The fundamental shortcoming of the American health-care system is that we spend more than other developed nations, yet are less healthy. One reason for this, policy analysts have long argued, is that so many of the nation’s poor, lacking health insurance, can’t afford regular visits to the doctor and the sort of preventive care that comes with it. They allow chronic problems to grow into acute ones and then go to the emergency room, where they can’t be turned away—but where care is particularly expensive. This is clearly a very inefficient way to treat medical problems.
President Obama and others, in making the case for the Affordable Care Act, repeatedly argued that expanding insurance would steer people away from the emergency room and into doctors’ offices. They’d be healthier as a result, and the cost of their care would drop. A paper published today in the journal Science—just as Obamacare starts to cover people—calls this assumption into question.
The study, by researchers from Harvard, MIT, Columbia, the Providence Portland Medical Center, and the National Bureau of Economic Research, took advantage of a limited Medicaid expansion instituted in Oregon. Starting in 2008, the state selected 30,000 new Medicaid beneficiaries from a waiting list of 90,000, using a lottery. Because the selection was random, it allowed the researchers to isolate the effect of getting health insurance (in the form of Medicaid) the same way a randomized drug trial is meant to isolate a drug’s effects. Such “natural experiments” are all the rage in social science.
The researchers found that—according to hospital records (and contra President Obama)—expanding Medicaid didn’t decrease emergency department visits; it increased them by 40 percent. “We looked across a wide range of groups of people and types of visits and we see increases across the range,” says Sarah Taubman, an epidemiologist at the NBER and one of the authors. “Nowhere do we see decreases.”
From a basic economics point of view, this outcome makes sense. Medicaid covers emergency room visits, and if you make something cheaper, people will consume more of it. Taubman and her co-authors found that the new Medicaid recipients reported that they were going to primary care facilities more frequently, too. In other words, they were going to the doctor more and they were going to the ER more. Perhaps most frustrating for Obamacare proponents, consuming all this additional health care doesn’t seem to have made people physically healthier. In earlier research, Taubman et al didn’t find any statistically significant improvements in how well the new Medicaid recipients managed their high blood pressure, cholesterol, or diabetes.
The Science paper is a major addition to a fairly thin and contradictory literature on health insurance and emergency room visits. Research on the Romneycare health-insurance expansion in Massachusetts found that ER visits either stayed the same or went down after the law took effect, while the landmark RAND Health Insurance Experiment from the 1970s found that more comprehensive coverage increased emergency department use.
The Oregon study in question looked only at the first 18 months after the state expanded Medicaid. Behavior may have changed as people adjusted to having health insurance and not needing to rely on emergency departments—facilities meant for use as a last resort. And Taubman et al did find that, in self-reports, the new Oregon Medicaid recipients said they felt less depressed and were suffering less financial strain than earlier. Contradicting the ambiguous diagnostic results, the new recipients also said they felt healthier. Health, of course, is a complex thing; there may be benefits from feeling better, even if they don’t show up in blood work.
Moreover, in a country where medical bills are a leading cause (PDF) of personal bankruptcy, freeing people from some portion of that burden is an accomplishment. It’s what health insurance is designed to do. It would be nice to think that strengthening the health-care safety net would also bring costs down—and, for good measure, improve care—but that may be asking too much.