Impact of Integrated Care Delivery on Emergency Department Use Among Female Medicaid Enrollees: A Natural Experiment in Oregon's Coordinated Care Organizations

Objectives: Oregon’s coordinated care organizations (CCOs) transform the way healthcare is delivered for Medicaid enrollees, by integrating the provision of medical, behavioral, and oral health services.  This study examines whether implementation of CCOs has affected emergency department (ED) use among women of reproductive age (15-44 years old).  Reducing avoidable ED use among this large population of adult Medicaid enrollees was a goal of the CCO reform. 

Methods: Using enrollment and claims records from January 2011 to July 2013, we created person-month panel data on 665,650 female Oregon Medicaid enrollees (N = 7,985,846).  Utilizing the fact that some CCOs began operations earlier than others, we perform a quasi-experimental analysis that compares the difference in ED visits between pre- and post-CCO implementation periods for CCO enrollees compared to the difference for non-CCO Medicaid enrollees.  Two-part difference-in-differences count data models are estimated. The models control for individual heterogeneity.

Results: The monthly probability of ED use among Oregon women of reproductive age enrolled in Medicaid decreased on average by 0.14 percentage points following the implementation of CCOs (p < .01)–an approximately 1 percent reduction in the probability of ED use before CCO implementation.  The magnitude of the relationship varies by CCO, and is larger for nonwhites.  On average the frequency of ED use decreased after CCOs began operating although this relationship is not statistically significant.

Conclusion: During the first year of implementation, Oregon CCOs led to a reduction in ED use. Future evaluations can benefit from understanding factors contributing to variation across CCOs.