Target trial emulation: applying principles of randomised trials to observational studies

The randomised trial is the preferred study design for evaluating the effectiveness and safety of interventions. Yet such trials can be prohibitively expensive, unethical, or take too long. When it is not possible to carry out a randomised trial, observational data can be used to answer similar questions. Here, we describe the process of using observational data to emulate a target trial, which applies the study design principles of randomised trials to observational studies that aim to estimate the causal effect of an intervention. The target trial provides a formal framework to help avoid self-inflicted biases common to observational studies.