On the Oregon Medicaid study: “Keep calm and collect more data”

MedicaidLogoThe New England Journal of Medicine sent out a press release Wednesday evening with the click-bait title of “The Oregon Experiment — Effects of Medicaid on Clinical Outcomes” (abstract here, the full paper is gated). Some background: because of a lack of funding the state of Oregon conducted a lottery system for Medicaid enrollment, and the ensuing results are being studied as a rare randomized controlled trial (RCT). The lottery has allowed healthcare researchers to, essentially, study the effects of Medicaid coverage in a way that people can begin to talk in terms of empirical causality. The first results (see here & here) were promising; for those who ‘won’ the lottery Medicaid appeared to provide increased health benefits and greater income security. This second result, however, is a little bit more mixed:

Let’s review. The good: Medicaid improved rates of diagnosis of depression, increased the use of preventive services, and improved the financial outlook for enrollees. The bad: It did not significantly effect the A1C levels of people with diabetes or levels of hypertension or cholesterol.

There is much, much more on this from folks smarter and more able to suss out the details behind this study than I can. For a wonderfully BS-free version go see Matt Yglesias’ take. I’ve read them all and so should you. For what it’s worth, on matters such as these, at the very least you will should be reading the folks at The Incidental Economist (where the above quote comes from). They’re serious, fair, and do this sort of thing for a living.

Unfortunately it was also an opportunity for some conservatives to, what in some cases I can only describe as gloatingly, declare everything they’ve ever argued negatively about Medicaid as validated. That is, yet another opportunity for them to delegitimize the program. I won’t bother linking to those simply because their arguments are (to be polite) disingenuous at best, or (to be impolite) a massive dollop of conservative derp at worst. For a more level-headed conservative reaction, here’s Ross Douthat:

Obviously the policy questions are vexing and complicated, and no single study is likely to dislodge people’s commitments no matter the state of the debate. But in that (hypothetical) world, the Oregon study’s evidence that Medicaid helps people financially but doesn’t do that much for health outcomes seems like it makes a much stronger case for the (hypothetical) Republican health care alternative than for the current Medicaid-driven, comprehensive-coverage-seeking Democratic approach. After all, if insurance is mostly beneficial because it shields low-income people “from medically induced financial catastrophe,” then why not make that the policy goal, rather than trying to expand Medicaid’s deeply flawed form of single-payer from the bottom up … ?

You could also read Tyler Cowen, I suppose. Their both level-headed, yet flawed. The ideological debate over Medicaid and the expansion as a part of the Affordable Care Act mirrors other public policy debates; insomuch as it contains an Inception-level mountain of preexisting bias, assumptions and deceptions that would make even Christopher Nolan say “Whoa, that’s too much.” Like any other aspect of arguing politics and policy, it often seems like it’s not worth it. At my most idealistic I can only think to describe it as “fighting the good fight,” or at the very least it’s the act of not letting others define the things you’ll care enough about to advocate.

So why bother? I don’t know. I’ve spent many posts in the past few years dealing with these same issues, so maybe it’s just another step along the way in this personal journey of interest. Here are my simplified thoughts:

  • Above all else, this was a study — to be sure a very good one with high internal validity — covering 12,229 people who had Medicaid for 17 months in one city in America. It should go without saying that one study an entire case about providing millions of low-income people should not make. This is a point that’s escaping many who should know better.
  • There are some things this study definitely does not prove — that Medicaid is worse for you than nothing at all, and that having Medicaid produces no health benefits. If you start from that point then most detractors have little substance to add and the headlines are especially pernicious.
  • Medicaid is an insurance program. It enables (comparatively) affordable access to our healthcare system and protects against catastrophic costs. In the latter way it’s much like any other insurance scheme, and according to this study works quite well.
  • One common response, then, from the left to criticisms has roughly been “If you’re so confident that Medicaid as health insurance has no positive health benefits then why bother keeping your own?” On the surface this seems trite, but it has substance if (some of) the negligible statistical effects on the conditions examined in the study are more a result of the basic structure of health insurance in general. The onus is on detractors to provide evidence that this is especially a result of Medicaid. They cannot, for reasons I’ve discussed before, so the response strikes me as valid.
  • Contra many of the conservative arguments I read, the debate isn’t over Medicaid (and the expansion) and their idealized alternatives in 2014. Ostensibly it’s over Medicaid and nothing else (or at least no expansion). There is no serious movement by Republican elected officials to work out a compromise that brings conservative healthcare wonk proposals to the table. Medicaid exists now. The expansion exists, in law, right now.

There is one other line of thought for which the subject matter is troubling for me to think on, brought up by Jeff Spross yesterday (in fact, just go read his succinct tweets starting here), on the morality of implicitly discussing whether Medicaid is “worth it:”

Sorry, but I find chin stroking about whether we *really* need to spend all this money on HC coverage for the poor, by people who have great coverage through their employers and aren’t poor, to be pretty morally repugnant.

Like so many pressing policy debates today, Medicaid is at its root about public spending on the poor. If I was explaining this debate to my four year old I would describe it as “Some want more, others want less.” Yet within the adult realm this conversation should include some verisimilitude of humility, of contextual self-awareness that goes beyond recognizing the socioeconomic status of the poor, but of yourself as well.

(Hat tip to Andrianna McIntyre from Project Millennial for the quoted portion of this post’s headline. Also go read her roundup and comments in today’s Wonkbites)


One response to “On the Oregon Medicaid study: “Keep calm and collect more data”

  1. Pingback: Douthat and generalizing the OHIE | Punditocracy·

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s