Causal inference in a clinical trial: A comparative example

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Recently there has been much interest in methods for analyzing clinical trials of treatments that are subject to noncompliance. In this paper I study a small, simple dataset from a clinical trial of immunosuppressive therapy in the treatment of multiple sclerosis. I apply and compare a range of methods: the as-randomized (intention-to-treat) analysis, the as-treated analysis, estimates based on a nonignorable selection model, and Rubin's causal model. The results differ substantially even in this small dataset that exhibits modest noncompliance. For this reason, data analysts should be clear about which parameters are of greatest importance in the analysis of a clinical trial. Control Clin Trials 1999;20:309-318 Copyright (C) 1999 Elsevier Science Inc.

Original languageEnglish (US)
Pages (from-to)309-318
Number of pages10
JournalControlled Clinical Trials
Volume20
Issue number4
DOIs
StatePublished - Jan 1 1999

Fingerprint

Clinical Trials
Intention to Treat Analysis
Immunosuppressive Agents
Multiple Sclerosis
Datasets
Therapeutics

Keywords

  • Intention-to-treat analysis
  • Noncompliance
  • Nonignorable model

ASJC Scopus subject areas

  • Pharmacology

Cite this

Causal inference in a clinical trial : A comparative example. / Heitjan, Daniel F.

In: Controlled Clinical Trials, Vol. 20, No. 4, 01.01.1999, p. 309-318.

Research output: Contribution to journalArticle

@article{3593ac9c0666445a8b90ebbde2ffb9bd,
title = "Causal inference in a clinical trial: A comparative example",
abstract = "Recently there has been much interest in methods for analyzing clinical trials of treatments that are subject to noncompliance. In this paper I study a small, simple dataset from a clinical trial of immunosuppressive therapy in the treatment of multiple sclerosis. I apply and compare a range of methods: the as-randomized (intention-to-treat) analysis, the as-treated analysis, estimates based on a nonignorable selection model, and Rubin's causal model. The results differ substantially even in this small dataset that exhibits modest noncompliance. For this reason, data analysts should be clear about which parameters are of greatest importance in the analysis of a clinical trial. Control Clin Trials 1999;20:309-318 Copyright (C) 1999 Elsevier Science Inc.",
keywords = "Intention-to-treat analysis, Noncompliance, Nonignorable model",
author = "Heitjan, {Daniel F.}",
year = "1999",
month = "1",
day = "1",
doi = "10.1016/S0197-2456(99)00012-4",
language = "English (US)",
volume = "20",
pages = "309--318",
journal = "Controlled Clinical Trials",
issn = "0197-2456",
publisher = "Elsevier BV",
number = "4",

}

TY - JOUR

T1 - Causal inference in a clinical trial

T2 - A comparative example

AU - Heitjan, Daniel F.

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Recently there has been much interest in methods for analyzing clinical trials of treatments that are subject to noncompliance. In this paper I study a small, simple dataset from a clinical trial of immunosuppressive therapy in the treatment of multiple sclerosis. I apply and compare a range of methods: the as-randomized (intention-to-treat) analysis, the as-treated analysis, estimates based on a nonignorable selection model, and Rubin's causal model. The results differ substantially even in this small dataset that exhibits modest noncompliance. For this reason, data analysts should be clear about which parameters are of greatest importance in the analysis of a clinical trial. Control Clin Trials 1999;20:309-318 Copyright (C) 1999 Elsevier Science Inc.

AB - Recently there has been much interest in methods for analyzing clinical trials of treatments that are subject to noncompliance. In this paper I study a small, simple dataset from a clinical trial of immunosuppressive therapy in the treatment of multiple sclerosis. I apply and compare a range of methods: the as-randomized (intention-to-treat) analysis, the as-treated analysis, estimates based on a nonignorable selection model, and Rubin's causal model. The results differ substantially even in this small dataset that exhibits modest noncompliance. For this reason, data analysts should be clear about which parameters are of greatest importance in the analysis of a clinical trial. Control Clin Trials 1999;20:309-318 Copyright (C) 1999 Elsevier Science Inc.

KW - Intention-to-treat analysis

KW - Noncompliance

KW - Nonignorable model

UR - http://www.scopus.com/inward/record.url?scp=0032988557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032988557&partnerID=8YFLogxK

U2 - 10.1016/S0197-2456(99)00012-4

DO - 10.1016/S0197-2456(99)00012-4

M3 - Article

C2 - 10440558

AN - SCOPUS:0032988557

VL - 20

SP - 309

EP - 318

JO - Controlled Clinical Trials

JF - Controlled Clinical Trials

SN - 0197-2456

IS - 4

ER -