Capping Consumption in the Medicaid Market

In an effort to reduce frequent emergency department (ED) use among fee-for-service (FFS) Medicaid recipients, the state of Florida enacted a policy in 2012 limiting payment for ED use to 6 visits annually and no more than 2 visits per month for non-pregnant individuals age 21 and over. Medicaid recipients who anticipated facing a 100% coinsurance rate at some point in the calendar year may have changed health care consumption by reducing their overall ED utilization, avoiding ED use early in the year, or changing the intensity and types of services obtained within an ED encounter. While this unusual and blunt policy instrument attracted significant popular media and federal regulator attention, it is unknown what, if any response, it actually elicited from Medicaid beneficiaries.

Our analysis utilizes the universe of ED visits from the state of Florida for 2009 through 2013. These detailed records include a rich set of variables, including diagnosis and procedure codes, type of insurance, demographic information, and post-discharge treatment plans. We supplement these records with county-level data on Medicaid enrollment, including managed care penetration, prior to the policy change.

We use a difference-in-differences approach to estimate the effect of the policy change on the behavior of FFS Medicaid patients relative to patient-payer groups not targeted by the legislation. We further build on this approach by using a triple difference model that exploits the variation in FFS penetration across Florida counties to explore heterogeneity consistent with a dose-response relationship.

Early findings indicate that aggregate ED visit volume decreased for adult (non-pregnant) patients enrolled in a Medicaid FFS plan following the policy change; however, the declines are largely confined to areas predominantly relying on FFS Medicaid plans. Interestingly, there is a larger and sharper ED utilization divergence between FFS and Medicaid managed care (MMC) children under age 18, though FFS enrollees under age 21 were exempt from the cap. However, our detailed insurance enrollment analyses reveal that more Medicaid beneficiaries opt into MMC plans following the state legislature’s policy announcement. In this way, the observed changes in FFS utilization behavior are largely driven by a strategic shift across Medicaid plan types, as opposed to restrained use of ED care. The findings are consistent with existing and new Medicaid enrollees being aware of and concerned by the policy, though with incomplete understanding of policy exemption rules. Ongoing analyses are looking at the health problem mix and treatment flows to Medicaid FFS as well as MMC patients after policy implementation when compared against unaffected payer groups.