Programmes of cluster-randomised trials for optimising learning healthcare systems

Background to the project

Healthcare systems consist of hierarchies of individuals and organisations. For example, in the UK, primary care is made up of patients within general practitioners (GPs), in general practices, within primary care networks (PCNs), in regions (Integrated Care Boards). Likewise, secondary care is made up of patients within clinicians, in teams, in hospitals, in trusts, in regions. Implementation science studies ways of promoting the systematic uptake of research findings, and other evidence-based practices, into routine practice [1]. One of its goals, therefore, is to optimise the care provided to patients through a learning healthcare systems approach. An example of an implementation strategy is clinical audit and feedback. Healthcare systems are audited and then feedback is provided on performance against a standard, potentially at multiple levels in the system. Typically, one cluster-randomised trial is conducted in which different types of feedback are randomised to one set of clusters (e.g. general practices or trusts) and then outcomes are assessed on patients. This only allows the effects of feedback at a single point in time to be evaluated. In routine practice, however, National Clinical Audits (NCAs) are conducted repeatedly over time, with a view to optimising learning in the system. Ivers and Grimshaw [2] argued for a shift in research design from a series of unrelated cluster-randomised trials to “implementation laboratories” that embed programmes of cluster-randomised trials in existing, large-scale initiatives, such as NCAs. Grimshaw et al [3] suggest a series of “head-to-head” parallel-group cluster-randomised trials, taking forward the most effective strategy from the previous trial and comparing it to a refined alternative. Other options are available, however. Where the same clusters (e.g. general practices or trusts) are randomised repeatedly over time, determining the type of feedback provided at each point, it is possible to obtain evidence on how healthcare systems learn over time. Moreover, it is possible to explore the effects of systematically changing the audit topic over time to obtain evidence on how learning generalises across topics. In the Design of Experiments (DoE) community, these programmes of cluster-randomised trials are referred to as “split-block” designs and are a specific type of “multi-stratum” design [4]. This studentship aims to explore statistical trial designs for optimising learning healthcare systems using ideas behind implementation laboratories as motivation.

What the studentship will encompass

The exact plan for the studentship will be determined by both the student’s skills/interests, but it is likely to include some, or all, of the following:

1    Review the methodological literature on approaches to designing and analysing programmes of cluster-randomised trials;

2    Re-design previous programmes of cluster-randomised trials across primary and secondary care settings (PILL and AFFINITIE) using these competing approaches;

3    Conduct a simulation study to compare the statistical efficiency of the competing approaches under ideal and realistic scenarios;

4    Consider the practical benefits and limitations of each competing approach for multiple stakeholders;

5    Explore pragmatic approaches that allow for the complexities faced in health and social care research.

 

Supervision

The proposed supervisory team includes a range of expertise: RW, an Associate Professor of Clinical Trial Methodology and NIHR Advanced Fellow, directs the Methodology Division at the Leeds Institute of Clinical Trials Research (LICTR) and leads the TMRP DoE Group within the Statistical Analysis Working Group; SA, an Associate Professor of Primary Care, General Practitioner and NIHR Advanced Fellow, leads a portfolio of trials optimising audit and feedback; BC, a Principal Statistician in the Complex Interventions Division in LICTR, leads the Implementation Science trials portfolio; SG, a Professor of Statistics in Department of Mathematics at King’s College London, is an international expert in multi-stratum experimental designs.

Planned fieldwork/secondments/placements

Shadowing an ongoing trial of audit and feedback in LICTR and a placement at King’s College London (up to one month) will be incorporated. Fieldwork and secondments are not applicable.

Detail of any PPI

PPIE will be an integral component of this studentship, which will be embedded in RW’s NIHR Advanced Fellowship and BC’s Trials Portfolio. Existing PPIE contributors will be approached and asked to convene a PPIE group for this project, meeting on an annual basis.

HOW TO APPLY

You are applying for a PhD studentship from the MRC TMRP DTP. A list of potential projects and the application form is available online at:

https://mrctmrpdtp.com/current-opportunities/

Please complete the form fully. Incomplete forms will not be considered. CVs will not be accepted for this scheme.

Please apply giving details for your first choice project. You can provide details of up to two other TMRP DTP projects you may be interested in at section B of the application form.

Before making an application, applicants should contact the project supervisor to find out more about the project and to discuss their interests in the research before 06 January 2025.

The deadline for applications is 12 noon (GMT) 13 January 2025. Late applications will not be considered.

Completed application forms must be returned to: enquiries@methodologyhubs.mrc.ac.uk

Informal enquiries may be made to Dr Walwyn – R.E.A.Walwyn@leeds.ac.uk