Michigan’s Medicaid expansion improved both health and finances

Just over a decade ago, Michigan expanded its Medicaid health coverage program, opening it to all adults with very low incomes through the Healthy Michigan Plan.

Now, a new University of Michigan report shows long term benefits of this expansion – not just for individuals but for primary care clinics and hospitals that serve all Michiganders.

It also finds that some of the unique features of Michigan’s Medicaid expansion showed mixed results, such as cost-sharing in the form of copays and monthly premiums, and financial incentives for healthy behaviors.

The findings could inform the 39 other states that have expanded Medicaid as they implement federal changes to Medicaid policy, including requiring enrollees to show they are working if they are able, and copays for some enrollees.

After analyzing a large amount of data from multiple sources, the report shows a positive impact of the first 10 years of Michigan’s Medicaid expansion from 2014 to 2023, including: 

Those with a substantial health burden (a composite measure developed by the U-M team) were less likely to be employed.

But even among those with substantial health burden who enrolled right after the expansion took effect in 2014, employment increased from 19% in 2016 to 32% in 2018.

Using data submitted by Medicaid managed care organizations to the state government, the U-M team was able to look at changes over time in what the Medicaid managed care plans that enroll HMP enrollees pay for administrative costs.

They found that preparing for the community engagement or “work requirement” provision that was implemented very briefly in early 2020 led to increased administrative costs for health plans.

In general, administrative costs accounted for around 12% of health plans’ total amount received from the state.

But in fiscal year 2020, as the plans prepared to meet the provisions of the work requirement, that percentage rose to nearly 14%, and dropped to 13% the following year.

The community engagement requirement was suspended by court order in March 2020 and officially removed from state law in April 2025.

Uninsurance

The new report shows that Medicaid expansion in Michigan was highly effective at reducing uninsurance.

In 2022, the percentage of all Michigan adults who were uninsured was 6.7%, the lowest in the 14-year study period, and about the same as the rate in other states that expanded Medicaid.

All areas of the state saw a drop in uninsurance, mainly among low-income adults, and the differences between regions shrank.

Before expansion, 22% of adults in the northeast section of the state were uninsured, but this dropped to 9% as of 2022.

In that region, 30% of adults are now covered by Medicaid, compared with 15% in the Ann Arbor-Jackson area.

Personal debt, credit ratings and bankruptcy

The new report shows a significant and consistent reduction in HMP enrollees’ amount of medical debt in collections for up to seven years after enrolling.

The researchers used credit report data from 2013 to 2021 for 704,000 HMP enrollees ages 26 to 62 as of 2014, with credit outcomes measured every six months.

They compared them with a sample of 723,000 similar adults with low incomes living in non-expansion states.

They looked at medical debt that had been sent to a collection agency, as well as non-medical debt in collections, subprime credit scores under 600 and bankruptcy in the last two years.

HMP beneficiaries overall had large reductions in their medical debt in collections after enrollment, compared to people with low incomes in non-expansion states.

Enrollees who enrolled for the first time in 2014 and 2015 also experienced reductions in non-medical debt in collections and were less likely to have a subprime credit score.

However, HMP enrollees were more likely to declare bankruptcy after enrollment compared with those in non-expansion states.

These increases in bankruptcies relative to people in non-expansion states occurred in the first two years after enrolling, but then began to drop again.

The report’s authors note that this may be an indicator that those who got covered were able to declare bankruptcy to discharge the medical debt they acquired before getting covered.

Uncompensated care at hospitals

One of the key drivers for Medicaid expansion in Michigan in 2014 was the amount of care that hospitals were providing to people who didn’t have health insurance, which meant that hospitals weren’t getting paid for some or all of the cost of that care.

The difference between cost and payment is called uncompensated care.

The new report confirms that Michigan’s Medicaid expansion was highly effective at reducing uncompensated care for hospitals, cutting it by half and sustaining it at lower levels through 2021, both in terms of total dollars and the percentage of hospital spending.

The report also shows that uncompensated care dropped for hospitals in other expansion states. Meanwhile, it did not drop for hospitals in non-expansion states.

In addition, the proportion of hospitalizations involving self-pay patients – those who are uninsured – dropped by 74% in Michigan from 2013 to 2015, and 69% in other expansion states, while dropping only 11% in non-expansion states.

Michigan and other expansion states saw little change in self-pay hospital stays over the next seven years, the percentage rose in non-expansion states.

Health care safety net

Clinics, health centers and hospitals that serve people with Medicaid coverage also often serve as safety-net providers for people without insurance, or for those with insurance that does not cover the full cost of their care.

The report concludes that expanding Medicaid helped sustain Michigan’s safety-net providers and allowed them to innovate by addressing both the health and social needs of patients.

It also concludes that Medicaid expansion helped the state weather the COVID-19 public health emergency by providing access to health coverage to those who lost jobs and job-related insurance during the first part of the pandemic.

The report also concludes that HMP allowed safety-net organizations to better serve those with COVID-19 and with substance use disorders.

