A strong majority of Americans oppose proposed Medicaid funding cuts, with bipartisan agreement that the program is essential to health care access and mental health treatment, according to a new NAMI-Ipsos poll released Thursday.
The findings come as Congress debates a House-passed budget resolution that would reduce federal Medicaid spending. According to the National Alliance on Mental Illness (NAMI), 71% of respondents said they disagree with cutting Medicaid to curb government spending, and 78% believe such cuts would worsen health outcomes for low-income individuals and families.
“Across party lines, Americans agree on two important priorities: protecting Medicaid and doing more to address mental health,” said Daniel H. Gillison, Jr., CEO of NAMI. “We are facing a mental health crisis, and the message from the public is clear—they want our policymakers to do more, not less.”
Mental health tied closely to Medicaid access
Medicaid covers more than 72 million people in the U.S. and pays for one in four dollars spent on mental health and substance use care. It also insures 40% of non-elderly adults with a mental health or substance use condition, according to NAMI.
Three in four Americans support expanding Medicaid to increase access to affordable mental health care, and 85% support protecting federal Medicaid funding for that purpose. The survey also found that Americans view Medicaid more favorably (71%) than private or employer-sponsored health insurance (64%).
Growing dissatisfaction with mental health care
Nearly eight in 10 Americans are dissatisfied with the state of mental health care in the country, and a majority (60%) believe Congress is doing too little to address the issue—an increase from last year’s findings.
Only 52% believe people with mental health conditions are getting better treatment now than a year ago, and three in five Americans agree that Medicaid and mental health care should be among the federal government’s top funding priorities.
The poll was conducted March 21–23, 2025, with a nationally representative sample of 2,049 adults. The margin of error is ±2.2 percentage points.
