Comparison of data analysis strategies for intent-to-treat analysis in pre-test-post-test designs with substantial dropout rates - PubMed (original) (raw)
Comparative Study
. 2008 Sep 30;160(3):335-45.
doi: 10.1016/j.psychres.2007.08.005. Epub 2008 Aug 20.
Affiliations
- PMID: 18718673
- DOI: 10.1016/j.psychres.2007.08.005
Comparative Study
Comparison of data analysis strategies for intent-to-treat analysis in pre-test-post-test designs with substantial dropout rates
Agus Salim et al. Psychiatry Res. 2008.
Abstract
The pre-test-post-test design (PPD) is predominant in trials of psychotherapeutic treatments. Missing data due to withdrawals present an even bigger challenge in assessing treatment effectiveness under the PPD than under designs with more observations since dropout implies an absence of information about response to treatment. When confronted with missing data, often it is reasonable to assume that the mechanism underlying missingness is related to observed but not to unobserved outcomes (missing at random, MAR). Previous simulation and theoretical studies have shown that, under MAR, modern techniques such as maximum-likelihood (ML) based methods and multiple imputation (MI) can be used to produce unbiased estimates of treatment effects. In practice, however, ad hoc methods such as last observation carried forward (LOCF) imputation and complete-case (CC) analysis continue to be used. In order to better understand the behaviour of these methods in the PPD, we compare the performance of traditional approaches (LOCF, CC) and theoretically sound techniques (MI, ML), under various MAR mechanisms. We show that the LOCF method is seriously biased and conclude that its use should be abandoned. Complete-case analysis produces unbiased estimates only when the dropout mechanism does not depend on pre-test values even when dropout is related to fixed covariates including treatment group (covariate-dependent: CD). However, CC analysis is generally biased under MAR. The magnitude of the bias is largest when the correlation of post- and pre-test is relatively low.
Comment in
- Complete-cases analysis is appropriate for randomised trials with pre-test-post-test designs.
White IR, Horton N, Carpenter J. White IR, et al. Psychiatry Res. 2009 Aug 15;168(3):268; author reply 269. doi: 10.1016/j.psychres.2009.01.003. Epub 2009 Jun 25. Psychiatry Res. 2009. PMID: 19556011 No abstract available.
Similar articles
- Sensitivity analysis of intention-to-treat estimates when withdrawals are related to unobserved compliance status.
Salim A, Mackinnon A, Griffiths K. Salim A, et al. Stat Med. 2008 Apr 15;27(8):1164-79. doi: 10.1002/sim.3025. Stat Med. 2008. PMID: 17724782 - Comparisons of methods for analysis of repeated binary responses with missing data.
Frank Liu G, Zhan X. Frank Liu G, et al. J Biopharm Stat. 2011 May;21(3):371-92. doi: 10.1080/10543401003687129. J Biopharm Stat. 2011. PMID: 21442514 - Comparison of imputation and modelling methods in the analysis of a physical activity trial with missing outcomes.
Wood AM, White IR, Hillsdon M, Carpenter J. Wood AM, et al. Int J Epidemiol. 2005 Feb;34(1):89-99. doi: 10.1093/ije/dyh297. Epub 2004 Aug 27. Int J Epidemiol. 2005. PMID: 15333619 - An overview of practical approaches for handling missing data in clinical trials.
DeSouza CM, Legedza AT, Sankoh AJ. DeSouza CM, et al. J Biopharm Stat. 2009 Nov;19(6):1055-73. doi: 10.1080/10543400903242795. J Biopharm Stat. 2009. PMID: 20183464 Review. - Last-observation-carried-forward imputation method in clinical efficacy trials: review of 352 antidepressant studies.
Woolley SB, Cardoni AA, Goethe JW. Woolley SB, et al. Pharmacotherapy. 2009 Dec;29(12):1408-16. doi: 10.1592/phco.29.12.1408. Pharmacotherapy. 2009. PMID: 19947800 Review.
Cited by
- Effectiveness of an eHealth intervention for reducing psychological distress and increasing COVID-19 knowledge and protective behaviors among racialized sexual and gender minority adults: A quasi-experimental study (#SafeHandsSafeHearts).
Newman PA, Chakrapani V, Massaquoi N, Williams CC, Tharao W, Tepjan S, Roungprakhon S, Forbes J, Sebastian S, Akkakanjanasupar P, Aden M. Newman PA, et al. PLoS One. 2024 May 3;19(5):e0280710. doi: 10.1371/journal.pone.0280710. eCollection 2024. PLoS One. 2024. PMID: 38701074 Free PMC article. - Enhanced Skills Training in Affective and Interpersonal Regulation versus Treatment as Usual for ICD-11 Complex PTSD: A Pilot Randomised Controlled Trial (The RESTORE Trial).
Karatzias T, Shevlin M, Cloitre M, Busuttil W, Graham K, Hendrikx L, Hyland P, Biscoe N, Murphy D. Karatzias T, et al. Psychother Psychosom. 2024;93(3):203-215. doi: 10.1159/000538428. Epub 2024 Apr 30. Psychother Psychosom. 2024. PMID: 38688242 Free PMC article. Clinical Trial. - Testing the 'Faith Moves Mountains model' to increase Alzheimer's disease awareness, detection, and diagnosis among rural, racially, and ethnically diverse older adults.
Wiese LK, Williams IC, Holt JK, Williams CL, Lingler J, Galvin JE, Schoenberg NE. Wiese LK, et al. Aging Ment Health. 2024 Jun;28(6):943-956. doi: 10.1080/13607863.2023.2294062. Epub 2023 Dec 21. Aging Ment Health. 2024. PMID: 38127408 - Pulmonary Rehabilitation with and without a Cognitive Behavioral Intervention for Breathlessness in People Living with Chronic Obstructive Pulmonary Disease: Randomized Controlled Trial.
Williams MT, Lewthwaite H, Paquet C, Cafarella P, Frith P. Williams MT, et al. J Clin Med. 2023 Nov 24;12(23):7286. doi: 10.3390/jcm12237286. J Clin Med. 2023. PMID: 38068337 Free PMC article. - Cerebrolysin for acute ischaemic stroke.
Ziganshina LE, Abakumova T, Nurkhametova D, Ivanchenko K. Ziganshina LE, et al. Cochrane Database Syst Rev. 2023 Oct 11;10(10):CD007026. doi: 10.1002/14651858.CD007026.pub7. Cochrane Database Syst Rev. 2023. PMID: 37818733 Free PMC article. Review.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Medical