistics

AU - Bell, Melanie L

AU - King, Madeleine T.

AU - Fairclough, Diane L.

PY - 2014

Y1 - 2014

N2 - A common approach to the analysis of longitudinal patient reported outcomes (PROs) is the use of summary measures such as area under the time curve (AUC). However, it is not clear how missing data affects the validity of AUC analysis. This study aimed to compare the use of AUC summary measures (in individuals) with AUC summary statistics (on groups, calculated from the estimated parameters of a mixed model) when data are complete, missing at random, and missing not at random. A simulation experiment based on a two-armed randomized trial was carried out to investigate the precision and bias of AUC in longitudinal analysis where missingness, trajectory, and missingness allocation were varied. Summary measures AUC with ad hoc approaches to missing data were compared with mixed model AUC summary statistics. AUC summary statistics were consistently superior to AUC summary measures in terms of precision and bias. The bias of AUC summary statistic approach was very small, even when data were missing not at random and when differential attrition between groups existed. AUC summary measures on individuals should not be used to analyze longitudinal PRO data in the presence of missing data.

AB - A common approach to the analysis of longitudinal patient reported outcomes (PROs) is the use of summary measures such as area under the time curve (AUC). However, it is not clear how missing data affects the validity of AUC analysis. This study aimed to compare the use of AUC summary measures (in individuals) with AUC summary statistics (on groups, calculated from the estimated parameters of a mixed model) when data are complete, missing at random, and missing not at random. A simulation experiment based on a two-armed randomized trial was carried out to investigate the precision and bias of AUC in longitudinal analysis where missingness, trajectory, and missingness allocation were varied. Summary measures AUC with ad hoc approaches to missing data were compared with mixed model AUC summary statistics. AUC summary statistics were consistently superior to AUC summary measures in terms of precision and bias. The bias of AUC summary statistic approach was very small, even when data were missing not at random and when differential attrition between groups existed. AUC summary measures on individuals should not be used to analyze longitudinal PRO data in the presence of missing data.

KW - Applied psychology

KW - Data processing and interpretation

KW - Health psychology

KW - Psychology

KW - Reliability and validity

KW - Research methods

KW - Social sciences

KW - Statistical theory and tests

UR - http://www.scopus.com/inward/record.url?scp=84907289366&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907289366&partnerID=8YFLogxK

U2 - 10.1177/2158244014534858

DO - 10.1177/2158244014534858

M3 - Article

AN - SCOPUS:84907289366

VL - 4

JO - SAGE Open

JF - SAGE Open

SN - 2158-2440

IS - 2

ER -