Background
England’s cancer registry data in theory offer a wealth of data for comparing the effects of cancer treatments[1, 2]. Linking England’s cancer registry data with other healthcare datasets allows researchers to estimate how well treatments perform in real-world settings, benefiting patients, clinicians, and policymakers. However, drawing reliable conclusions from observational data can be challenging as biases inherent to observational studies and data
quality issues can lead to misleading results [2, 3].
Miguel Hernan and James Robins (2016) proposed the Target Trial (TT) framework as a more robust approach to estimate comparative treatment effectiveness using observational data
[4]. This framework leverages counterfactual theories and principles from randomised controlled trials (RCTs), considered the gold standard for evaluating treatments[5]. However, the reliability of the estimates obtained from applying the TT framework to analyse England's cancer registry data remains unknown.
Addressing the Uncertainty: A Benchmarking Approach
This PhD project will address this uncertainty through a benchmarking process. Benchmarking involves comparing estimates derived from observational analyses against established gold standard evidence from RCTs [5]. By designing and emulating three TTs leveraged on three carefully selected existing RCTs (LUX-Lung 7 [6], KEYNOTE-024 [7], and TNT [8]) using England’s cancer registry data within the TT framework, this project will assess the reliability of treatment effectiveness estimates using this approach.
Expected Impact
This study has the potential to provide valuable insights into the reliability of comparative effectiveness estimates derived from England's cancer registry data. These findings will inform clinicians, policymakers, and patients on the appropriate use and interpretation of results from future comparative effectiveness studies utilising this data source.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)