Oregon Experiment with Medicaid Accountable Care and Pediatric Preventive Care Utilization Among Infants

Abstract

Research ObjectiveIn August 2012, Oregon began enrolling Medicaid beneficiaries in coordinated care organizations (CCOs), a unique mandatory‐enrollment accountable care organization (ACO) model with payment methods strongly tied to preventive care and care coordination through patient‐centered medical homes. No study has examined the impact of CCOs on preventive care utilization among infants. The main goal of this study was to examine the impact of this new delivery model on the utilization of preventive care services among infants enrolled in Medicaid during 2 years after birth. We also investigated heterogeneous impacts of the CCO model for preterm and full‐term infants as well as over time.Study DesignWe linked Oregon 2008‐14 birth certificates to Medicaid eligibility and claims data, and created the pre‐ and post‐CCO cohorts of Medicaid infants. Medicaid infants were defined as being enrolled in Medicaid for at least 80% of the two years after birth. Pediatric preventive care outcomes included indicators of preventive care adequacy as recommended by American Academy of Pediatrics: six well‐child visits by the first birthday, nine well‐child visits by the second birthday, and annual developmental screening. We estimated logistic regression models and linear probability models, adjusting for time trends of outcomes and a comprehensive set of maternal and infant characteristics. We utilized a difference‐in‐differences approach to examine heterogeneous effects of CCOs for preterm and full‐term infants. Spline regression was used to test the effect of CCOs on the outcomes for every six months from the start of the CCO implementation. We conducted a range of sensitivity analyses, such as different thresholds of Medicaid enrollment for the definition of Medicaid infants and excluding infants of mothers in the Medicaid expansion group.Population Studied36 546 pre‐CCO infants born between August 2008 and July 2010 and 47 973 post‐CCO infants born between August 2012 and December 2014Principal FindingsDuring the 2 years after birth, infants enrolled in Medicaid CCOs had higher probability of well‐child visit adequacy: 10% points (95%CI: 8.2; 11.8) by the first birthday and 5.7% point (95%CI: 4.3; 7.1) by the second birthday. Annual probability of developmental screening increased by 23.1% points (95%CI: 21.6; 24.6) after the CCO implementation. No difference in the CCOs’ impact was found between preterm infants and full‐term infants. Increase in preventive care was found to start in the second year of the CCO implementation.ConclusionsThe CCO model was associated with an increase in preventive care utilization among infants enrolled in Medicaid during the first 2 years of birth.Implications for Policy or PracticeIncrease in preventive services among low‐income infants after CCO implementation implies that the Oregon Medicaid ACO model could improve quality of care for infants in their early years. A growing number of states have been adopting accountable care models for their Medicaid enrollees. Our findings suggest that other states benefit from studying the Oregon approach to accountable care for Medicaid enrollees that integrates financing and delivery of care, and implements a payment methodology with clear performance targets for pediatric preventive services.Primary Funding SourceCenters for Disease Control and Prevention.