Use of primary care, preventive care and emergency care

The report shows that HMP spurred increased use of primary care services, improved management of chronic conditions, and decreased emergency department visits.

It uses national data to show that low-income Michiganders were significantly more likely than people in non-expansion states to get a checkup and a flu shot after HMP went into effect.

After HMP went into effect, Michigan primary care providers reported offering more same-day and after-hours appointments and following up with patients after an emergency department visit.

They also increased hiring of care managers and community health workers to help those with the highest needs to manage their health conditions and avoid ED visits.

At each survey, about one in three HMP enrollees reported an ED visit in the past year.

In the first year after enrolling in HMP, 2% of enrollees went to the ED more than five times in a year.

This dropped slightly by the second year of enrollment, especially among those who saw a primary care provider.

ED visits also decreased among those who had asthma, cardiovascular disease, chronic lung disease, or diabetes.

Surveys showed that 74% of enrollees who had gone to an ED didn’t contact their primary care provider before going to the ED.

Many said their provider’s office was closed when they needed care urgently, or that they went to the ED because they couldn’t get an appointment with their regular provider.

Nearly two-thirds of those who made contact with their primary care provider before going to the ED said the provider had instructed them to go to the ED.

Overall health burden improvement

The new report includes an analysis of the health burden faced by HMP enrollees based on survey results from more than 9,000 people who had been enrolled at least a year, with some enrollees surveyed multiple times and a subset completing further in-depth interviews.

Surveys were conducted by phone in English, Spanish and Arabic.

The U-M team created a health burden score, based on each person’s overall self-reported health, the number of days in the last month where they said their physical or mental health was not good, and the number of days where they said their health interfered with usual activities.

In all, 17% of HMP enrollees had a substantial health burden and 13% had a moderate health burden.

Half of those who had substantial or moderate health burden showed improvement two to three years later.

More than half of all enrollees with a substantial health burden said they had gone to the emergency department in the last 12 months – a much higher percentage than in the group with minimal health burden.

Encouraging healthy behaviors

The HMP’s original design encouraged all enrollees to fill out a health risk assessment or HRA, a tool designed to help their providers understand opportunities to lower their risk of future health problems by engaging in preventive care and healthy behaviors.

The new report shows that HMP enrollees with more health problems at the time of enrollment reported improvements as time went on.

Primary care providers interviewed by the U-M team reported that this improvement was linked to behavior change, which benefits from sustained engagement with the health care team.

But the HRA form itself didn’t prompt PCPs to identify which patients should get more support from other members of the team or a program to encourage more activity, stopping tobacco use or eating healthier diets.

And many HRAs completed by enrollees didn’t get entered into their electronic medical record.

In all, the report concludes, HRA completion wasn’t linked to getting more preventive care. Instead, the key factor was whether or not the person saw their PCP regularly.

The report also shows that most enrollees and providers didn’t know there was a financial incentive for completing the HRA or engaging in a healthy behavior.

But even among those who knew this, the financial incentive wasn’t the motivation for behavior change, the report finds.

The report recommends that states seeking to encourage healthy behaviors among Medicaid enrollees should focus on sustained healthy behavior engagement, increased integration of health risk tools into electronic medical records, and communication on a regular basis to enrollees and providers about the goals, processes and incentives of the program.

Data sources

In addition to the health care utilization data, credit reporting data and surveys described above, the report includes information from interviews with enrollees, health care providers, and current and former Michigan health care leaders.

It also includes data from the federal Behavioral Risk Factor Surveillance System health survey data for low income individuals in Michigan and non-expansion and selected expansion states, federal Healthcare Cost and Utilization Project hospital discharge data, Medicare cost reports, and American Community Survey data on insurance status for Michigan and for states that expanded Medicaid before 2014 or had not expanded as of 2022.

Additional authors: In addition to Ayanian, contributors to the report include Erin Beathard, M.P.H., M.S.W.; Nora Becker, M.D., Ph.D.; Nicholas Box, M.P.A.; Sarah Clark, M.P.H.; Lisa Cohn, M.S.; Anne Cowan, M.P.H.; Acham Gebremariam, M.S.; Susan Goold, M.D., MHSA, M.A.; Richard Hirth, Ph.D., M.A.; Matthias Kirch, M.S.; Sunghee Lee, Ph.D., M.S.; Helen Levy, Ph.D.; Minal Patel, Ph.D., M.P.H.; Erica Solway, Ph.D., M.S.W., M.P.H.; Andrei Stefanescu, Ph.D, M.S.; and Renuka Tipirneni, M.D., M.Sc.

Citation: Healthy Michigan Plan Summative Evaluation Report for the Demonstration Period Ending December 31, 2023. DOI: 10.7302/28712

Sign up for Health Lab newsletters today. Get medical tips from top experts and learn about new scientific discoveries every week

Sign up for the Health Lab Podcast. Add us wherever you listen to your favorite shows.