When Congress passed the Affordable Care Act in 2010, a Harvard professor, Dr. Benjamin Sommers, wondered whether the law's directive to the states to expand Medicaid would actually benefit anyone.
"We know the Medicaid expansion is coming. What can we conclude about the potential impact on population health?" was the question Sommers sought to answer, he said in an telephone interview Friday.
There was a laboratory of sorts: Three states, in very different regions of the country, had expanded the health-care program on their own in the early 2000s, covering so called "childless adults" excluded everywhere else, including Alaska.
Sommers and two colleagues published their study in 2012 in the New England Journal of Medicine. It showed significant improvements in health, more visits to doctors and reductions in mortality in the study group of Arizona, New York and Maine.
But the study was no longer an academic exercise. The Supreme Court ruled that Congress couldn't compel the states to expand Medicaid -- it would have to be voluntary.
"The paper ended up coming out a month or so after the Supreme Court ruling and suddenly the states were making this decision," said Sommers, a physician and assistant professor of health policy and economics. "Certainly the relevance got ratcheted up a notch as states are now having to make a big decision that two years ago no one knew they were going to have to make."
Sommers was interviewed from his Boston office at the Harvard School of Public Health just as Gov. Sean Parnell was announcing his decision on the other side of the continent Friday that Alaska would not expand Medicaid.
If Alaska is anything like Arizona, New York or Maine, Parnell's decision may directly result in more deaths here, the research showed.
"We set out when we did the study to figure out which states had done something like the ACA in the decade before so we'd have enough time to see the aftereffects," Sommers said.
Looking at 19- to 64-year-olds over the five years before the three states expanded Medicaid and the five years after, the study group comprised about 500,000 new enrollees, Sommers said.
"Overall, the decline in mortality amounted to about 2,800 deaths prevented per year in these states," Sommers said. "For every 176 adults that gained coverage through Medicaid, our study identified a decline of one death per year."
But Sommers cautioned that variations among state populations, Medicaid rules and other factors makes applying the study to Alaska or any other state inappropriate.
"I wouldn't say that our study is predicting specific numbers for any given state," Sommers said. "It's saying, here's what we observed in these three states for a population that is similar for what the ACA is trying to do."
Conservative critics of the study point out that it showed Maine's mortality rate actually increased compared to neighboring New Hampshire's, which did not expand Medicaid at the time. And they faulted the study for comparing the three subject states with others that had significant differences in populations, like New York to Pennsylvania.
Sommers has been a senior adviser to the U.S. Department of Health and Human Services in the Obama administration but said the study was financed by the Harvard School of Public Health. One of the co-authors, Katherine Baicker, served on President George W. Bush's Council of Economic Advisors.