Can Synthetic Controls Improve Causal Inference in Interrupted Time Series Evaluations of...

History bias—confounding by unexpected events occurring at the same time of the intervention—threatens the validity and limits the causal inference of interrupted time series designs. In a new study led by PISC International Scholar, Michelle Degli Esposti, PhD, along with senior author and PISC External Advisory Board Member, David Humphreys, PhD, Dr. Degli Esposti, et al. evaluate if and when synthetic controls can strengthen an interrupted time series design. In this study they (1) summarize the main observational study designs used in evaluative research, (2) outline when the use of synthetic controls can strengthen interrupted time series studies and when their combined use may be problematic, (3) provide recommendations for using synthetic controls in interrupted time series and illustrate the potential pitfalls of using a data-driven approach, and (4) emphasize the importance of theoretical approaches for informing study design and argue that synthetic control methods are not always well suited for generating a counterfactual that minimizes critical threats to interrupted time series studies. Incorporating synthetic controls in interrupted time series studies may not always nullify important threats to validity nor improve causal inference, but advances in synthetic control methods bring new opportunities to conduct rigorous research in evaluating public health interventions.

Go to Article

Back to News