Skip to content

Medicaid Expansion Reduces Mortality, Study Finds

As states decide whether to expand their Medicaid programs to cover low-income childless adults, the impact of their choices became clearer today in a study showing a reduction of mortality in states that have already made that move.

The research published in the New England Journal of Medicine found a 6.1 percent reduction in mortality among low-income adults between the ages of 20 and 64 in Maine, New York and Arizona — three states that expanded coverage since 2000, compared with similar adults in New Hampshire, Pennsylvania, Nevada and New Mexico, neighboring states that did not do so.

The decline in mortality, by an overall 19.6 deaths per 100,000 adults, was especially pronounced among older individuals, minorities and residents of the poorest counties.  The researchers analyzed data spanning  five-year periods before and after the three states extended their Medicaid coverage to poor, childless adults.

The study also found “improved coverage, access to care and self-reported health” among the newly covered adults.

“It seems intuitive, but there’s been surprisingly little evidence so far,” said lead researcher Benjamin D. Sommers, M.D., Ph.D., an assistant professor of health policy and economics at the Harvard School of Public Health.  “There’s been some [research] on pregnant women and children, but much less on adults.  And right now there are a significant number of people arguing that Medicaid is worse than nothing at all.”

The Supreme Court on June 28 struck down as unduly coercive a provision of the 2010 federal  health care law that sought to force all states to extend Medicaid coverage to everyone with incomes up to 133 percent of the federal poverty level — currently  $14,856 for individuals and $25,390 for a family of three. Although the federal government will pay the full cost of the expanded coverage for three years starting in 2014, and at least 90 percent thereafter, a number of state governors have said they will not approve the wider coverage.

The study’s authors — Sommers, Katherine Baicker, Ph.D. and Arnold M. Epstein, M.D. — said their research results are consistent with previous analyses finding an 8.5 percent reduction in infant mortality and a 5.1 percent drop in child mortality  as a result of Medicaid expansions in the 1980s.

The authors cautioned that their study “cannot definitively show causality,” because other factors might have contributed to the reduction in death rates in the population newly covered by Medicaid. Among those factors, they said, was the possibility that “expanding coverage had positive spillover effects through increased funding to providers, particularly safety-net hospitals and clinics.” But they said they were not aware of any large-scale changes in health policy in the three states they studied.

“This answers the question of what happens when you give people Medicaid who didn’t already have coverage, as opposed to comparing people who have Medicaid with people who have something else,” said Sommers.  “The latter is not apples to apples, because Medicaid recipients are usually sicker and with worse socioeconomic conditions.”