Selection of Out-of-State Comparison Groups and the Synthetic Control Method
U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services, Center for Medicaid and CHIP Services
Evaluators can use out-of-state comparison groups when no in-state comparison group is available or to corroborate findings from other evaluation strategies. This white paper provides practical suggestions for selecting or constructing out-of-state comparison groups to support the use of this strategy. When choosing out-of-state comparison groups, states and their evaluators should select other states with observed characteristics and pre-intervention trends in the outcomes of interest that are similar to those of the demonstration state. This paper describes several types of characteristics to consider. If an out-of-state comparison group is difficult to find because no other states are similar to the demonstration state, the synthetic control method can help with constructing a comparison group. This method has been used in published research on several Medicaid-related outcomes.