Background to the project
Missing data are a common issue in randomised controlled trials (RCTs) to evaluate the relative effects of interventions. For example, participants may withdraw or be lost to follow-up, questionnaires may be unreturned, and responses to individual questionnaire items may be illegible non-existent. Handling missing data inappropriately can seriously undermine the validity of trial-based cost-effectiveness analyses (CEA) because costs or health outcomes in individuals with missing data may be systematically different from those with fully observed information. Both costs and quality-adjusted life-years (QALYs) are cumulative measures, meaning that missing data for one item or at one timepoint results in patient-level missingness. In addition, they are non-normally distributed and tend to be correlated which means that guidelines developed for statistical analysis of other types of trial data might not be transferrable to economic analysis. If the wrong assumption is made, the obtained treatment effect and its standard error will be biased, resulting in misleading inferences. Therefore, it is necessary to assess the impact of untestable and unavoidable assumptions about any unobserved data on key inferences. Missing data are often addressed under a ‘missing at random’ (MAR) assumption. However, this assumption is often questionable and guidelines1 recommend conducting sensitivity analyses to assess the robustness of conclusions to plausible departures from MAR. Recent reviews have highlighted that around 1/3 of trials with missing data report sensitivity analyses indicating a large gap between methodological developments and practical application2,3.
The aim of this studentship is to develop a practical framework on handling missing data in trial-based CEA focusing on multiple imputation (MI), delta-based and reference-based approaches, with software. This will be complementary to existing best practice guidance for the conduct and reporting of health economic evaluations. Objectives are to 1) implement quantitative methods to handle missing data in year 1, 2) understand the mechanisms and reasons for missing data using qualitative methods in year 2, and 3) integrate data to inform the framework using mixed methods approach in year 3. The results of the project will be complementary to existing best practice guidance for the conducting and reporting the health economic evaluations. This studentship will also provide facilities, resources and training needed to undertake this project.
A multidisciplinary supervisory team has been established. Primary supervisor, Dr Ge Yu, senior health economist with expertise in applied microeconometrics, is currently holding an NIHR senior fellowship and NIHR ARC fellowship. Ge is a member of both MRC-NIHR Trials Methodology Research Partnership (TMRP) Health Economics and Statistical Analysis Working Groups (WGs). Co-supervisors: Dr Yu Fu, senior research associate with expertise in mixed methods methodology, service evaluation and implementation, is well linked to the NHS Beneficial Changes Network and regional Academic Health Science Network. Professor Luke Vale (HEG Director) will provide methodological advice on study design and conduct. Supervision meeting will take place weekly during the first three months and monthly thereafter. The student will be introduced to both TMRP WGs to collaborate and disseminate learning captured throughout the project and contribute to the growing power of the methodological solution to improve trial-based economic evaluation.
As case studies, the student will use data on two NIHR funded trials where the HEG has full access to the data and original analyses. The recently completed PHOTO trial in bladder cancer surgery will provide a case study illustrating the application of the guidance. The OPEN trial in management of urethral strictures will provide another case that has challenges relating to the complexity of its data collection utilities (data were collected for acute events as well as at same time point).
Patients who contributed to the design of the PHOTO trial will be invited as patient representatives to inform the methods development. Meetings will take place quarterly to enable a robust co-production including reviewing context/ethics and planning/delivering dissemination activities
A Masters in a quantitative subject with a statistics component with knowledge of programming languages such as Stata, R or Python is required. Prior knowledge of clinical trials and qualitative research is not required but would be advantageous.
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:
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 primary supervisor to find out more about the project and to discuss their interests in the research.
The deadline for applications is 4pm (GMT) 18 February 2022. Late applications will not be considered.
Completed application forms must be returned to: [Email Address Removed]
Informal enquiries may be made to [Email Address Removed]