Insurance Churning

Back to Policy Briefs

November 2016

Policy Brief courtesy of the University of Michigan Center for Healthcare Research and Transformation

Elizabeth Austic, Emily Lawton, Melissa Riba and Marianne Udow-Phillips
Center for Healthcare Research & Transformation

Download printable version (Adobe PDF)

Introduction

The uninsured rate has declined substantially since provisions of the Patient Protection and Affordable Care Act (ACA) came into effect. Nevertheless, many individuals continue to experience instability in insurance coverage. Transitions between different insurance plans, as well as between insured and uninsured status, are often referred to as “insurance churning.” The causes of insurance churning vary. Changes in job status may result in loss of coverage or transition to a new insurance plan. Eligibility for Medicaid or plans with marketplace subsidies may change based on changes in family composition or fluctuations in income. Nationally, three to five percent of members are dropped each month by health plans offered on the health insurance exchange due to non-payment of premiums.1

Insurance churning can affect quality, cost, and continuity of care.2 Individuals may avoid seeking health care when they need it during gaps in insurance coverage. Even if consumers maintain continuous coverage while transitioning between different insurance plans, they may find that their regular health care providers do not accept their new insurance plan. In addition, when they change health plans, research shows that medication compliance is often disrupted.2 These and other churning-related problems tend to be exacerbated by uncertainty about what new co-pays or deductibles might be as well as by known increased costs associated with new insurance plans.

A recent study found no evidence of significant increases or decreases in broader indicators of insurance churning since the introduction of the ACA in three states (Texas, Kentucky, and Arkansas).2 Under the new administration, depending on which provisions of the ACA are repealed, replaced, de-funded, or retained in the coming years and how these changes are implemented, rates of insurance churning could change dramatically. Uncertainty and changes in availability of coverage and out-of- pocket costs could contribute to greater churning during the implementation of new health care policies.3

Between 2013 and 2015, data from the Census Bureau show that the proportion of Michiganders who reported no source of health insurance declined by five percentage points—from 11 percent to 6 percent.4 Using data from the Center for Healthcare Research & Transformation’s Cover Michigan Survey, this brief explores consumer experiences with insurance churning and access to care within the state of Michigan for approximately a one-year period in 2014–2015.

Key Findings

  • Medicaid recipients had the most instability in their coverage status of all respondents to the survey. Medicaid recipients were also seven times more likely to have experienced a temporary uninsured period in the past year compared to respondents with employer-sponsored or individual coverage.
  • Those with individually purchased coverage in 2014 were the most likely to switch to a different type of coverage in 2015. Among respondents with an individually purchased plan in 2014, less than half re-enrolled in the same plan in 2015, and nearly a third transitioned to Medicare or to an employer-sponsored plan in 2015.
  • Those with employer-sponsored coverage experienced the least amount of churning compared to respondents with other coverage. Ninety-four percent of respondents with employer-sponsored coverage remained continuously insured from 2014 to 2015.

Transitions in Medicaid Coverage, 2014–2015

Twenty-two percent of respondents with Medicaid experienced a temporary period of being uninsured, compared to only 4 percent of respondents with employer-sponsored and individually purchased coverage. Eight percent of respondents with Medicaid in 2014 were uninsured by the time they were surveyed in 2015—a higher proportion than the overall state average of 5 percent.5 Figure 1

Figure of Transitions in Medicaid Coverage, 2014-2015

Sixty-two percent were continuously enrolled in Medicaid for the entire year and 22 percent gained insurance coverage after temporarily being uninsured; of these, 19 percent re-enrolled in Medicaid and 3 percent secured employer-sponsored coverage. By comparison, 84 percent of Americans with Medicaid maintained continuous coverage all year.6

Transitions in Individually Purchased Coverage, 2014–2015

Compared to respondents with other types of coverage, those with individually purchased coverage in 2014 were the most likely to switch to a different form of coverage in 2015. Thirty-one percent switched to a plan outside of the individual market, while an additional 3 percent made this switch after being temporarily uninsured. Figure 2 A total of 5 percent of those with individually purchased coverage were temporarily uninsured during the year prior to the survey. Six percent of those with an individual plan in 2014 became and remained uninsured at the time of the survey in 2015.

Only 48 percent of respondents who had individual coverage in 2014 re-enrolled in the same plan in 2015 (1 percent after an uninsured period). Eleven percent switched directly from one individual plan to another, while an additional 1 percent made this switch after a temporary uninsured period.

Figure of Transitions in Individually Purchased Coverage, 2014-2015

Transitions in Employer-Sponsored Coverage, 2014–2015

Those with employer-sponsored coverage in 2014 experienced the least amount of churning compared to respondents with other coverage. They also had the lowest rates of becoming and remaining uninsured—just 2 percent. Figure 3 Ninety-four percent of respondents with employer-sponsored plans maintained coverage continuously all year (2014 to 2015). This is comparable to national data indicating that around 92 percent of Americans with employer-sponsored plans had continuous coverage from year to year.5

Seventy-two percent of Michigan respondents were continuously covered by the same employer-sponsored plan, 16 percent directly switched to a different employer-sponsored plan and 6 percent gained coverage through either an individually purchased plan,
Medicaid, or Medicare.

Just 4 percent reported being temporarily uninsured during the previous 12 months. Among this group, less than 1 percent re-enrolled in the employer-sponsored plan they had previously, while 3 percent enrolled in a different employer plan by the time of the survey in 2015.

