Change score or follow-up score? Choice of mean difference estimates could impact meta-analysis conclusions

Rongwei (Rochelle) Fu, Haley K. Holmer

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Objectives: In randomized controlled clinical trials, continuous outcomes are typically measured at both baseline and follow-up, and mean difference could be estimated using the change scores from baseline or the follow-up scores. This study assesses the impact of using change score vs. follow-up score on the conclusions of meta-analyses. Study Design and Setting: A total of 63 meta-analyses from six comparative effectiveness reviews were included. The combined mean difference was estimated using a random-effects model, and we also evaluated whether the impact qualitatively varied by alternative random-effects estimates. Results: Based on the Dersimonian-Laird (DL) method, using the change vs. the follow-up score led to five meta-analyses (7.9%) showing discrepancy in conclusions. Based on the profile likelihood (PL) method, nine (14.3%) showed discrepancy in conclusions. Using change score was more likely to show a significant difference in effects between interventions (DL method: 4 of 5; PL method: 7 of 9). A significant difference in baseline scores did not necessarily lead to discrepancies in conclusions. Conclusions: Using the change vs. the follow-up score could lead to important discrepancies in conclusions. Sensitivity analyses should be conducted to check the robustness of results to the choice of mean difference estimates.

Original languageEnglish (US)
JournalJournal of Clinical Epidemiology
DOIs
StateAccepted/In press - 2016

Fingerprint

Meta-Analysis
Randomized Controlled Trials

Keywords

  • Baseline difference
  • Change score
  • Follow-up score
  • Mean difference
  • Meta-analysis
  • Random-effects estimates

ASJC Scopus subject areas

  • Epidemiology

Cite this

@article{ff4a1e8525eb4401ad7379375ed337d9,
title = "Change score or follow-up score? Choice of mean difference estimates could impact meta-analysis conclusions",
abstract = "Objectives: In randomized controlled clinical trials, continuous outcomes are typically measured at both baseline and follow-up, and mean difference could be estimated using the change scores from baseline or the follow-up scores. This study assesses the impact of using change score vs. follow-up score on the conclusions of meta-analyses. Study Design and Setting: A total of 63 meta-analyses from six comparative effectiveness reviews were included. The combined mean difference was estimated using a random-effects model, and we also evaluated whether the impact qualitatively varied by alternative random-effects estimates. Results: Based on the Dersimonian-Laird (DL) method, using the change vs. the follow-up score led to five meta-analyses (7.9{\%}) showing discrepancy in conclusions. Based on the profile likelihood (PL) method, nine (14.3{\%}) showed discrepancy in conclusions. Using change score was more likely to show a significant difference in effects between interventions (DL method: 4 of 5; PL method: 7 of 9). A significant difference in baseline scores did not necessarily lead to discrepancies in conclusions. Conclusions: Using the change vs. the follow-up score could lead to important discrepancies in conclusions. Sensitivity analyses should be conducted to check the robustness of results to the choice of mean difference estimates.",
keywords = "Baseline difference, Change score, Follow-up score, Mean difference, Meta-analysis, Random-effects estimates",
author = "Fu, {Rongwei (Rochelle)} and Holmer, {Haley K.}",
year = "2016",
doi = "10.1016/j.jclinepi.2016.01.034",
language = "English (US)",
journal = "Journal of Clinical Epidemiology",
issn = "0895-4356",
publisher = "Elsevier USA",

}

TY - JOUR

T1 - Change score or follow-up score? Choice of mean difference estimates could impact meta-analysis conclusions

AU - Fu, Rongwei (Rochelle)

AU - Holmer, Haley K.

PY - 2016

Y1 - 2016

N2 - Objectives: In randomized controlled clinical trials, continuous outcomes are typically measured at both baseline and follow-up, and mean difference could be estimated using the change scores from baseline or the follow-up scores. This study assesses the impact of using change score vs. follow-up score on the conclusions of meta-analyses. Study Design and Setting: A total of 63 meta-analyses from six comparative effectiveness reviews were included. The combined mean difference was estimated using a random-effects model, and we also evaluated whether the impact qualitatively varied by alternative random-effects estimates. Results: Based on the Dersimonian-Laird (DL) method, using the change vs. the follow-up score led to five meta-analyses (7.9%) showing discrepancy in conclusions. Based on the profile likelihood (PL) method, nine (14.3%) showed discrepancy in conclusions. Using change score was more likely to show a significant difference in effects between interventions (DL method: 4 of 5; PL method: 7 of 9). A significant difference in baseline scores did not necessarily lead to discrepancies in conclusions. Conclusions: Using the change vs. the follow-up score could lead to important discrepancies in conclusions. Sensitivity analyses should be conducted to check the robustness of results to the choice of mean difference estimates.

AB - Objectives: In randomized controlled clinical trials, continuous outcomes are typically measured at both baseline and follow-up, and mean difference could be estimated using the change scores from baseline or the follow-up scores. This study assesses the impact of using change score vs. follow-up score on the conclusions of meta-analyses. Study Design and Setting: A total of 63 meta-analyses from six comparative effectiveness reviews were included. The combined mean difference was estimated using a random-effects model, and we also evaluated whether the impact qualitatively varied by alternative random-effects estimates. Results: Based on the Dersimonian-Laird (DL) method, using the change vs. the follow-up score led to five meta-analyses (7.9%) showing discrepancy in conclusions. Based on the profile likelihood (PL) method, nine (14.3%) showed discrepancy in conclusions. Using change score was more likely to show a significant difference in effects between interventions (DL method: 4 of 5; PL method: 7 of 9). A significant difference in baseline scores did not necessarily lead to discrepancies in conclusions. Conclusions: Using the change vs. the follow-up score could lead to important discrepancies in conclusions. Sensitivity analyses should be conducted to check the robustness of results to the choice of mean difference estimates.

KW - Baseline difference

KW - Change score

KW - Follow-up score

KW - Mean difference

KW - Meta-analysis

KW - Random-effects estimates

UR - http://www.scopus.com/inward/record.url?scp=84964305223&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964305223&partnerID=8YFLogxK

U2 - 10.1016/j.jclinepi.2016.01.034

DO - 10.1016/j.jclinepi.2016.01.034

M3 - Article

C2 - 26931293

AN - SCOPUS:84964305223

JO - Journal of Clinical Epidemiology

JF - Journal of Clinical Epidemiology

SN - 0895-4356

ER -