Project title: Developing causal modelling methods for cluster randomised trials for complex interventions
Supervisors: Dr Chris Roberts & Dr Richard Emsley
Deadline: Monday 4th April
The objective of this fully-funded 4-year PhD project is to consider statistical methods for causal modelling in cluster randomised trials.
The studentship provides full support for tuition fees, all associated research costs and an annual tax-free stipend at Research Council rates (£13,590 in 2010). The project is due to commence October 2011 and is open to UK/EU nationals only due to the nature of the funding.
Causal modelling has been at the forefront of biostatistics research in recent decades, but its application to cluster randomised trials has so far been limited. Complex multi-component interventions imply a more complex causal pathway between intervention and outcome than pharmacological interventions. For example, in cluster randomised trials non-compliance may arise at the subject or cluster level and vary between components of the intervention.
The project will have four components:
*Review of methodology related to causal modelling in cluster randomised trials.
*A methodological review to ground the project in an area of application. The review will consider ways in which non-adherence in complex multi-component interventions may arise and how they may be measured. An area of application of cluster randomised trial methodology in public health will be selected (e.g. environmental and occupational hazards, obesity and exercise, substance abuse) which will enable dataset and data structures to be identified for the next stage.
*Variation in the causal treatment effect may bias intention to treat estimation (ITT). The implications of non-compliance for ITT estimation of treatment effects will be evaluated through simulation and analytical work.
*The development and testing of methods of causal modelling for cluster randomised trials will be carried out through simulation and analytical work.
The successful candidate will benefit from an extensive support network based in the Health Sciences Research Group, which has leading expertise in causal modelling and cluster randomised trials. Training will be provided in advanced hierarchical modelling techniques and methods for statistical programming and simulation.
Given the portfolio of biostatistical skills developed in this PhD, it will provide an ideal platform to progress onto a career path within post-doctoral research in a related area.
Applicants should hold (or expect to obtain) a minimum upper-second honours degree (or equivalent) in mathematics or statistics.
Please direct applications in the following format to Dr Chris Roberts ([log in to unmask]):
*Official academic transcripts
*Contact details for two suitable referees
*A personal statement (750 words maximum) outlining your suitability for the study, what you hope to achieve from the PhD and your research experience to date.
Any enquiries relating to the project and/or suitability should be directed to Dr Roberts at the address above. Applications are invited up to and including Monday 4 April 2011.
You may leave the list at any time by sending the command
to [log in to unmask], leaving the subject line blank.