Comment on Frimpong et al.: Impact of the 1115 Behavioral Health Medicaid Waiver on Adult Medicaid Beneficiaries in New York State

This comment provides a critical assessment of Frimpong et al. "Impact of the 1115 Behavioral Health Medicaid Waiver on Adult Medicaid Beneficiaries in New York State," published in Health Services Research (2021) [DOI: 10.1111/1475-6773.13657]. The authors reported a decline in inpatient utilization due to the Waiver. However, they failed to disclose that inpatient utilization did not decline when using an alternative difference-in-differences regression specification. Therefore, the decline in inpatient utilization appears to be an artifact of the authors' regression specification. I provide an explanation of these discrepant findings, drawing from recently published methodological studies on differences-in-differences and matching on outcomes.