Weibull prediction of event times in clinical trials

Gui Shuang Ying, Daniel F. Heitjan

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

In clinical trials with interim analyses planned at pre-specified event counts, one may wish to predict the times of these landmark events as a tool for logistical planning. Currently available methods use either a parametric approach based on an exponential model for survival (Bagiella and Heitjan, Statistics in Medicine 2001; 20:2055) or a non-parametric approach based on the Kaplan-Meier estimate (Ying et al., Clinical Trials 2004; 1:352). Ying et al. (2004) demonstrated the trade-off between bias and variance in these models; the exponential method is highly efficient when its assumptions hold buy potentially biased when they do not, whereas the non-parametric method has minimal bias and is well calibrated under a range of survival models but typical gives wider prediction intervals and may fail to produce useful predictions early in the trial. As a potential compromise, we propose here to make predictions under a Weibull survival model. Computations are somewhat more difficult than with the simpler exponential model, but Monte Carlo studies show that predictions are robust under a broader range of assumptions. We demonstrate the method using data from a trial of immunotherapy for chronic granulomatous disease.

Original languageEnglish (US)
Pages (from-to)107-120
Number of pages14
JournalPharmaceutical Statistics
Volume7
Issue number2
DOIs
StatePublished - Apr 1 2008

Fingerprint

Weibull
Clinical Trials
Survival Model
Exponential Model
Prediction
Kaplan-Meier Estimate
Immunotherapy
Weibull Model
Chronic Disease
Prediction Interval
Nonparametric Methods
Monte Carlo Study
Landmarks
Chronic Granulomatous Disease
Range of data
Medicine
Biased
Count
Trade-offs
Planning

Keywords

  • Interim analysis
  • Predictie power
  • Time to event

ASJC Scopus subject areas

  • Pharmaceutical Science
  • Statistics and Probability

Cite this

Weibull prediction of event times in clinical trials. / Ying, Gui Shuang; Heitjan, Daniel F.

In: Pharmaceutical Statistics, Vol. 7, No. 2, 01.04.2008, p. 107-120.

Research output: Contribution to journalArticle

@article{bc2a9aeef71741308715515e7d4cb2f8,
title = "Weibull prediction of event times in clinical trials",
abstract = "In clinical trials with interim analyses planned at pre-specified event counts, one may wish to predict the times of these landmark events as a tool for logistical planning. Currently available methods use either a parametric approach based on an exponential model for survival (Bagiella and Heitjan, Statistics in Medicine 2001; 20:2055) or a non-parametric approach based on the Kaplan-Meier estimate (Ying et al., Clinical Trials 2004; 1:352). Ying et al. (2004) demonstrated the trade-off between bias and variance in these models; the exponential method is highly efficient when its assumptions hold buy potentially biased when they do not, whereas the non-parametric method has minimal bias and is well calibrated under a range of survival models but typical gives wider prediction intervals and may fail to produce useful predictions early in the trial. As a potential compromise, we propose here to make predictions under a Weibull survival model. Computations are somewhat more difficult than with the simpler exponential model, but Monte Carlo studies show that predictions are robust under a broader range of assumptions. We demonstrate the method using data from a trial of immunotherapy for chronic granulomatous disease.",
keywords = "Interim analysis, Predictie power, Time to event",
author = "Ying, {Gui Shuang} and Heitjan, {Daniel F.}",
year = "2008",
month = "4",
day = "1",
doi = "10.1002/pst.271",
language = "English (US)",
volume = "7",
pages = "107--120",
journal = "Pharmaceutical Statistics",
issn = "1539-1604",
publisher = "John Wiley and Sons Ltd",
number = "2",

}

TY - JOUR

T1 - Weibull prediction of event times in clinical trials

AU - Ying, Gui Shuang

AU - Heitjan, Daniel F.

PY - 2008/4/1

Y1 - 2008/4/1

N2 - In clinical trials with interim analyses planned at pre-specified event counts, one may wish to predict the times of these landmark events as a tool for logistical planning. Currently available methods use either a parametric approach based on an exponential model for survival (Bagiella and Heitjan, Statistics in Medicine 2001; 20:2055) or a non-parametric approach based on the Kaplan-Meier estimate (Ying et al., Clinical Trials 2004; 1:352). Ying et al. (2004) demonstrated the trade-off between bias and variance in these models; the exponential method is highly efficient when its assumptions hold buy potentially biased when they do not, whereas the non-parametric method has minimal bias and is well calibrated under a range of survival models but typical gives wider prediction intervals and may fail to produce useful predictions early in the trial. As a potential compromise, we propose here to make predictions under a Weibull survival model. Computations are somewhat more difficult than with the simpler exponential model, but Monte Carlo studies show that predictions are robust under a broader range of assumptions. We demonstrate the method using data from a trial of immunotherapy for chronic granulomatous disease.

AB - In clinical trials with interim analyses planned at pre-specified event counts, one may wish to predict the times of these landmark events as a tool for logistical planning. Currently available methods use either a parametric approach based on an exponential model for survival (Bagiella and Heitjan, Statistics in Medicine 2001; 20:2055) or a non-parametric approach based on the Kaplan-Meier estimate (Ying et al., Clinical Trials 2004; 1:352). Ying et al. (2004) demonstrated the trade-off between bias and variance in these models; the exponential method is highly efficient when its assumptions hold buy potentially biased when they do not, whereas the non-parametric method has minimal bias and is well calibrated under a range of survival models but typical gives wider prediction intervals and may fail to produce useful predictions early in the trial. As a potential compromise, we propose here to make predictions under a Weibull survival model. Computations are somewhat more difficult than with the simpler exponential model, but Monte Carlo studies show that predictions are robust under a broader range of assumptions. We demonstrate the method using data from a trial of immunotherapy for chronic granulomatous disease.

KW - Interim analysis

KW - Predictie power

KW - Time to event

UR - http://www.scopus.com/inward/record.url?scp=46449103064&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46449103064&partnerID=8YFLogxK

U2 - 10.1002/pst.271

DO - 10.1002/pst.271

M3 - Article

C2 - 17377932

AN - SCOPUS:46449103064

VL - 7

SP - 107

EP - 120

JO - Pharmaceutical Statistics

JF - Pharmaceutical Statistics

SN - 1539-1604

IS - 2

ER -