People newly enrolled in a health insurance program for the poor were more likely to seek emergency room treatment than people who remained uninsured, Boston-area researchers have found, providing the best evidence to date that the national Medicaid expansion that began this week is unlikely to lead to a decline in emergency services.

The study was published online Thursday by the journal Science, just as millions of Americans have become newly eligible for Medicaid coverage under the Affordable Care Act. Some politicians have suggested that uninsured people who did not have a regular doctor or who delayed treatment as their condition worsened would, once they had coverage, get more of their care in a doctor’s office, rather than the emergency room, where providing it is often more costly.

Previous research on what happened to emergency room usage in Massachusetts, which expanded its Medicaid program and required that most residents have health insurance in 2006, have reached conflicting conclusions. But the new study, of about 25,000 low-income adults randomly selected in 2008 to enroll in Oregon’s Medicaid program, found that the newly insured increased their visits to emergency rooms.

Advertisement - Continue Reading Below

Over an 18-month period, about 42 percent of the new Medicaid enrollees visited the emergency department. In the same period, about 35 percent of those who did not receive Medicaid visited the emergency department. Similar increases were reported by the new Medicaid recipients in an earlier study that looked at use of prescription drugs, hospital stays, and outpatient visits.

“Basic economic theory is, if you lower the price, people use it more,” said Amy Finkelstein, a Massachusetts Institute of Technology economist and a senior author on the paper.

It was not a foregone conclusion in this case, however, that reducing the cost of an emergency room visit would increase use, she said, because the cost of a visit to a primary care doctor and of preventive services that may have helped them avoid the emergency room also decreased with insurance coverage.

Yet, emergency department use among those on Medicaid increased during businesses hours, nights, and weekends. While there was no increase in visits classified as non-preventable emergencies, there was an increase for visits deemed preventable or treatable by a primary care doctor.

A 2011 study found that overall emergency department visits increased in Massachusetts in the two years after the state expanded insurance coverage under the 2006 state health care law, though visits for “low severity” problems declined slightly. Dr. Peter Smulowitz, an emergency physician at Beth Israel Deaconess Medical Center and lead author on that study, said he and colleagues have more recently reviewed emergency department use across Massachusetts and found a small increase in pockets of the state that had seen the largest gains in insurance coverage. The study is pending publication.

Data published in the New England Journal of Medicine in 2011, however, found that ER usage was already increasing in Massachusetts and nearby states before the expansion of health insurance coverage here, and that the law did not change the trend in Massachusetts when compared to the other states.

The Oregon study is unique in that it is a randomized controlled study, considered the gold standard in medical research but rarely feasible in health policy research. The state of Oregon created a valuable study scenario when, because the state had money only for a small expansion of the program, it held a lottery for Medicaid coverage, providing insurance to some people and leaving others uninsured.

Past work by principal investigators Finkelstein and Katherine Baicker, a professor of health economics at Harvard School of Public Health, and their colleagues at the National Bureau of Economic Research in Cambridge has found that the lottery winners were more likely to report feeling better about their mental and physical health and had less financial strain, including fewer bills sent to collection. But there was no improvement in key health factors, such as blood pressure or blood sugar levels, as compared with the uninsured.

With the latest study, Baicker said, the body of research out of Oregon has disproved both the worst and best predictions for Medicaid—that it is an expensive program that does little to improve access to care and overall health, or alternatively, that it is a money-saving program that clearly improves health. The results are far more nuanced.

“Policymakers should make decisions based on this evidence that the program has real costs. It’s not free,” she said. “And, it has real benefits. Beneficiaries are clearly better off.”