Amrita G Shenoy1, Charles E Begley2, Lee Revere3, Stephen H Linder4, Stephen P Daiger5. 1. Math, Science and Technology Department at the University of Minnesota Crookston, United States. Electronic address: amritashenoy@gmail.com. 2. Center for Health Services Research, Department of Management, Policy and Community Health, UTHealth School of Public Health, United States. 3. Department of Management, Policy and Community Health, UTHealth School of Public Health, United States; Fleming Center for Healthcare Management, UTHealth School of Public Health, United States. 4. Institute for Health Policy, Department of Management, Policy and Community Health, UTHealth School of Public Health, United States; Health Policy Institute, Texas Medical Center, United States. 5. Human Genetics Center, UTHealth School of Public Health, United States; Ruiz Dept. of Ophthalmology and Visual Science, Univ. of Texas Health Science Center - Houston, United States.
Abstract
INTRODUCTION: Adoption of Medicaid Section 1115 waiver is one of the many ways of innovating healthcare delivery system. The Delivery System Reform Incentive Payment (DSRIP) pool, one of the two funding pools of the waiver has four categories viz. infrastructure development, program innovation and redesign, quality improvement reporting and lastly, bringing about population health improvement. BACKGROUND: A metric of the fourth category, preventable hospitalization (PH) rate was analyzed in the context of eight conditions for two time periods, pre-reporting years (2010-2012) and post-reporting years (2013-2015) for two hospital cohorts, DSRIP participating and non-participating hospitals. The study explains how DSRIP impacted Preventable Hospitalization (PH) rates of eight conditions for both hospital cohorts within two time periods. METHODS: Eight PH rates were regressed as the dependent variable with time, intervention and post-DSRIP Intervention as independent variables. PH rates of eight conditions were then consolidated into one rate for regressing with the above independent variables to evaluate overall impact of DSRIP. An interrupted time series regression was performed after accounting for auto-correlation, stationarity and seasonality in the dataset. RESULTS: In the individual regression model, PH rates showed statistically significant coefficients for seven out of eight conditions in DSRIP participating hospitals. In the combined regression model, the coefficient of the PH rate showed a statistically significant decrease with negative p-values for regression coefficients in DSRIP participating hospitals compared to positive/increased p-values for regression coefficients in DSRIP non-participating hospitals. CONCLUSION AND IMPLICATIONS: Several macro- and micro-level factors may have likely contributed DSRIP hospitals outperforming DSRIP non-participating hospitals. Healthcare organization/provider collaboration, support from healthcare professionals, DSRIP's design, state reimbursement and coordination in care delivery methods may have led to likely success of DSRIP. LEVEL OF EVIDENCE: IV, a retrospective cohort study based on longitudinal data.
INTRODUCTION: Adoption of Medicaid Section 1115 waiver is one of the many ways of innovating healthcare delivery system. The Delivery System Reform Incentive Payment (DSRIP) pool, one of the two funding pools of the waiver has four categories viz. infrastructure development, program innovation and redesign, quality improvement reporting and lastly, bringing about population health improvement. BACKGROUND: A metric of the fourth category, preventable hospitalization (PH) rate was analyzed in the context of eight conditions for two time periods, pre-reporting years (2010-2012) and post-reporting years (2013-2015) for two hospital cohorts, DSRIP participating and non-participating hospitals. The study explains how DSRIP impacted Preventable Hospitalization (PH) rates of eight conditions for both hospital cohorts within two time periods. METHODS: Eight PH rates were regressed as the dependent variable with time, intervention and post-DSRIP Intervention as independent variables. PH rates of eight conditions were then consolidated into one rate for regressing with the above independent variables to evaluate overall impact of DSRIP. An interrupted time series regression was performed after accounting for auto-correlation, stationarity and seasonality in the dataset. RESULTS: In the individual regression model, PH rates showed statistically significant coefficients for seven out of eight conditions in DSRIP participating hospitals. In the combined regression model, the coefficient of the PH rate showed a statistically significant decrease with negative p-values for regression coefficients in DSRIP participating hospitals compared to positive/increased p-values for regression coefficients in DSRIP non-participating hospitals. CONCLUSION AND IMPLICATIONS: Several macro- and micro-level factors may have likely contributed DSRIP hospitals outperforming DSRIP non-participating hospitals. Healthcare organization/provider collaboration, support from healthcare professionals, DSRIP's design, state reimbursement and coordination in care delivery methods may have led to likely success of DSRIP. LEVEL OF EVIDENCE: IV, a retrospective cohort study based on longitudinal data.
Keywords:
DSRIP Policy; Impact of DSRIP on Preventable Hospitalization Rates; Interrupted Time Series Regression Analysis; Macro- and Micro- care delivery environmental factors on DSRIP policy; Preventable Hospitalization Rate Change: DSRIP and PH rates