Deviation from intention to treat analysis in randomised trials and treatment effect estimates: Meta-epidemiological study

Iosief Abraha, Antonio Cherubini, Francesco Cozzolino, Rita De Florio, Maria Laura Luchetta, Joseph M. Rimland, Ilenia Folletti, Mauro Marchesi, Antonella Germani, Massimiliano Orso, Paolo Eusebi, Alessandro Montedori

Research output: Contribution to journalArticle

62 Citations (Scopus)

Abstract

Objective: To examine whether deviation from the standard intention to treat analysis has an influence on treatment effect estimates of randomised trials. Design: Meta-epidemiological study. Data sources: Medline, via PubMed, searched between 2006 and 2010; 43 systematic reviews of interventions and 310 randomised trials were included. Eligibility criteria for selecting studies: From each year searched, random selection of 5% of intervention reviews with a meta-analysis that included at least one trial that deviated from the standard intention to treat approach. Basic characteristics of the systematic reviews and randomised trials were extracted. Information on the reporting of intention to treat analysis, outcome data, risk of bias items, post-randomisation exclusions, and funding were extracted from each trial. Trials were classified as: ITT (reporting the standard intention to treat approach), mITT (reporting a deviation from the standard approach), and no ITT (reporting no approach). Within each meta-analysis, treatment effects were compared between mITT and ITT trials, and between mITT and no ITT trials. The ratio of odds ratios was calculated (value 2=0.13). Adjustments for sample size, type of centre, funding, items of risk of bias, post-randomisation exclusions, and variance of log odds ratio yielded consistent results (0.80 (0.69 to 0.94), P=0.005; τ2=0.08). After exclusion of five influential studies, results remained consistent (0.85 (0.75 to 0.98); τ2=0.08). The comparison between mITT trials and no ITT trials showed no statistical difference between the two groups (adjusted ratio of odds ratios 0.92 (0.70 to 1.23); τ2=0.57). Conclusions: Trials that deviated from the intention to treat analysis showed larger intervention effects than trials that reported the standard approach. Where an intention to treat analysis is impossible to perform, authors should clearly report who is included in the analysis and attempt to perform multiple imputations.

Original languageEnglish
Article numberh2445
JournalBritish Medical Journal
Volume350
DOIs
Publication statusPublished - May 27 2015

Fingerprint

Intention to Treat Analysis
Epidemiologic Studies
Odds Ratio
Random Allocation
Meta-Analysis
Information Storage and Retrieval
PubMed
Sample Size

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Deviation from intention to treat analysis in randomised trials and treatment effect estimates : Meta-epidemiological study. / Abraha, Iosief; Cherubini, Antonio; Cozzolino, Francesco; De Florio, Rita; Luchetta, Maria Laura; Rimland, Joseph M.; Folletti, Ilenia; Marchesi, Mauro; Germani, Antonella; Orso, Massimiliano; Eusebi, Paolo; Montedori, Alessandro.

In: British Medical Journal, Vol. 350, h2445, 27.05.2015.

Research output: Contribution to journalArticle

Abraha, I, Cherubini, A, Cozzolino, F, De Florio, R, Luchetta, ML, Rimland, JM, Folletti, I, Marchesi, M, Germani, A, Orso, M, Eusebi, P & Montedori, A 2015, 'Deviation from intention to treat analysis in randomised trials and treatment effect estimates: Meta-epidemiological study', British Medical Journal, vol. 350, h2445. https://doi.org/10.1136/bmj.h2445
Abraha, Iosief ; Cherubini, Antonio ; Cozzolino, Francesco ; De Florio, Rita ; Luchetta, Maria Laura ; Rimland, Joseph M. ; Folletti, Ilenia ; Marchesi, Mauro ; Germani, Antonella ; Orso, Massimiliano ; Eusebi, Paolo ; Montedori, Alessandro. / Deviation from intention to treat analysis in randomised trials and treatment effect estimates : Meta-epidemiological study. In: British Medical Journal. 2015 ; Vol. 350.
@article{1dbeef195d484fc7837fab8241bc14f0,
title = "Deviation from intention to treat analysis in randomised trials and treatment effect estimates: Meta-epidemiological study",
abstract = "Objective: To examine whether deviation from the standard intention to treat analysis has an influence on treatment effect estimates of randomised trials. Design: Meta-epidemiological study. Data sources: Medline, via PubMed, searched between 2006 and 2010; 43 systematic reviews of interventions and 310 randomised trials were included. Eligibility criteria for selecting studies: From each year searched, random selection of 5{\%} of intervention reviews with a meta-analysis that included at least one trial that deviated from the standard intention to treat approach. Basic characteristics of the systematic reviews and randomised trials were extracted. Information on the reporting of intention to treat analysis, outcome data, risk of bias items, post-randomisation exclusions, and funding were extracted from each trial. Trials were classified as: ITT (reporting the standard intention to treat approach), mITT (reporting a deviation from the standard approach), and no ITT (reporting no approach). Within each meta-analysis, treatment effects were compared between mITT and ITT trials, and between mITT and no ITT trials. The ratio of odds ratios was calculated (value 2=0.13). Adjustments for sample size, type of centre, funding, items of risk of bias, post-randomisation exclusions, and variance of log odds ratio yielded consistent results (0.80 (0.69 to 0.94), P=0.005; τ2=0.08). After exclusion of five influential studies, results remained consistent (0.85 (0.75 to 0.98); τ2=0.08). The comparison between mITT trials and no ITT trials showed no statistical difference between the two groups (adjusted ratio of odds ratios 0.92 (0.70 to 1.23); τ2=0.57). Conclusions: Trials that deviated from the intention to treat analysis showed larger intervention effects than trials that reported the standard approach. Where an intention to treat analysis is impossible to perform, authors should clearly report who is included in the analysis and attempt to perform multiple imputations.",
author = "Iosief Abraha and Antonio Cherubini and Francesco Cozzolino and {De Florio}, Rita and Luchetta, {Maria Laura} and Rimland, {Joseph M.} and Ilenia Folletti and Mauro Marchesi and Antonella Germani and Massimiliano Orso and Paolo Eusebi and Alessandro Montedori",
year = "2015",
month = "5",
day = "27",
doi = "10.1136/bmj.h2445",
language = "English",
volume = "350",
journal = "British Medical Journal",
issn = "0959-8146",
publisher = "BMJ Publishing Group",

