Article Text

Download PDFPDF

107 Can synthetic controls improve causal inference in interrupted time series evaluations of public health interventions?
  1. Michelle Degli Esposti1,
  2. Thees Spreckelsen1,
  3. Antonio Gasparrini2,
  4. Douglas J Wiebe3,
  5. Alexa R Yakubovich4,
  6. David K Humphreys1
  1. 1Department of Social Policy and Intervention, University of Oxford
  2. 2Department of Public Health, Environments and Society, London
  3. 3Department of Biostatistics and Epidemiology, University of Pennsylvania
  4. 4Centre for Urban Health Solutions, Li Ka Shing Knowledge Institute, St Michael’s Hospital, Toronto, Canada


Statement of Purpose Interrupted time series (ITS) designs are a valuable quasi-experimental approach for evaluating public health interventions. ITS extends a single group pre-post comparison by using multiple timepoints to control for underlying trends. But history bias – confounding by unexpected events occurring at the same time of the intervention – threatens the validity of this design and limits causal inference. Synthetic control methodology (SCM), a popular data-driven technique for deriving a control series from a pool of unexposed populations, is increasingly recommended. We aimed to evaluate if and when SCM can strengthen an ITS design.

Methods/Approach First, we summarise the main observational study designs used in evaluative research, highlighting their respective uses, strengths, biases, and design extensions. Second, we outline when the use of SCM can strengthen ITS studies and when their combined use may be problematic. Third, we provide recommendations for using SCM in ITS and, using a real-world example of an evaluation of Florida’s Stand Your Ground laws on homicides, we illustrate the potential pitfalls of using a data-driven approach to identify a suitable control series.

Results Our real-world evaluation demonstrates that the benefits of SCM in ITS depends on the nature of the time-varying confounding which presents the most plausible threat to the study’s validity. We emphasise the importance of theoretical approaches for informing study design and argue that synthetic control methods are not always well-suited for minimising critical threats to ITS studies.

Conclusions Advances in SCM bring new opportunities to conduct rigorous research in evaluating public health interventions. However, incorporating synthetic controls in ITS studies may not always nullify important threats to validity nor improve causal inference.

Significance and Contributions to Injury and Violence Prevention Science We provide important methodological recommendations to guide advancement in the science of injury and violence prevention.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.