Disenrollment under ACA: Lessons learned from a health care coverage initiative in California

This study examined factors associated with disenrollment, specifically the role of service utilization under the Health Care Coverage Initiative (HCCI) program. HCCI was implemented under a California §1115 Medicaid Waiver from September 2007 to August 2010. HCCI was later expanded statewide as the early implementation of the Affordable Care Act (ACA).

HCCI enrollees were low-income (up to 200% of the federal poverty level) adults ages 18-64, legally residing in ten participating counties, had lived more than 5 years in the U.S., and did not qualify for other public programs. The sample (N=57,000) included program enrollees (ages 18-62) who enrolled during the first year of the program in seven HCCI counties with comprehensive enrollment and utilization data.

Disenrollment was defined as: (1) dropping-out completely; or (2) “churning,” or enrollment interruptions. We applied a random-effect logistic regression model to evaluate the likelihood of disenrollment (both churn and drop-out) after controlling for individual-level characteristics (including age, gender, race, language, federal poverty level and chronic conditions) and service utilization (including annual rates of primary care provider (PCP) visits, emergency department (ED) visits or hospitalizations) in the first 12 months of the program year. We then compared the differences in rates of drop-out vs. churn to identify which populations are more likely to churn.

Half (50%) of enrollees in year 1 remained enrolled for all three program years. The odds of continuous enrollment in the second and third program years were 23% higher for enrollees who had a higher rate of PCP visits in year 1. The odds of disenrollment were higher if enrollees had higher rates of ED visits (42%) or hospitalizations (71%) in year 1. Also, the odds of continuous enrollment were higher for older and female enrollees as well as those with chronic conditions (asthma, diabetes, heart disease, hyperlipidemia, hypertension, osteoarthritis or a mental health condition). Those who dropped out and those who churned were similar in most characteristics and utilization patterns, except more individuals with at least three chronic conditions were in drop-out group.  

Our findings have important implications for retention efforts for Medicaid and exchanges under the Affordable Care Act (ACA). Retention efforts should include linking enrollees to primary care services. Retention efforts should also focus on younger, male, and healthier enrollees who are more likely to disenroll. Targeted retention policies and programs for newly covered individuals in Medicaid and exchanges will ultimately enhance the success of the ACA.