Evaluating the New Jersey Hospital DSRIP Program: Lessons for policy reforms structuring safety net delivery system transformation and financing

Research Objective and Background: The Delivery System Reform Incentive Payment (DSRIP) program currently underway in thirteen states is an increasingly utilized payment innovation reflecting the federal government’s focus on pay-for-performance (P4P) arrangements to drive delivery system reform. However, outside of state reports, there is limited evidence around the effectiveness of the decade-long DSRIP experience. We present findings from the evaluation of the New Jersey DSRIP program to examine the extent to which the program helped hospitals improve patient health outcomes and promote population health. Like many other states, the program transitioned Medicaid supplementary subsidies to a P4P system and our study also sheds light on the impact of such payment transformation on health care utilization and outcomes.

Data: Comprehensive Medicaid claims over 2011-2017 were analyzed to examine health outcomes that were the focus of hospital DSRIP programs and, separately, ambulatory care quality and spending reflecting overall population health. A hospital web survey and key informant interviews captured stakeholder perceptions on program potential and impact.

Methods: Difference-in-differences estimation was utilized to identify the impact of the DSRIP policy on various outcomes, relating specifically to hospital-adopted programs or overall quality and efficiency of care. Analyzing claims, we assessed rates of ambulatory care sensitive (ACS) hospitalizations; readmissions after initial hospitalizations for heart failure, AMI and pneumonia; follow-up rates after mental health hospitalizations; and treatment for alcohol and other drug dependencies. For hospital-based outcomes, patients were assigned to the intervention group if the hospital participated in the program. For area-based ambulatory or primary care quality outcomes, patient program exposure depended on DSRIP participation among the relevant hospitals in a zip code.

Results: Since inception, hospital stakeholders were enthusiastic about the chronic disease management interventions and perceived positive impacts on enrolled patients. Claims-based analysis revealed lower likelihood of asthma-related ED visits and hospitalizations for patients undergoing asthma management. Rates of avoidable ED visits and associated costs increased from the baseline to the end of the fifth DSRIP demonstration year. We did not detect a statistically significant impact on hospital readmissions, follow up after discharge, and rates of ACS hospitalizations. Hospital-reported metrics measuring recommended and preventive care showed improvement during the DSRIP implementation years. There was no negative impact of DSRIP on overall hospital margins.

Conclusions: The New Jersey DSRIP program is particularly unique and innovative since unlike several other states, it adopted a participatory model that opened eligibility to all hospitals. Our findings allay concerns that such a funding allocation and the P4P shift could adversely impact outcomes, and is an important case-study for structuring payment reform initiatives in the safety-net space. As the DSRIP program structure continues to evolve as a value-based P4P arrangement, our findings will provide valuable guidance for current and future policies leveraging payment reforms to implement system-wide delivery system transformation.