Into the maelstrom of debate over whether
Medicaid should cover more people comes a new study by
Harvard researchers who found that when states expanded their Medicaid programs and gave more poor people
health insurance, fewer people died.
Medicaid
expansions are controversial, not just because they cost states money,
but also because some critics, primarily conservatives, contend the
program does not improve the health of recipients and may even be
associated with worse health. Attempts to research that issue have
encountered the vexing problem of how to compare people who sign up for
Medicaid with those who are eligible but remain uninsured. People who
choose to enroll may be sicker, or they may be healthier and simply be
more motivated to see doctors.
The
New England Journal study reflects a recent effort by researchers to
get around that problem and allow policy makers to make “evidence-based
decisions,” said Katherine Baicker, an investigator on the study who
served on former President George W. Bush’s Council of Economic
Advisers.
“I
think it’s a very significant study in part because of the paucity of
studies that have really looked at health outcomes of insurance
coverage,” said Karen Davis, the president of the Commonwealth Fund, a
nonpartisan research foundation. “Actual mortality studies are few and
far between. This is a well-done study: timely, adds to the evidence
base, and certainly should raise concern about the failure to expand
Medicaid coverage to people most at risk of not getting the care that
they need.”
The
study, conducted by researchers from Harvard’s School of Public Health,
analyzed data from three states that had expanded their programs in the
last decade to cover a population not normally eligible for Medicaid:
low-income adults without children or disabilities. The new law also
expands coverage to a similar population nationally.
Researchers
looked at mortality rates in those states — New York, Maine and Arizona
— five years before and after the Medicaid expansions, and compared
them with those in four neighboring states — Pennsylvania, Nevada, New
Mexico and New Hampshire — that did not put such expansions in place.
The
number of deaths for people age 20 to 64 — adults too young to be
considered elderly by the researchers — decreased in the three states
with expanded coverage by about 1,500 combined per year, after adjusting
for population growth in those states, said Dr. Benjamin D. Sommers, a
physician and an assistant professor of health policy and economics who
was an author of the study.
In
the five years before the expansion, there were about 46,400 deaths per
year, while in the five years after the expansion, there were about
44,900 deaths per year. During the same period, death rates in the four
comparison states increased, said Dr. Sommers, who began a yearlong
stint as an adviser to the federal Department of Health and Human
Services after research for the study was completed, but before its
publication.
When
researchers adjusted the data for economic factors like income and
unemployment rates and population characteristics like age, sex and
race, and then compared those numbers with neighboring states, they
estimated that the Medicaid expansions were associated with a decline of
6.1 percent in deaths, or about 2,840 per year for every 500,000 adults
added.
While
the data included all deaths, not just deaths of Medicaid recipients,
the decline in mortality was greatest among nonwhites and people living
in poorer counties, groups most affected by expanded Medicaid coverage.
“I
can’t tell you for sure that this is a cause-and-effect relationship,”
that the Medicaid expansion caused fewer non-elderly adults to die, Dr.
Sommers said. “I can tell you we did everything we could to rule out
alternative explanations.”
Several experts with varying views on the Affordable Care Act said the
study, which was completed long before the Supreme Court hearings on the
law, was conducted by highly qualified researchers who carefully
analyzed the available data. Still, they and the study’s authors pointed
to limitations in the data, noting that the mortality figures represent
county-level statistics, not individual deaths.