RCTs
Randomised controlled trials (RCTs) are the gold-standard for CI,
and although they provide valuable evidence for predicting
treatment efficacy, they are costly and time-consuming.
How to use Model Estimated Controls
MECs and other CI tools
Other CI tools have been developed to estimate causal effects from observational datasets.
Other CI tools
G-estimation and Targeted Maximum Likelihood Estimations (TMLEs) are also CI tools that have been developed. These tools aim to estimate an average treatment effect and require patient level observational datasets from both the control group and the experimental treatment group. G-estimation and TMLE compare the expected outcome under each treatment group and compare these expectations to derive an average treatment effect.
How do MECs compare to other CI tools?
MECs can act as a control group, and so an observational dataset for a control treatment group is not required. MECs compare the observed response from an experimental group against the predicted response from the control. Other tools such as G-computation typically compare the expected response from each treatment group