Skip Navigation

Biometrika 1999 86(2):365-379; doi:10.1093/biomet/86.2.365
© 1999 by Biometrika Trust
This Article
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Frangakis, C.
Right arrow Articles by Rubin, D.
Right arrow Search for Related Content
Social Bookmarking
 Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Addressing complications of intention-to-treat analysis in the combined presence of all-or-none treatment-noncompliance and subsequent missing outcomes

CE FrangakisA1 and DB RubinA2

Department of Statistics, Harvard University, 1 Oxford St, Cambridge, MA 02138, USA A1 E-mail: frangaki@hustat.harvard.edu A2 E-mail: rubin@stat.harvard.edu

We study the combined impact that all-or-none compliance and subsequent missing outcomes can have on the estimation of the intention-to-treat effect of assignment in randomised studies. In this setting, a standard analysis, which drops subjects with missing outcomes and ignores compliance information, can be biased for the intention-to-treat effect. To address all-or-none compliance that is followed by missing outcomes, we construct a new estimation procedure for the intention-to-treat effect that maintains good randomisation-based properties under more plausible, nonignorable noncompliance and nonignorable missing-outcome conditions: the 'compound exclusion restriction' on the effect of assignment and the 'latent ignorability' of the missing data mechanism. We present both theoretical results and a simulation study. Moreover, we show how the two key concepts of compound exclusion and latent ignorability are relevant in more complicated settings, such as right censoring of a time-to-event outcome.

Key Words: Compound exclusion restriction; Intention-to-treat; Latent ignorability; Noncompliance; Nonignorability; Rubin causal model.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
Clin TrialsHome page
L. Comte, S. Vansteelandt, E. Tousset, G. Baxter, and B. Vrijens
Linear and loglinear structural mean models to evaluate the benefits of an on-demand dosing regimen
Clinical Trials, October 1, 2009; 6(5): 403 - 415.
[Abstract] [PDF]


Home page
BiometrikaHome page
O. Harel and J. L. Schafer
Partial and latent ignorability in missing-data problems
Biometrika, March 1, 2009; 96(1): 37 - 50.
[Abstract] [PDF]


Home page
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICSHome page
B. Jo
Bias Mechanisms in Intention-to-Treat Analysis With Data Subject to Treatment Noncompliance and Missing Outcomes
Journal of Educational and Behavioral Statistics, June 1, 2008; 33(2): 158 - 185.
[Abstract] [Full Text] [PDF]


Home page
Clin TrialsHome page
S. L Bellamy, J. Y Lin, and T. R T. Have
An introduction to causal modeling in clinical trials
Clinical Trials, February 1, 2007; 4(1): 58 - 73.
[Abstract] [PDF]


Home page
J. Am. Soc. Nephrol.Home page
T. Greene, J. Daugirdas, T. Depner, M. Allon, G. Beck, C. Chumlea, J. Delmez, F. Gotch, J. W. Kusek, N. Levin, et al.
Association of Achieved Dialysis Dose with Mortality in the Hemodialysis Study: An Example of "Dose-Targeting Bias"
J. Am. Soc. Nephrol., November 1, 2005; 16(11): 3371 - 3380.
[Abstract] [Full Text] [PDF]


Home page
Stat Methods Med ResHome page
I. R White
Uses and limitations of randomization-based efficacy estimators
Statistical Methods in Medical Research, August 1, 2005; 14(4): 327 - 347.
[Abstract] [PDF]


Home page
Stat Methods Med ResHome page
S. G Baker and B. S Kramer
Simple maximum likelihood estimates of efficacy in randomized trials and before-and-after studies, with implications for meta-analysis
Statistical Methods in Medical Research, August 1, 2005; 14(4): 349 - 367.
[Abstract] [PDF]


Home page
Stat Methods Med ResHome page
G. Dunn, M. Maracy, and B. Tomenson
Estimating treatment effects from randomized clinical trials with noncompliance and loss to follow-up: the role of instrumental variable methods
Statistical Methods in Medical Research, August 1, 2005; 14(4): 369 - 395.
[Abstract] [PDF]


Home page
Stat Methods Med ResHome page
F. Li and C. E. Frangakis
Designs in partially controlled studies: messages from a review
Statistical Methods in Medical Research, August 1, 2005; 14(4): 417 - 431.
[Abstract] [PDF]


Home page
Br. J. PsychiatryHome page
G. DUNN, M. MARACY, C. DOWRICK, J. L. AYUSO-MATEOS, O. S. DALGARD, H. PAGE, V. LEHTINEN, P. CASEY, C. WILKINSON, J. L. VAZQUEZ-BARQUERO, et al.
Estimating psychological treatment effects from a randomised controlled trial with both non-compliance and loss to follow-up
The British Journal of Psychiatry, October 1, 2003; 183(4): 323 - 331.
[Abstract] [Full Text] [PDF]


Home page
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICSHome page
F. Mealli and D. B. Rubin
Discussion of "Estimation of Intervention Effects with Noncompliance: Alternative Model Specifications" by Booil Jo
Journal of Educational and Behavioral Statistics, January 1, 2002; 27(4): 411 - 415.
[PDF]


Home page
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICSHome page
J. L. Hill
Accommodating Missing Data in Mixture Models for Classification by Opinion-Changing Behavior
Journal of Educational and Behavioral Statistics, January 1, 2001; 26(2): 233 - 268.
[Abstract] [PDF]


Home page
Am J EpidemiolHome page
A. I. Sashegyi, K. Stephen Brown, and P. J. Farrell
Application of a Generalized Random Effects Regression Model for Cluster-correlated Longitudinal Data to a School-based Smoking Prevention Trial
Am. J. Epidemiol., December 15, 2000; 152(12): 1192 - 1200.
[Abstract] [Full Text] [PDF]


Home page
Stat Methods Med ResHome page
L. Sheiner and J. Wakefield
Population modelling in drug development
Statistical Methods in Medical Research, June 1, 1999; 8(3): 183 - 193.
[Abstract] [PDF]



Disclaimer: Please note that abstracts for content published before 1996 were created through digital scanning and may therefore not exactly replicate the text of the original print issues. All efforts have been made to ensure accuracy, but the Publisher will not be held responsible for any remaining inaccuracies. If you require any further clarification, please contact our Customer Services Department.