Synthetic Control Methods for the Evaluation of Single-Unit Interventions in Epidemiology: A Tutorial

Area of Science:

• Public Health
• Health Policy Evaluation
• Epidemiology

Background:

• Evaluating population-level interventions, such as legislative changes, is challenging due to the impracticality of randomized controlled trials.
• Quasi-experimental designs often rely on finding suitable control groups that match the treated unit's pre-intervention characteristics and trends, which can be difficult.
• The limited number of available control regions further complicates the selection of appropriate comparisons for aggregate-level interventions.

Purpose of the Study:

• To provide a nontechnical tutorial on the synthetic control method for health researchers.
• To address the gap in accessible literature on the synthetic control method within health research.
• To illustrate the application and considerations of the synthetic control method using a real-world example.

Main Methods:

• The synthetic control method utilizes a data-driven algorithm to construct an optimal, weighted control unit from available control units.
• This synthetic control unit is based on pre-intervention data, aiming to closely replicate the treated unit's characteristics and outcome trends.
• A worked example evaluating Florida's 'stand your ground' law is used to demonstrate the methodological and practical aspects of the approach.

Main Results:

• The synthetic control method provides a data-driven solution for creating a suitable counterfactual in the absence of randomized experiments.
• The worked example demonstrates how the method can be applied to evaluate the impact of a specific state-level policy.
• The tutorial highlights key methodological and practical considerations for implementing this technique in health research.

Conclusions:

• The synthetic control method is a valuable tool for health researchers to evaluate population-level interventions when traditional experimental designs are not feasible.
• This method enhances the validity of quasi-experimental studies by creating a data-driven, optimal synthetic control group.
• Increased accessibility and application of the synthetic control method can improve the evaluation of health policies and interventions.