Decision making during a Phase III randomized controlled trial

Donald A. Berry, Mark C. Wolff, David Allen Sack

Research output: Contribution to journalArticle

Abstract

Experiments such as clinical trials should be carried out with specific objectives. For example, in a trial designed to prevent disease, specific considerations should be made concerning the impact of the trial on the health of the target population, including the participants in the trial. These objectives should be assessed continually in light of data accumulating from the trial. Accumulating evidence should be judged in the context of changing circumstances external to the trial, and the trial's design possibly modified. An important type of modification is stopping the trial. This is a sequential decision problem that can be addressed using a Bayesian approach and the methods of dynamic programming. As an example we consider a vaccine trial for the prevention of haemophilus influenzae type b. The objective we consider is minimizing the number of cases of this disease in a Native American population over a specified horizon. We assess the prior probability distribution of vaccine efficacy. We also assess the probability of regulatory approval for widespread use of the vaccine, depending on the data presented to the regulatory officials. In deciding whether to continue the trial we weigh the impact of the possible future results by their (predictive) probabilities. We address the sensitivity of the optimal stopping policy to the priorprobability distribution, to the assessed probability of regulatory approval, and to the horizon.

Original languageEnglish (US)
Pages (from-to)360-378
Number of pages19
JournalControlled Clinical Trials
Volume15
Issue number5
DOIs
StatePublished - 1994

Fingerprint

Decision Making
Randomized Controlled Trials
Bayes Theorem
Vaccines
Haemophilus influenzae type b
North American Indians
Health Services Needs and Demand
Clinical Trials
Health
Population

Keywords

  • assessing prior information
  • decision analysis
  • dynamic programming
  • Monitoring randomized clinical trials
  • optimal stopping
  • sequential design
  • weighing information gain with effective therapy

ASJC Scopus subject areas

  • Pharmacology

Cite this

Decision making during a Phase III randomized controlled trial. / Berry, Donald A.; Wolff, Mark C.; Sack, David Allen.

In: Controlled Clinical Trials, Vol. 15, No. 5, 1994, p. 360-378.

Research output: Contribution to journalArticle

Berry, Donald A. ; Wolff, Mark C. ; Sack, David Allen. / Decision making during a Phase III randomized controlled trial. In: Controlled Clinical Trials. 1994 ; Vol. 15, No. 5. pp. 360-378.
@article{89d0dbbfbd8b413883183a9eb394e010,
title = "Decision making during a Phase III randomized controlled trial",
abstract = "Experiments such as clinical trials should be carried out with specific objectives. For example, in a trial designed to prevent disease, specific considerations should be made concerning the impact of the trial on the health of the target population, including the participants in the trial. These objectives should be assessed continually in light of data accumulating from the trial. Accumulating evidence should be judged in the context of changing circumstances external to the trial, and the trial's design possibly modified. An important type of modification is stopping the trial. This is a sequential decision problem that can be addressed using a Bayesian approach and the methods of dynamic programming. As an example we consider a vaccine trial for the prevention of haemophilus influenzae type b. The objective we consider is minimizing the number of cases of this disease in a Native American population over a specified horizon. We assess the prior probability distribution of vaccine efficacy. We also assess the probability of regulatory approval for widespread use of the vaccine, depending on the data presented to the regulatory officials. In deciding whether to continue the trial we weigh the impact of the possible future results by their (predictive) probabilities. We address the sensitivity of the optimal stopping policy to the priorprobability distribution, to the assessed probability of regulatory approval, and to the horizon.",
keywords = "assessing prior information, decision analysis, dynamic programming, Monitoring randomized clinical trials, optimal stopping, sequential design, weighing information gain with effective therapy",
author = "Berry, {Donald A.} and Wolff, {Mark C.} and Sack, {David Allen}",
year = "1994",
doi = "10.1016/0197-2456(94)90033-7",
language = "English (US)",
volume = "15",
pages = "360--378",
journal = "Controlled Clinical Trials",
issn = "0197-2456",
publisher = "Elsevier BV",
number = "5",

}

TY - JOUR

T1 - Decision making during a Phase III randomized controlled trial

AU - Berry, Donald A.

AU - Wolff, Mark C.

AU - Sack, David Allen

PY - 1994

Y1 - 1994

N2 - Experiments such as clinical trials should be carried out with specific objectives. For example, in a trial designed to prevent disease, specific considerations should be made concerning the impact of the trial on the health of the target population, including the participants in the trial. These objectives should be assessed continually in light of data accumulating from the trial. Accumulating evidence should be judged in the context of changing circumstances external to the trial, and the trial's design possibly modified. An important type of modification is stopping the trial. This is a sequential decision problem that can be addressed using a Bayesian approach and the methods of dynamic programming. As an example we consider a vaccine trial for the prevention of haemophilus influenzae type b. The objective we consider is minimizing the number of cases of this disease in a Native American population over a specified horizon. We assess the prior probability distribution of vaccine efficacy. We also assess the probability of regulatory approval for widespread use of the vaccine, depending on the data presented to the regulatory officials. In deciding whether to continue the trial we weigh the impact of the possible future results by their (predictive) probabilities. We address the sensitivity of the optimal stopping policy to the priorprobability distribution, to the assessed probability of regulatory approval, and to the horizon.

AB - Experiments such as clinical trials should be carried out with specific objectives. For example, in a trial designed to prevent disease, specific considerations should be made concerning the impact of the trial on the health of the target population, including the participants in the trial. These objectives should be assessed continually in light of data accumulating from the trial. Accumulating evidence should be judged in the context of changing circumstances external to the trial, and the trial's design possibly modified. An important type of modification is stopping the trial. This is a sequential decision problem that can be addressed using a Bayesian approach and the methods of dynamic programming. As an example we consider a vaccine trial for the prevention of haemophilus influenzae type b. The objective we consider is minimizing the number of cases of this disease in a Native American population over a specified horizon. We assess the prior probability distribution of vaccine efficacy. We also assess the probability of regulatory approval for widespread use of the vaccine, depending on the data presented to the regulatory officials. In deciding whether to continue the trial we weigh the impact of the possible future results by their (predictive) probabilities. We address the sensitivity of the optimal stopping policy to the priorprobability distribution, to the assessed probability of regulatory approval, and to the horizon.

KW - assessing prior information

KW - decision analysis

KW - dynamic programming

KW - Monitoring randomized clinical trials

KW - optimal stopping

KW - sequential design

KW - weighing information gain with effective therapy

UR - http://www.scopus.com/inward/record.url?scp=0028046097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028046097&partnerID=8YFLogxK

U2 - 10.1016/0197-2456(94)90033-7

DO - 10.1016/0197-2456(94)90033-7

M3 - Article

C2 - 8001357

AN - SCOPUS:0028046097

VL - 15

SP - 360

EP - 378

JO - Controlled Clinical Trials

JF - Controlled Clinical Trials

SN - 0197-2456

IS - 5

ER -