Multivariate outcomes in trials evaluating complex interventions

Background

MRC guidance flags the number of outcomes as one of the key dimensions of complexity when evaluating complex interventions, stating that “a single primary outcome and a small number of secondary outcomes are the most straight-forward for statistical analysis but may not represent the best use of the data or provide an adequate assessment of the success or otherwise of an intervention that has effects across a range of domains”. Often what is of interest to clinicians is the profile of treatment effects across outcomes. Outcomes may provide an assessment of success across a range of domains relating to the patient (e.g. symptoms, disability, adherence), regarding different targets (e.g. patient, carer, provider, teacher), or in a single domain informed my multiple informants (e.g. Quality of Life). The latter is common in trials of vulnerable populations (e.g. children or adults with dementia). Here different perspectives provide data on the range of views people who know the patient well might have. Outcomes may be only collected from informants other than the patient (‘proxies’), with potential bias arising, or they may be obtained by researchers through interviews, possibly introducing rater effects. To facilitate uptake of more sophisticated approaches to the analysis of complex intervention trials, guidance is needed on the pros and cons of different approaches, illustrated using a range of examples.

Objectives

The candidate will (i) review relevant methodological and clinical literature, (ii) explore and re-analyse trial datasets in one or more of the following areas: care homes, child and adolescent mental health, palliative care, stroke, elderly care; (iii) consider different approaches to presenting results to various audiences, (iv) explore methods for handling multiplicity, mixed outcome types, missing item data, detection bias and sample size across scenarios, providing recommendations and guidance for the primary analyses of future trials, (v) consider novel design options, and (vi) reflect on practical considerations.

Summary

Complex healthcare interventions are widespread within the NHS. One of the key dimensions of complexity when evaluating complex interventions in clinical trials is the number of outcomes assessed. Outcomes may relate to different patient domains (e.g. symptoms, disability, adherence), to different targets (e.g. patient, carer, provider, teacher), to a single domain but multiple informants (e.g. Quality of Life) common in trials of vulnerable populations (e.g. children or adults with dementia). They may be collected from informants other than the patient (‘proxies’) or by researchers through interviews. A general framework is needed that facilitates the uptake of more sophisticated statistical methods, making more use of existing data. The candidate will review the methodological and clinical literatures, re-analyse existing datasets, consider issues that arise and explore practical and analysis considerations. This project sits within a larger programme of work on developing methods for evaluating complex healthcare interventions.

Eligibility

You should hold a first degree equivalent to at least a UK upper second class honours degree in a relevant subject (mathematics, statistics, psychology). Ideally candidates should have, or be about to complete, an MSc in a relevant subject.

Candidates whose first language is not English must provide evidence that their English language is sufficient to meet the specific demands of their study, the Faculty minimum requirements are:

• British Council IELTS - score of 6.5 overall, with no element less than 6.0
• TOEFL iBT - overall score of 92 with the listening and reading element no less than 21, writing element no less than 22 and the speaking element no less than 23.

How to apply

To apply for this scholarship applicants should complete a Faculty Scholarship Application form using the link below http://medhealth.leeds.ac.uk/download/2376/fmh_scholarship_application_16_17_entry and send this alongside a full academic CV, degree transcripts (or marks so far if still studying) and degree certificates to the Faculty Graduate School [Email Address Removed]

We also require 2 academic references to support your application. Please ask your referees to send these references on your behalf, directly to [Email Address Removed] by no later than Tuesday 6 December 2016.

Any queries regarding the application process should be directed to [Email Address Removed]

For further information about projects contact Dr Rebecca Walwyn by emailing [Email Address Removed] and include your brief CV.