Randomized control trials (RCTs) are considered as the gold standard design for establishing causal inference.1 However, many interventions are not feasible or considered unethical, e.g. deliberately exposing a group of people to harmful work exposures. Therefore, observational studies are needed to estimate the effects of such exposures on health outcomes. Observational studies are, however, prone to selection bias and confounding.2 Pseudo-trials are considered as a stronger way to analyze observational data, providing an opportunity to control for confounding factors and mimic an RCT.1 Observational studies with repeated measurements of the same participants create a design that can be analyzed as a pseudo-trial. It is, therefore, possible to examine, whether a change in exposure results in a change in outcome of interest. This design thus better helps judge causal order, making sure that the change in exposure precedes the change in outcome. A proper use of pseudo-trials in public health research could be improving the opportunities to show causality.