}

TY - JOUR

T1 - Deviation from intention to treat analysis in randomised trials and treatment effect estimates

T2 - Meta-epidemiological study

AU - Abraha, Iosief

AU - Cherubini, Antonio

AU - Cozzolino, Francesco

AU - De Florio, Rita

AU - Luchetta, Maria Laura

AU - Rimland, Joseph M.

AU - Folletti, Ilenia

AU - Marchesi, Mauro

AU - Germani, Antonella

AU - Orso, Massimiliano

AU - Eusebi, Paolo

AU - Montedori, Alessandro

PY - 2015/5/27

Y1 - 2015/5/27

N2 - Objective: To examine whether deviation from the standard intention to treat analysis has an influence on treatment effect estimates of randomised trials. Design: Meta-epidemiological study. Data sources: Medline, via PubMed, searched between 2006 and 2010; 43 systematic reviews of interventions and 310 randomised trials were included. Eligibility criteria for selecting studies: From each year searched, random selection of 5% of intervention reviews with a meta-analysis that included at least one trial that deviated from the standard intention to treat approach. Basic characteristics of the systematic reviews and randomised trials were extracted. Information on the reporting of intention to treat analysis, outcome data, risk of bias items, post-randomisation exclusions, and funding were extracted from each trial. Trials were classified as: ITT (reporting the standard intention to treat approach), mITT (reporting a deviation from the standard approach), and no ITT (reporting no approach). Within each meta-analysis, treatment effects were compared between mITT and ITT trials, and between mITT and no ITT trials. The ratio of odds ratios was calculated (value 2=0.13). Adjustments for sample size, type of centre, funding, items of risk of bias, post-randomisation exclusions, and variance of log odds ratio yielded consistent results (0.80 (0.69 to 0.94), P=0.005; τ2=0.08). After exclusion of five influential studies, results remained consistent (0.85 (0.75 to 0.98); τ2=0.08). The comparison between mITT trials and no ITT trials showed no statistical difference between the two groups (adjusted ratio of odds ratios 0.92 (0.70 to 1.23); τ2=0.57). Conclusions: Trials that deviated from the intention to treat analysis showed larger intervention effects than trials that reported the standard approach. Where an intention to treat analysis is impossible to perform, authors should clearly report who is included in the analysis and attempt to perform multiple imputations.

AB - Objective: To examine whether deviation from the standard intention to treat analysis has an influence on treatment effect estimates of randomised trials. Design: Meta-epidemiological study. Data sources: Medline, via PubMed, searched between 2006 and 2010; 43 systematic reviews of interventions and 310 randomised trials were included. Eligibility criteria for selecting studies: From each year searched, random selection of 5% of intervention reviews with a meta-analysis that included at least one trial that deviated from the standard intention to treat approach. Basic characteristics of the systematic reviews and randomised trials were extracted. Information on the reporting of intention to treat analysis, outcome data, risk of bias items, post-randomisation exclusions, and funding were extracted from each trial. Trials were classified as: ITT (reporting the standard intention to treat approach), mITT (reporting a deviation from the standard approach), and no ITT (reporting no approach). Within each meta-analysis, treatment effects were compared between mITT and ITT trials, and between mITT and no ITT trials. The ratio of odds ratios was calculated (value 2=0.13). Adjustments for sample size, type of centre, funding, items of risk of bias, post-randomisation exclusions, and variance of log odds ratio yielded consistent results (0.80 (0.69 to 0.94), P=0.005; τ2=0.08). After exclusion of five influential studies, results remained consistent (0.85 (0.75 to 0.98); τ2=0.08). The comparison between mITT trials and no ITT trials showed no statistical difference between the two groups (adjusted ratio of odds ratios 0.92 (0.70 to 1.23); τ2=0.57). Conclusions: Trials that deviated from the intention to treat analysis showed larger intervention effects than trials that reported the standard approach. Where an intention to treat analysis is impossible to perform, authors should clearly report who is included in the analysis and attempt to perform multiple imputations.

UR - http://www.scopus.com/inward/record.url?scp=84936818702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936818702&partnerID=8YFLogxK

U2 - 10.1136/bmj.h2445

DO - 10.1136/bmj.h2445

M3 - Article

C2 - 26016488

AN - SCOPUS:84936818702

VL - 350

JO - British Medical Journal

JF - British Medical Journal

SN - 0959-8146

M1 - h2445

ER -