When fully implemented, the Patient Protection and Affordable Care Act (ACA) will increase the number of Americans with health insurance by over 30 million. Roughly half of the newly insured will gain coverage through Medicaid. Although such an increase in coverage alone will be a notable achievement, the true measure of success of the coverage expansion will depend on the extent to which it increases access to providers, increases appropriate and needed utilization of services, reduces unnecessary care, reduces program costs, and improves outcomes. The pursuit of these goals will take place in a very challenging budgetary context, necessitating an innovative approach to program design. The first two papers in this session will provide important new evidence on two key program design decisions: the choice to impose utilization caps on certain types of care, and how best to build a data infrastructure that facilitates the targeting of interventions to subgroups of enrollees most likely to benefit. The first paper in the panel uses a quasi-experimental design to assess the plausibly causal impacts of service caps on Medicaid beneficiaries with serious mental illness (SMI). Understanding how such caps impact utilization – for example, whether or not the exhaustion of outpatient mental health benefits leads to greater psychiatric emergency room visits – is of critical importance given the SMI population’s disproportionately high expenditure levels. The paper will provide the first evaluation of the impact of Medicaid psychotherapy limits on mental health care use among beneficiaries with SMI. In addition to the imposition of utilization caps, Medicaid programs also attempt to control costs by adopting a targeted approach to the provision of resource-intensive interventions. Such an approach requires a data infrastructure that facilitates the prospective stratification of patients into clinically meaningful subgroups. The second paper in the panel evaluates a recent effort in Wisconsin to gather the requisite data for such stratification efforts. In expanding Medicaid coverage to childless adults in 2009 via a waiver application, the State required that all applicants complete a self-reported health needs assessment (HNA) in addition to providing the sociodemographic information typically required for program enrollment. The paper will assess the predictive capacity of the HNA measures, providing the first exploration of the promise of using Medicaid enrollment systems to collect health-related information. As the cost-effectiveness paradigm continues to guide health care decision-making across payers – including Medicaid programs – it becomes increasingly important to provide careful, causal evidence regarding the benefits of Medicaid eligibility. While the effects of eligibility are most likely to be cumulative in nature, existing evidence largely focuses on contemporaneous health outcomes and/or short-term changes in outcomes. The third paper is one of very few in the literature to adopt a longer-term perspective, exploiting the historic discontinuity in Medicaid eligibility thresholds across young children and older children. It finds that Medicaid eligibility – when operationalized as access to coverage over the entire lifecourse of childhood – has effects along a very profound health margin: mortality of teenagers from internal causes.