Using missing data methods to enable well informed interim decision making in adaptive clinical trials
Applications are invited for an exciting PhD topic on addressing missing data arising in adaptive clinical trials, working under the supervision of Dr. Robin Mitra (Cardiff University), Dr. Kirstin Stokorb (Cardiff University), and Dr. Thomas Burnett (University of Bath).
Adaptive designs for clinical trials add flexibility and efficiency to the clinical development process through the addition of interim analyses, allowing pre-planned alterations to a trial in progress. The benefits of such flexibility have been highlighted during the Covid-19 pandemic with a large amount of uncertainty while designing trials.
In most trials there is a delay between administration of the treatment and observing the patient’s response. Thus, clinical end points will not be observed for all recruited when interim decisions need to be made. It is crucial to make well informed decisions at the interim analyses to ensure the best treatments are made available to patients, while avoiding delays to the trial.
Typically, either ad hoc methods are used or very strong distributional assumptions need to be made, often implausible in practice. You will explore this problem from a missing data perspective, developing principled statistical methodology to improve the quality of decision making in adaptive clinical trials, through the incorporation of all available information. You will also engage with the project partners in the pharmaceutical industry, collaborating with statisticians in Roche and AstraZeneca, paving the way in developing methodology that is suitable in practical settings.
Year 1 – Review relevant literature; construct general framework for the research in collaboration with the project partners in the pharmaceutical industry; identify limitations with the current approaches in this setting, both theoretically and empirically (via the simulations).
Year 2 – Develop new methodology appropriate to this setting utilising missing data techniques; rigorously evaluate the performance of the proposed methodology, comparing and contrasting this to the performance of the existing methods; begin dissemination of the research to the relevant communities; prepare a paper for publication in a reputable Statistics Journal.
Year 3 – Determine mathematical properties of the proposed methodology; identify theoretical advantages the proposed methods offer over existing methods; consider possible extensions to the research findings so far; disseminate the research extensively via presentations and preparing further publications; prepare the thesis.
You will be based in the Statistics group within the School of Mathematics. The group has a diverse range of research interests including experimental design and missing data, and you will benefit from group interactions. There are also close links between the Schools of Mathematics and Computer Science and Informatics via the Data Science Academy, and both Schools will be moving into a brand new building in Autumn 2021.
Cardiff University is also home to the Centre for Trials Research (CTR), a centre of excellence in adaptive trial design and this research would provide an excellent opportunity to establish synergies between the CTR and the Statistics group. You will also have the opportunity to engage with the vibrant academic community at the University of Bath.
The postgraduate research student community in the department comprise a diverse body with research interests across the whole spectrum of Mathematical Sciences. We particularly welcome applications from females and those belonging to minority ethnic groups.
Training and Development Opportunities
You will have the opportunity to attend various national training courses in Statistics. These will include courses run by the Academy for PhD Training in Statistics as well as the Statisticans in the Pharmaceutical Industry. The advanced methods in Statistics and Clinical Trials you will learn in these courses will allow you to develop key research skills appropriate for your PhD, as well as being useful for your career in general going forward.
You will also have various networking opportunities within both academic communities and the pharmaceutical industry. This will involve attending and presenting at workshops, conferences, and seminars. There will also be further career development opportunities through attending events such as the Young Statisticians’ Meeting and the Research Students’ Conference.
More generally you will gain experience in teaching through assisting in delivering undergraduate and postgraduate courses. You will also gain experience in academic writing through preparing publications on your research and submitting these to reputable Statistics journals.
Interested applicants are encouraged to contact either Dr. Robin Mitra ([Email Address Removed]) or Dr. Thomas Burnett ([Email Address Removed]) prior to submitting their application to discuss this opportunity further.
HOW TO APPLY
Applicants should apply through the Cardiff University online application portal, for a Doctor of Philosophy in Mathematics with an entry point of October 2021
In the research proposal section of your application, please specify the project title and supervisors of this project.
There is no requirement to submit a research proposal
In the funding section, please select "I will be applying for a scholarship / grant" and specify that you are applying for advertised funding from EPRSC Maths DTP.