: A randomized-controlled study published in the New England Journal of Medicine by a group of nation's top health policy scholars has found that Medicaid has no measurable effect on any of the objectively measured physical health outcomes the study examined.
In its second-year results, the Oregon Health Insurance Experiment, which randomly assigned Medicaid to 10,000 people in Oregon, and compared them with a randomly selected control group, found that those who got Medicaid did not on average have healthier blood pressure, cholesterol levels, or diabetic blood pressure control than those who did not get Medicaid. Those with Medicaid did see some reduction in out of pocket health expenses. They were also less likely to be diagnosed with depression.
The Medicaid recipients also ended up utilizing a lot more health care—care that has to be paid for—than those who didn't get coverage. But they didn't use the emergency room any less than the control group.
This study is perhaps the best and most important study of Medicaid's health effects ever conducted, and it has huge implications for public policy—in particular for Obamacare's Medicaid expansion, which is supposed to account for about half of the law's increase in health coverage. Obamacare supporters had used the results from the study's first year, which showed large gains in self-reported health, to argue that the law's expansion of Medicaid was justified. The second-year results significantly complicate that argument.
But there was always good reason to be skeptical that the study would not reveal that Medicaid coverage improves health on the health markers measured. As I noted when the first results from the study were published, even though it is true that self-reported health status rose amongst the population assigned Medicaid, the bulk of the improvement in self-reporting occurred prior to the provision of any care. Just because the Medicaid recipients said they felt better, in other words, did not necessarily suggest that they were measurably healthier. And now, it turns out, they weren't, at least not by any measure the study examined.
