The following is co-authored by Aaron and Austin.
They just won’t die. Evidently, the House Republican budget is going to take another whack at Medicaid reform. Today, from the WaPo:
Medicaid, which provides health coverage to low-income families, is the object of a sharply worded review. “Medicaid coverage has little effect on patients’ health,” the report says, adding that it imposes an “implicit tax on beneficiaries,” “crowds out private insurance” and “increases the likelihood of receiving welfare benefits.”
There are studies documenting circumstances under which Medicaid can substantially “crowd out” private insurance. But, as has been explained on TIE, those circumstances don’t necessarily apply to the ACA. Moreover, many people at the low end of the socio-economic spectrum have the option of Medicaid or nothing. They make less than 138% of the poverty line. They aren’t able to afford insurance without massive subsidies.
But, as always, we want to focus on the first statement, the one that declares that Medicaid doesn’t improve patients’ health. That’s not true.
Does anyone really dispute that having health insurance is better than not having health insurance? Anyone who does should put their money where their mouth is. Mediaid isn’t welfare. You don’t get cash. It pays for health care if you need it. And, like all health insurance, it makes people healthier and saves lives. Lots of people say so. Studies confirm this.
That research could be improved with the use of better research design, and methodologically stronger studies have shown that Medicaid is good for HIV mortality, child health, infant mortality, and more.
Which brings us to the Oregon Health Study, an actual randomized controlled trial of Medicaid. We have both written on early results. We’ve also commented on the later results, which are the ones people often seize upon to discredit Medicaid. Again.
People say that it does little to improve the health of people who have diabetes, who are at risk for heart disease, who have high cholesterol, or who have high blood pressure. There are real problems with those assertions. The Oregon study was not powered to detect improvements in those domains. We’re sorry, but it wasn’t. Here’s Austin explaining how it wasn’t set up to detect major changes in cholesterol or the Framingham Risk Score. Here’s Aaron talking about how it couldn’t detect changes in hypertension because the vast majority of people didn’t have it, and the assumptions that underlie arguments for being able to see a change aren’t on point. Same goes for diabetes.
Here’s a summary of those issues.
Why do people have insurance? Most people have it to protect themselves from financial ruin should they get really ill. But they also get it because it provides them the ability and incentive to get health care if they need it. Medicaid is about access. It’s just the first step in the chain of events that leads to better health and wellbeing. It’s not sufficient, but it is often necessary.
Many who argue that insurance should immediately and significantly make a population healthier are glossing over these other issues. They also seem not to care that there are no good RCTs proving that private insurance (or Medicare) do this.
There are lots of legitimate claims to make against Medicaid. It under-reimburses physicians, for instance, causing access problems in some areas and for some beneficiaries. (Guess what. Those problems are even worse for the uninsured, though.) But the natural response to saying docs don’t get paid enough would be to increase Medicaid funding to improve that. Gutting the program will do the opposite.
We look forward to a continuing and lively debate on how to reform the health care system. But declaring that health insurance in the form of Medicaid hurts people or “doesn’t work” ignores the real good that it does for so many people. (And, come on, health insurance is just pushing money around—it isn’t medicine or procedures.) Let’s listen to each other’s arguments and respond to them, instead of repeating talking points past each other.