Trials Methodology Research Partnership (TMRP) PhD Studentship, University of Nottingham 2020 (MED1560)

Host Centres

The TMRP studentship at the University of Nottingham is co-funded by the Nottingham Clinical Trials Unit (NCTU, www.nottingham.ac.uk/nctu), the Centre of Evidence-Based Dermatology (CEBD, www.nottinham.ac.uk/dermatology), and the School of Medicine. The studentship will be co-supervised by experienced academic staff in NCTU and CEBD. PhD projects are based primarily on application of methods to dermatology trials, although generalisability to trials in other clinical areas is also considered.

Nottingham Clinical Trials Unit (NCTU)

The Nottingham Clinical Trials Unit (NCTU) is a UK Clinical Research Collaboration registered CTU (http://www.ukcrc-ctu.org.uk). The unit’s mission is to conduct high quality, high impact multicentre trials to improve health and well-being. NCTU is based in the School of Medicine at the University of Nottingham, and has close links with local NHS Trusts, the Institute of Mental Health and the School of Health Sciences.

NCTU has a broad portfolio of trials including studies evaluating medical devices, investigational medicinal products and complex interventions. Particular strengths currently are trials of dermatology, obstetrics, gastroenterology, respiratory medicine, stroke, mental health, surgery (particularly hand surgery), emergency trials, and genitourinary medicine. Other research includes methodological studies to improve the efficiency and quality of trial conduct, and systematic reviews and meta-analysis.

The Unit comprises a multidisciplinary group of trial managers, statisticians, database programmers, data managers, and quality assurance managers, as well as experienced academic trialists. We welcome PhD students with an interest in improving how we do clinical trials by conducting high quality trials methodology research.

The Centre of Evidence Based Dermatology (CEBD)

The Centre of Evidence Based Dermatology (CEBD) is a centre of excellence for applied dermatology research. It combines the editorial base of the international Cochrane Skin Group (which reconciles clinical dilemmas and identifies research gaps), the UK Dermatology Clinical Trials Network (which conducts national clinical trials to address those gaps) and a programme of related methodological and epidemiological work which underpins our work.  

The CEBD is an extremely successful and productive research group drawing in around £1-2million of grant income per year and 4* publications in top journals. The main areas of interest for the group include childhood skin diseases, skin disease in older age, rare skin conditions, diagnosis and prevention of skin disease, and outcomes and diagnostics research. We are especially interested in independent research, which typically includes evidence synthesis, randomised controlled trials and associated methodological work into disease definitions and outcome measures. We ensure that patients and healthcare professionals are active partners in our work. (See www.nottingham.ac.uk/dermatology for more information).

We are looking for PhD students with an eagerness to learn and a commitment to conducting research of the highest methodological standards that would complement and expand existing skills at the Centre.

PhD Projects for 2020

1. Methods for handling missing data in patient-reported outcomes

Patient-reported outcome measures (PROMs) are frequently used in randomised controlled trials (RCTs) to assess the impact of treatment from the patient perspective. PROMs, being generally multi-item scales, are inevitably susceptible to missing data due to unanswered items.  Various approaches have been developed for handling missing data, with maximum likelihood (ML) estimation and multiple imputation (MI) being the two widely used principal methods. However, ML estimation is less flexible with item-level missing data, particularly when the goal is to compute and analyse scale scores. MI provides a flexible mechanism for dealing with item-level missingness, however, its implementation can be challenging1. Recent authors have proposed imputing missing item data at the level of the individual item rather than at the sub-scale or total score level2.Few studies have compared simple methods of imputation with MI for handling multi-item scales with contradicting results2 3. 

This PhD aims to provide clear guidance on how to impute missing item data for PROMS in RCTs. The PhD will include: (1) a literature review; (2) case studies to illustrate the sensitivity of trial results to the choice of method for imputing missing item data; (3) simulation studies to investigate the effects of varying different parameters such as prevalence of missing data, patterns of missing data, sample size, the number of items within the PROM, numbers of levels within the individual items, and planned analyses.

1.   Rombach I, Gray AM, Jenkinson C, Murray DW, Rivero-Arias O. Multiple imputation for patient reported outcome measures in randomised controlled trials: advantages and disadvantages of imputing at the item, subscale or composite score level. BMC Med Res Methodol. 2018;18(1):87-87.

2.   Eekhout I, de Vet HCW, Twisk JWR, Brand JPL, de Boer MR, Heymans MW. Missing data in a multi-item instrument were best handled by multiple imputation at the item score level. Journal of Clinical Epidemiology. 2014/03/01/ 2014;67(3):335-342.

3.   Bell ML, Fairclough DL, Fiero MH, Butow PN. Handling missing items in the Hospital Anxiety and Depression Scale (HADS): a simulation study. BMC Research Notes. 2016/10/22 2016;9(1):479.

2. Strategies to improve retention and external validity of clinical trials

It is well established that different strategies, such as offer of an incentive, can be used to improve retention within clinical trials. There are a number of studies within a trial (SWAT) that investigate methods for improving retention and maximising follow-up. 

However, an often overlooked consequence is the possibility that methods to improve retention may influence the characteristics of the followed-up population. This could, theoretically at least, have a positive or negative impact on the external generalisability of the trial. 

This project will investigate the impact of different retention strategies on the characteristics of the followed-up participants. The student will have access to existing clinical trial datasets hosted by Nottingham Clinical Trials Unit to allow empirical exploration of these issues, but the project may also involve data sharing, or primary data collection (quantitative and/or qualitative) depending on the interests of the student.

