Changes in insurance over time -"churning" -may lead to adverse consequences, but there has been little evidence to date on churning since the implementation of the Affordable Care Act (ACA). We explored the frequency and implications of churning with a survey of over 3000 low-income adults in three states with different ACA policies: Arkansas, Kentucky, and Texas. We also compared 2015 churning rates in these states to pre-ACA survey data from 2013. Overall, 20-25% of respondents experienced a change in coverage in the previous 12 months. While frequent, this rate was lower than some pre-ACA predictions. Churning rates were similar after Kentucky's Medicaid expansion and Arkansas' private option, compared to Texas, which did not expand. Common causes were job-related changes and loss of Medicaid/Marketplace eligibility. Churning was associated with disruptions in physician care and medications, trouble obtaining primary and specialty care appointments, and more ED use. Overall, 35-40% of churners felt that it had adversely impacted their quality of care and health. Outcomes were worst among those experiencing gaps in coverage, but even those who churned without becoming uninsured reported adverse effects. Our results indicate policies are needed to reduce churning frequency and to mitigate negative impacts when it occurs.3