Birkbeck, University of London Featured PhD Programmes
Norwich Research Park Featured PhD Programmes
Norwich Research Park Featured PhD Programmes
Norwich Research Park Featured PhD Programmes
European Molecular Biology Laboratory (Heidelberg) Featured PhD Programmes

Improving statistical methods for trial emulations using observational data

  • Full or part time
    Dr L Su
    Dr S Seaman
  • Application Deadline
    Tuesday, January 07, 2020
  • Competition Funded PhD Project (Students Worldwide)
    Competition Funded PhD Project (Students Worldwide)

Project Description

Randomised controlled trials (RCTs) are regarded as the gold standard for estimating causal effects of treatments on health outcomes. By randomising patients to receive treatment or non-treatment, these trials ensure that the treated group of patients and the untreated group of patients are comparable. However, RCTs are not always feasible, because of time, budget or ethical constraints, or have not been used to study particular populations, e.g. RCTs often exclude patients with multiple diseases.

Observational data such as electronic health records (EHR) offer an alternative way to estimate causal effects of treatments. However, estimating the causal effect of a treatment from such databases is not straightforward. For example, the individuals in the EHR databases are usually not randomly assigned to the treatments they received, meaning that any outcome differences in these groups may be fully or partly explained by differences between the individuals in each group, rather than differential effects of treatments (i.e., the treatment effects are confounded). Recently, the ‘emulating a target trial’ framework was proposed by Hernán and Robins (1) as a formal structure for estimating treatment effects from large observational data sets. This involves specifying the protocol of a ‘target trial’ that one would like to carry out, identifying the individuals in the observational database who satisfy the eligibility criteria of this trial, and then comparing the outcomes between treated and untreated eligible individuals during the trial follow-up period. The lack of randomisation in treatment assignment can be at least partly accounted for by using techniques such as matching, weighting or regression adjustment. Other sources of bias that may need to be addressed include loss to follow-up, irregular measurement times and treatment switching.

The `emulating a target trial’ approach has received increasing attention. In addition to providing evidence about the benefits and risks of treatments, the results from trial emulations can be used to design future RCTs of promising treatments, e.g. trials of treatment combinations for tackling multiple diseases in multi-morbid patients.

In this PhD project, we will investigate
(1) how to improve statistical methods in order to address various biases in trial emulations, and
(2) how to conduct sensitivity analyses when unverifiable assumptions made in trial emulations are violated.

This research will build on the experience of Drs Li Su and Shaun Seaman in methods for addressing loss to follow-up and confounding, including time-dependent confounding, in complex observational data (2,3). The successful candidate will have opportunities to collaborate with Drs Sofía Villar and Steve Kiddle, who will provide guidance in terms of clinical trial designs and EHR data, respectively.

Funding Notes

The MRC Biostatistics Unit offers 4 fulltime PhDs funded by the Medical Research Council for commencement in April 2020 (UK applicants only) or October 2020 (all applicants). Academic and Residence eligibility criteria apply.

In order to be formally considered all applicants must complete a University of Cambridge application form. Informal enquiries are welcome to

Applications received via the University application system will all be considered as a gathered field after the closing date 7th January 2020

For all queries see our website for details View Website

References

1. Hernán and Robins (2016) https://doi.org/10.1093/aje/kwv254
2. Yiu and Su (2018) https://doi.org/10.1093/biomet/asy015
3. Seaman and Vansteelandt (2018) https://europepmc.org/abstract/med/29731541

Related Subjects

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully





FindAPhD. Copyright 2005-2019
All rights reserved.