Brueton VC, Tierney JF, Stenning S, et al. Strategies to improve retention in randomised trials: a Cochrane systematic review and meta-analysis BMJ Open 2014;4:e003821. doi: 10.1136/bmjopen-2013-003821

Hardy P, Bell JL, Brocklehurst P; Epidural and Position Trial Collaborative Group. Evaluation of the effects of an offer of a monetary incentive on the rate of questionnaire return during follow-up of a clinical trial: a randomised study within a trial.BMC Med Res Methodol. 2016;16:82. Published 2016 Jul 15. doi:10.1186/s12874-016-0180-9

3. Methods for randomisation

In a randomised controlled trial participants are randomly assigned to one of a number of groups to test a specific drug, treatment or intervention. In practice, however, assignment is rarely completely random. It is conventional to adopt a method that promotes balance in terms of number and characteristics of study participants between randomised groups. 

This project will assess the performance of the most commonly used methods (e.g. minimisation and randomly permuted blocks), as well as less frequently used methods such as adaptive randomisation. Randomisation methods will be appraised in terms of (i) balance achieved (ii) predictability and (iii) ease of implementation. Recommendations for their use will be developed. 

When trying to achieve balance between randomised groups in a continuous variable it is typical to categorise this variable for the purposes of carrying out minimisation or stratification. However, there is no consensus on how this should be incorporated at the analysis stage. For instance, should the categorisation be preserved in the analysis model? 

Literature review for randomisation methods 

Development of a taxonomy of randomisation methods 

Simulation studies comparing performance of methods for adjusting for continuous stratification variables

Altman DG, Bland MG. Treatment allocation by minimisation BMJ 2005; 330:843 

Kahan, B.C., Rushton, H., Morris, T.P. et al. A comparison of methods to adjust for continuous covariates in the analysis of randomised trials. BMC Med Res Methodol 16, 42 (2016).

4. Exploring the potential for a ‘citizen science’ approach to delivery of eczema trials

Following a James Lind Alliance Prioritisation exercise for eczema, research priorities have been identified by patients and healthcare professionals that are priorities for future research.   However, with the average cost of a large pragmatic randomised controlled trial typically being over £1million per trial, there is a need to be innovative and efficient in designing and delivering clinical trials.

 This PhD project would inform the development of a larger programme of work to establish a novel ‘citizen-science approach’ to delivering clinical trials of non-drug interventions rapidly and efficiently.

Building on the considerable expertise of colleagues at the Centre for Evidence Based Dermatology  and Nottingham clinical Trials Unit in the design and conduct of eczema trails (www.nottingham.ac.uk/dermatology), the project would initially look at the feasibility of developing an on-line platform for the delivery of rapid trials that are designed and conducted by people with eczema. 

One of the initial methodological questions to be examined through this approach would be to explore the optimum timing and frequency of outcome assessments. This work would explore the impact of frequency of collection of patient-reported outcomes on treatment effects and missing outcome data. This work is vital to inform the development of future trials across the citizen-science platform.

The student would have the opportunity to develop further projects related to the wider programme, including development of novel methods to ensure broad engagement in the project from seldom heard groups (eg different ethnic groups and social classes).

Further details of the James Lind Alliance Priority Setting Partnership can be found here:

https://www.nottingham.ac.uk/research/groups/cebd/projects/eczema-psp/index.aspx 

5. Development of Core Outcome Sets in skin disease

This PhD will support the development of core outcomes sets in dermatology.

A core outcome set is an internationally agreed standardised set of outcomes that should be measured and reported, as a minimum, in all clinical trials in specific areas of health or health care (see COMET website).

The Centre of Evidence Based Dermatology has an international reputation for the development of core outcome sets through the HOME initiative and Cochrane Skin COUSIN. 

The successful candidate will progress outcomes research in an area of dermatology of interest to the applicant, but will be expected to fit within current CEBD strategic priority areas, or existing outcomes work projects.

The applicant may choose to develop and validate a new outcome measurement instrument (if required as part of a core outcome set), or to validate and test existing instruments according to the needs of the specific Core outcome Set Group of interest. There may also be opportunities to explore the use of item banks for capturing patient reported outcomes such as the PROMIS item bank.

Core Outcome Measures in Effectiveness Trials (COMET): http://www.comet-initiative.org/ 

COnsensus-based Standards for the selection of health Measurement Instruments (COSMIN): https://www.cosmin.nl/

Applications

The studentship provides a yearly stipend at current UKRI rates (£15,285 for 2020-21) plus Home/EU tuition fees. Full time (3 years) or part time study (up to 6 years, pro rata funding) is possible. Research costs to a maximum total of £10,000 are included.

Applicants should have a first or upper second class honours degree in a relevant discipline. A relevant Master's degree is desirable. Applications from Home/EU candidates https://www.nottingham.ac.uk/fees/tuitionfees/202021/index2020-21.aspx only due to funding restrictions. The studentship is available from October 2020.

To apply, applicants should send the following information to ctu@nottingham.ac.uk:

A personal statement (maximum 1000 words) demonstrating how your experience to date prepares you to undertake PhD level research

A current CV (two pages maximum)

Two academic references

Copies of your degree certificates and transcripts, and if appropriate, professional registration details

Evidence of English Language proficiency (IELTS 6.5 overall, with no less than 6.0 in any element, or equivalent) if your first language is not English

Key contacts

For enquiries about any of the PhD projects contact Professor Alan Montgomery, Professor of Medical Statistics and Clinical Trials, and NCTU Director (alan.montgomery@nottingham.ac.uk) or Professor Kim Thomas, Professor of Applied Dermatology Research and Co-Director of CEBD (kim.thomas@nottingham.ac.uk).

Closing Date: 24 May 2020
Category: Studentships