This BMJ article presents a comprehensive update to the UK Medical Research Council (MRC) and National Institute for Health and Care Research (NIHR) guidance on designing, conducting, and evaluating natural experiments in public health and health systems research. Natural experiments, where exposure to an intervention or policy is not controlled by the researcher, are essential for studying the population-level effects of real-world changes that cannot be ethically or practically randomized. The paper provides a refined definition of natural experiments to clarify inclusion criteria and distinguish them from other observational studies. Besides, it provides a structured assessment of strengths, limitations, and potential biases across different natural experimental designs, offering detailed guidance on study planning, implementation, and interpretation. In addition, it lists a set of practical recommendations to improve the credibility, transparency, and reporting standards of natural experiments, supporting more robust policy-relevant evidence. Authors emphasize the need for multidisciplinary collaboration and the consideration of equity and unintended consequences when evaluating interventions.
Recommended on behalf of EASE by Silvia Maina, Italy