We’re now having “the great Medicaid debates.” I wish it were all fun and games, like Go. But it isn’t. It’s serious, on at least two levels. The level that will get most attention pertains to policy. What should we do about Medicaid financing? How much flexibility should be afforded to states? How generous should the program be?
In some sense there is no right answer to those questions. That brings me to the second level of the debate, the one I have spent and will spend more time on this blog writing about: what are the sound research methods for assessing the consequences of Medicaid policy (or any policy, for that matter)? There are right answers to this question. Some research methods are appropriate. Some are not.
I have focused on this aspect of the debate because you can find the policy debate so many other places. What more could I say that isn’t said elsewhere, and why should my policy preference matter to you? What you won’t find elsewhere is a deeper understanding of the research and what studies are most credible and why. In fact, elsewhere you will mostly — but not entirely! — find misuse of research, cherry picking of studies and results within them. That is not what I do. I think you should care about sound research methods. That’s why I do what I do.
Having said that, I recognize that I don’t always write about research in an accessible way. Some can follow. Some can’t. Yet I do want everyone to understand, so I’ll keep trying. I am encouraged and delighted that Jon Cohn understands. His new post nicely summarizes the key issues about studies of Medicaid outcomes.
The basis for the claim [that Medicaid is worse than no insurance] is a handful of studies, chief among them a clinical study from researchers at the University of Virginia, in which people on Medicaid end up with worse outcomes than people with no insurance at all. As a general rule, the problem with studies like these is the underlying differences in the groups being studied: Simply put, the uninsured, overall, tend to be healthier than people on Medicaid. That’s going to skew the raw results, with the uninsured getting better medical outcomes. The results will suggest correlation, not causation.
Good studies adjust for this fact and, to their credit, the University of Virginia researchers tried to do that. But, as Austin Frakt and Harold Pollack have written, the researchers controlled only for “observable” factors–age, disability, presence of certain medical conditions, that sort of thing–that were present in the medical records made available to them. Anybody who has studied the Medicaid population closely–and by that I include not just academics but also journalists, like myself, who have interviewed providers, patients, and social workers extensively over the years–will tell you that the differences in the populations go beyond these clinical markers.
The University of Virginia researchers actually acknowledged as much in their paper, mentioning a whole list of factors they couldn’t address. Hospital staff, for example, are more likely to help the sickest patients navigate the enrollment process into Medicaid, which can be famously difficult. Medicaid patients may also end up in worse health because they have fewer family and community supports to keep them healthy or to get them help if something goes wrong. One key sign that the University of Virginia paper obscures the true distinctions among its patient populations is that Medicare patients also ended up with worse outcomes than people with no insurance, even after their adjustments. If there’s a theory for why seniors would be better off uninsured than on Medicare, I’m eager to hear it.
Austin and Harold, bona fide experts both, offer more details in their exhaustive blog posts on the subject. In so doing, they draw on a large body of research published by some of the smartest and most respected scholars in the field.
There is much more in Cohn’s piece and I encourage you to read it in full. Trust me, the great Medicaid debates are not over. If you want to keep up, read Cohn. I’ll handle the science, he and others should (and I hope will) stay on top of the policy developments.