Predicting analysis times in randomized clinical trials

Emilia Bagiella, Daniel F. Heitjan

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Randomized clinical trial designs commonly include one or more planned interim analyses. At these times an external monitoring committee reviews the accumulated data and determines whether it is scientifically and ethically appropriate for the study to continue. With failure-time endpoints, it is common to schedule analyses at the times of occurrence of specified landmark events, such as the 50th event, the 100th event, and so on. Because interim analyses can impose considerable logistical burdens, it is worthwhile predicting their timing as accurately as possible. We describe two model-based methods for making such predictions during the course of a trial. First, we obtain a point prediction by extrapolating the cumulative mortality into the future and selecting the date when the expected number of deaths is equal to the landmark number. Second, we use a Bayesian simulation scheme to generate a predictive distribution of milestone times; prediction intervals are quantiles of this distribution. We illustrate our method with an analysis of data from a trial of immunotherapy in the treatment of chronic granulomatous disease.

Original languageEnglish (US)
Pages (from-to)2055-2063
Number of pages9
JournalStatistics in Medicine
Volume20
Issue number14
DOIs
StatePublished - Jul 30 2001

Fingerprint

Randomized Clinical Trial
Randomized Controlled Trials
Landmarks
Immunotherapy
Chronic Disease
Prediction Interval
Predictive Distribution
Prediction
Failure Time
Quantile
Chronic Granulomatous Disease
Date
Mortality
Timing
Schedule
Advisory Committees
Continue
Monitoring
Model-based
Appointments and Schedules

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Predicting analysis times in randomized clinical trials. / Bagiella, Emilia; Heitjan, Daniel F.

In: Statistics in Medicine, Vol. 20, No. 14, 30.07.2001, p. 2055-2063.

Research output: Contribution to journalArticle

@article{bfd2e03e574b4ca990c481afbd50a063,
title = "Predicting analysis times in randomized clinical trials",
abstract = "Randomized clinical trial designs commonly include one or more planned interim analyses. At these times an external monitoring committee reviews the accumulated data and determines whether it is scientifically and ethically appropriate for the study to continue. With failure-time endpoints, it is common to schedule analyses at the times of occurrence of specified landmark events, such as the 50th event, the 100th event, and so on. Because interim analyses can impose considerable logistical burdens, it is worthwhile predicting their timing as accurately as possible. We describe two model-based methods for making such predictions during the course of a trial. First, we obtain a point prediction by extrapolating the cumulative mortality into the future and selecting the date when the expected number of deaths is equal to the landmark number. Second, we use a Bayesian simulation scheme to generate a predictive distribution of milestone times; prediction intervals are quantiles of this distribution. We illustrate our method with an analysis of data from a trial of immunotherapy in the treatment of chronic granulomatous disease.",
author = "Emilia Bagiella and Heitjan, {Daniel F.}",
year = "2001",
month = "7",
day = "30",
doi = "10.1002/sim.843",
language = "English (US)",
volume = "20",
pages = "2055--2063",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "14",

}

TY - JOUR

T1 - Predicting analysis times in randomized clinical trials

AU - Bagiella, Emilia

AU - Heitjan, Daniel F.

PY - 2001/7/30

Y1 - 2001/7/30

N2 - Randomized clinical trial designs commonly include one or more planned interim analyses. At these times an external monitoring committee reviews the accumulated data and determines whether it is scientifically and ethically appropriate for the study to continue. With failure-time endpoints, it is common to schedule analyses at the times of occurrence of specified landmark events, such as the 50th event, the 100th event, and so on. Because interim analyses can impose considerable logistical burdens, it is worthwhile predicting their timing as accurately as possible. We describe two model-based methods for making such predictions during the course of a trial. First, we obtain a point prediction by extrapolating the cumulative mortality into the future and selecting the date when the expected number of deaths is equal to the landmark number. Second, we use a Bayesian simulation scheme to generate a predictive distribution of milestone times; prediction intervals are quantiles of this distribution. We illustrate our method with an analysis of data from a trial of immunotherapy in the treatment of chronic granulomatous disease.

AB - Randomized clinical trial designs commonly include one or more planned interim analyses. At these times an external monitoring committee reviews the accumulated data and determines whether it is scientifically and ethically appropriate for the study to continue. With failure-time endpoints, it is common to schedule analyses at the times of occurrence of specified landmark events, such as the 50th event, the 100th event, and so on. Because interim analyses can impose considerable logistical burdens, it is worthwhile predicting their timing as accurately as possible. We describe two model-based methods for making such predictions during the course of a trial. First, we obtain a point prediction by extrapolating the cumulative mortality into the future and selecting the date when the expected number of deaths is equal to the landmark number. Second, we use a Bayesian simulation scheme to generate a predictive distribution of milestone times; prediction intervals are quantiles of this distribution. We illustrate our method with an analysis of data from a trial of immunotherapy in the treatment of chronic granulomatous disease.

UR - http://www.scopus.com/inward/record.url?scp=0035974243&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035974243&partnerID=8YFLogxK

U2 - 10.1002/sim.843

DO - 10.1002/sim.843

M3 - Article

C2 - 11439420

AN - SCOPUS:0035974243

VL - 20

SP - 2055

EP - 2063

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 14

ER -