August 1, 2014

Statistical methods for adaptive clinical trials in tropical diseases

Oxford University Clinical Research Unit Vietnam PhD Programme 2014-15

Project 1


Statistical methods for adaptive clinical trials in tropical diseases

Project overview

Randomized clinical trials (RCTs) are the gold standard for testing medical interventions. Conventional RCTs randomly assign a pre-defined number of patients to one of two or more treatment arms (including a control arm) and evaluate a patient-relevant outcome during follow-up. At the end of the trial, when all patients have been randomized and evaluated, the outcomes are compared between the treatment groups based on statistical estimation and hypothesis testing.

Adaptive randomized trials are RCTs that allow for (pre-planned) adaptations while the trial is ongoing. Potential adaptations include dropping of study arms, changes in sample size or inclusion/exclusion criteria (to target the right population), early stopping for efficacy or futility, and changes in randomization ratios. Adaptive trials can be considerably more efficient in answering relevant research questions compared to conventional RCTs but are also more complicated to design and frequently require extensive trial simulations.

The purpose of this project is to develop statistical methodology as well as concept protocols for adaptive RCTs in infectious and tropical diseases. The PhD project will evaluate in detail methodological challenges and potential designs in several disease areas including typhoid, malaria, and tuberculous meningitis (TBM). Methodological challenges include the handling of multi-arm studies with efficient dropping of inefficacious arms, investigations regarding the value of adaptive randomization in diseases with changing resistance patterns, and the identification and efficient use of early markers of outcome in diseases where the primary endpoint requires substantial follow-up (e.g. 9-month mortality in TBM).

Training opportunities

General OUCRU PhD training programme

Extensive specialized training within the OUCRU biostatistics group

Possibility for overseas training (dependent on additional approval)




Dr Marcel Wolbers

Skills required in the candidate

MSc in biostatistics, statistics or mathematics

Strong interest and, ideally, experience in medical applications


Skip to content