Figure of Transitions in Employer-Sponsored Coverage, 2014-2015

Age Differences in Insurance Churning

Overall, 27 percent of Michigan adults with Medicaid, employer-sponsored, or individually purchased coverage in 2014 reported experiencing one or more types of insurance churning between 2014 and 2015. Types of churning included switching to a different insurance plan, being temporarily uninsured during the past 12 months, or being uninsured at the time of survey in 2015.

The highest rates of insurance churning—47 percent—were reported by the youngest group of respondents (18–24 year olds). Churning decreased significantly as age increased, with just 18 percent of those 55 and older having experienced churning. Figure 4

Figure of Age Differences in Insurance Churning, 2014-2015

Reasons for Changing Insurance Coverage

Overall, 22 percent of Michiganders with Medicaid, employer-sponsored, or individually purchased insurance in 2014 changed insurance coverage in 2015. By far, the most common reason for changing insurance coverage was a change in job (44 percent). Another 14 percent reported that their employer changed the coverage provided. Only 8 percent reported cost as the reason for this change.

Conclusion

Although the uninsured rate has declined substantially in Michigan since provisions of the ACA came into effect, there is still a great deal of instability in insurance coverage, particularly for younger respondents, and for those with individually purchased coverage or Medicaid.

Less than half of those with an individually purchased insurance plan in 2014 re-enrolled in the same plan in 2015. Many of these respondents may have purchased insurance for the first time during the first post-ACA open enrollment period in 2014 and then decided to switch to a different insurance plan in 2015. Some of the churning in this group may also be attributable to changes in eligibility for other types of insurance. Nearly a third of those with an individually purchased insurance plan in 2014 transitioned to Medicare or to an employer-sponsored plan in 2015.

Respondents with Medicaid in 2014 were less likely to switch plans, but more likely to be uninsured in 2015, or to report temporarily being without coverage in the previous 12 months. In contrast, respondents with an employer-sponsored health plan in 2014 were the least likely to become or remain uninsured.

These findings suggest Michigan consumers may experience different types of insurance churning depending on their type of coverage. Whether due to switching to a different plan or experiencing a temporary period of being uninsured, such disruptions in coverage may have implications for consumers being able to access health care and preferred health care providers. These findings show continuing opportunities for Michigan insurance companies to reach out to the uninsured and develop strategies to help reduce uninsured periods.

Methodology

The survey data presented in this brief were produced from a series of survey questions added to the Michigan State University Institute for Public Policy and Social Research (IPPSR) quarterly State of the State Survey. The Cover Michigan Survey 2015 was fielded between October and December 2015, and included a sample of 972 Michigan adults, with a 17 percent response rate. The margin of error for the entire sample was ±3.9 percent. The sampling design, a random stratified sample based on regions within the state, was a telephone survey conducted via landline and cellular phones of Michigan residents.

For analytical purposes, survey data were weighted to adjust for the unequal probabilities of selection for each stratum of the survey sample (for example, region of the state, listed vs. unlisted telephones). Additionally, data were weighted to adjust for non-response based on age, gender, and race, according to population distributions from the 2009–2013 American Community Survey data. Respondents who reported both Medicare and Medicaid coverage were considered Medicaid recipients for the purpose of this analysis. Because the survey did not distinguish between the different types of Medicare plans available to consumers (i.e., “original” Medicare, Medicare Advantage plans and Medicare Part D), we were not able to develop meaningful analyses of churning for this population, therefore respondents who reported that Medicare was their only form of coverage were excluded from this analysis. Results were analyzed using SPSS 24 software. Statistical significance of bivariate relationships was tested using z tests or chi-square tests for independence. All comparisons are statistically significant at p < 0.05 unless otherwise noted. A full report of the IPPSR State of the State Survey methodology can be found at: http://ippsr.msu.edu/soss/.

Footnotes

(1) V . Dickson, “Income-based ‘churn’ in coverage less common than feared,” Modern Healthcare, April 2015: http://www.modernhealthcare.com/article/20150422/NEWS/150429959 (accessed 6/6/2016).

(2) B.D. Sommers, R. Gourevitch, B. Maylone, R.J. Blendon, and A.M. Epstein, “Insurance Churning Rates For Low-Income Adults Under Health Reform: Lower Than Expected But Still Harmful For Many,” Health Affairs, Oct. 2016,doi: 10.1377/hlthaff.2016.0455

(3) E. Saltzman and C. Eibner, “Donald Trump’s Health Care Reform Proposals: Anticipated Effects on Insurance Coverage, Out-of-Pocket Costs, and the Federal Deficit,” The Commonwealth Fund, September 2016. http://www.commonwealthfund.org/Publications/Issue-Briefs/2016/Sep/Trump-Presidential-Health-Care-Proposal (accessed 11/16/2016).

(4) J.C. Barnett and M.S. Vornovitsky, “Health Insurance Coverage in the United States: 2015,” Current Population Reports, Sept. 2016.

(5) Healthy Michigan Plan Enrollment Statistics (Lansing, MI: Michigan Department of Community Health, Feb. 2015): http://www.michigan.gov/mdch/0,4612,7-132-2943_66797—,00.html (accessed 6/6/2016).

(6) J. Fangmeier and M. Udow-Phillips. Dynamics of Health Insurance Coverage in the United States, 2014. (Ann Arbor, MI: Center for Healthcare Research and Transformation, 2016). http://www.chrt.org/publication/changes-healthinsurance-coverage-united-states-2014/.