How Large Should a Clinical Trial Be?

Gittins, John; Pezeshk, Hamid

doi:10.1111/1467-9884.00228

Cited by 46 publications

(51 citation statements)

References 8 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If a drug company perspective is taken, then the EVSI must be replaced by the company's expected gain. Gittens and Pezeshk [13,14] and Pezeshk and Gittens [18] where is the proÿt from treating a patient with T and k R (·) is the probability of regulatory approval. Gittens and Pezeshk [13,14] and Pezeshk and Gittens [18] suggest functions of the …”

Section: Discussionmentioning

confidence: 99%

The value of information and optimal clinical trial design

2005

View full text Add to dashboard Cite

Traditional sample size calculations for randomized clinical trials depend on somewhat arbitrarily chosen factors, such as type I and II errors. Type I error, the probability of rejecting the null hypothesis of no difference when it is true, is most often set to 0.05, regardless of the cost of such an error. In addition, the traditional use of 0.2 for the type II error means that the money and effort spent on the trial will be wasted 20 per cent of the time, even when the true treatment difference is equal to the smallest clinically important one and, again, will not reflect the cost of making such an error. An effectiveness trial (otherwise known as a pragmatic trial or management trial) is essentially an effort to inform decision-making, i.e. should treatment be adopted over standard? As such, a decision theoretic approach will lead to an optimal sample size determination. Using incremental net benefit and the theory of the expected value of information, and taking a societal perspective, it is shown how to determine the sample size that maximizes the difference between the cost of doing the trial and the value of the information gained from the results. The methods are illustrated using examples from oncology and obstetrics.

show abstract

Section: Discussionmentioning

confidence: 99%

The value of information and optimal clinical trial design

2005

View full text Add to dashboard Cite

show abstract

“…In some circumstances, it will be reasonable to construct benefit utilities which relate directly to the effectiveness of potential future treatments in a well-defined clinical context, e.g. Gittins and Pezeshk (2000), Walker (2003). Such utility specifications will typically be highly problem specific and usually will require great care in taming the computational complexity required to identify good designs, as Bayesian design choice is a notoriously computer intensive problem, often requiring simulations (e.g.…”

Section: Costs and Benefitsmentioning

confidence: 99%

“…Lindley (1997), Bernardo (1997) and Adcock (1997) give a useful introduction to the area. Recent examples include Walker (2003), whose utility involves the consequence of an action such as choice of treatment, for the case of a simple random 2 sample, and Gittins and Pezeshk (2000), who propose both a "public health" utility, involving both the benefit of a new treatment and the number who will use it as well as costs, and a "commercial" utility, for the case of a clinical trial to compare two treatments. For a Bayes linear approach to finding sample sizes to obtain specified reductions in variance, see Goldstein and Wooff (1997) and Shaw and Goldstein (1999).…”

Section: Introductionmentioning

confidence: 99%

Trade-off sensitive experimental design: a multicriterion, decision theoretic, Bayes linear approach

Farrow¹,

Goldstein²

2006

Journal of Statistical Planning and Inference

View full text Add to dashboard Cite

Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details. We show how mutually utility independent hierarchies, which weigh the various costs of an experiment against benefits expressed through a mixed Bayes linear utility representing the potential gains in knowledge from the experiment, provide a flexible and intuitive methodology for experimental design which remains tractable even for complex multivariate problems. A key feature of the approach is that we allow imprecision in the trade-offs between the various costs and benefits. We identify the Pareto optimal designs under the imprecise specification and suggest a criterion for selecting between such designs. The approach is illustrated with respect to an experiment related to the oral glucose tolerance test.

show abstract

“…Pezeshk and Gittins (1999) introduce a simplified version of the problem of sample size determination for a clinical trial for which the solution may be expressed in algebraic terms. They also extend their fully Bayesian model to more realistic cases (see, e.g., Gittins and Pezeshk, 2000;Pezeshk and Gittins, 2006;Kikuchi et al, 2008;M'Lan et al, 2008). Moreover, a simulation-based method for sample size determination using hierarchical models is presented in Gelfand and Wang (2002).…”

Section: Introductionmentioning

confidence: 99%

An Optimization Approach to Calculating Sample Sizes With Binary Responses

Maroufy

Marriott

Pezeshk

2014

Journal of Biopharmaceutical Statistics

Self Cite

View full text Add to dashboard Cite

In this article, we discuss an optimization approach to the sample size question, founded on maximizing the value of information in comparison studies with binary responses. The expected value of perfect information (EVPI) is calculated and the optimal sample size is obtained by maximizing the expected net gain of sampling (ENGS), the difference between the expected value of sample information (EVSI) and the cost of conducting the trial. The data are assumed to come from two independent binomial distributions, while the parameter of interest is the difference between the two success probabilities, [Formula: see text]. To formulate our prior knowledge on the parameters, a Dirichlet prior is used. Monte Carlo integration is used in the computation and optimization of ENGS. We also compare the results of this approach with existing Bayesian methods and show how the new approach reduces the computational complexity considerably.

show abstract

How Large Should a Clinical Trial Be?

Cited by 46 publications

References 8 publications

The value of information and optimal clinical trial design

The value of information and optimal clinical trial design

Trade-off sensitive experimental design: a multicriterion, decision theoretic, Bayes linear approach

An Optimization Approach to Calculating Sample Sizes With Binary Responses

Contact Info

Product

Resources

About