That leaves policymakers with "a much more nuanced and complex picture" of the potential benefits of expanding Medicaid, said Baicker, an economist.
It also debunks a widespread perception that having Medicaid is no better, and maybe even worse, than being uninsured. A federal-state partnership, Medicaid covers more than 55 million low-income and severely disabled people, ranging from poor children to nursing home residents. It pays providers less than Medicare or private insurance.
Obama's health care law envisioned expanding Medicaid to anyone making up to 138 percent of the federal poverty line, or nearly $15,860 for an individual. About 15 million people — mostly adults with no children living at home — could eventually be covered if all states expand. But the Supreme Court last year gave states the right to reject the expansion without jeopardizing the rest of their federal Medicaid funds.
The study found that having Medicaid reduced rates of depression by 30 percent and virtually eliminated catastrophic medical expenses due to a serious accident or the sudden onset of a life-threatening illness. People with Medicaid had better access to doctors, preventive care, prescriptions and hospitals. They also used their benefits, consuming about $1,200 a year more per person in health care services than do the uninsured.
But Medicaid had no significant effect on blood-pressure readings, high cholesterol or elevated blood sugar levels, although it did increase the probability that people with diabetes would be diagnosed.
The study is unusual because it took advantage of a state policy decision to create a natural experiment. A 2008 Medicaid expansion in Oregon used lottery drawings from a waiting list to determine who would get coverage. That created two populations: those who got in, and those left out — a control group of people who remained mostly uninsured. Comparing the two randomly selected groups gives the research a high degree of scientific rigor.
Obama administration officials, in a written reaction published alongside the study, suggested that two years might not be enough time for differences to emerge in physical health results, particularly for patients with chronic conditions influenced by lifestyle choices. They also noted that the Oregon group was relatively small, about 12,000 people, compared to the millions who will gain coverage next year. The Oregon pool might not have been large enough to tease out valid results, they suggested.
"This study did not or could not address many important health benefits of health insurance, including early detection of cancer, a reduction in sick days from school or work, and a reduction in mortality," wrote Richard Kronick and Andrew Bindman, health policy experts with the federal Health and Human Services department.
Lead researcher Baicker said it's possible that having insurance alone isn't enough to get control of lifestyle-related health problems like high cholesterol. "That said," she emphasized, "there are known treatments for these conditions, and the clinical literature suggests improvements are gettable within less than two years."
Expanded Medicaid will be available starting Jan. 1, and uninsured people can start signing up this fall. So far, 21 states plus Washington, D.C., have accepted the expansion, while 14 states have turned it down. Another 15 states are still weighing options.
Nearly all the states refusing are led by Republicans. Several of the states accepting have Republican governors, but most are led by Democrats. Washington will pick up the entire cost of the expansion for the first three years, and 90 percent over the longer haul. It's estimated that less than $100 billion in state spending could trigger nearly $1 trillion in federal dollars over a decade.