The Anti-Empirical Turn

Senator Ron Johnson held a public hearing  and issued an accompanying report yesterday on unintended consequences of Medicaid expansion on the opioid crisis. You can read discussions of the event ranging from Vox’s condescending dismissal, to ThinkProgress’s righteous indignation, to Newsweek and the LA Times‘s tendentious rebuttals, to a fairly sympathetic writeup from USA Today, and a standard he-said she-said from the Washington Post.

Since I believe, I think without much delusion, that I started this ball rolling last March   and have kept it rolling pretty well since then, I obviously don’t have a totally dispassionate view. Even so, I recognize there are multiple, fairly   strong lines of argument against Medicaid playing a significant causal role in the divergence between expansion and non-expansion states. However, the basic fact is that there is a very large divergence to explain.


2014 to2016 change


None of the critical write-ups of the issue have included these basic statistics as a factor to explain or overcome. The Vox write-up by German Lopez calls the correlation between Medicaid expansion status & greater increases in overdoses since Obamacare passage a “Republican claim,”but the association itself is a simple fact. The question is if that correlation is due to the ACA or something else.


Look, I’m just a pseudonymous coward, and as I’ve said before, I’m not a public health expert or an expert in addiction and recovery or pharmaceuticals. I very well may be wrong in arguing that the Medicaid expansion had a large, causal effect on overdoses. However, the failure of a single academic public health expert to make a good faith effort to grapple with the existence of a very large divergence as a plausible effect of the ACA expansion and recognize it as an important empirical puzzle rather than an easily dismissed partisan claim is deeply disturbing to me. The ability of almost every single high-status news organization similarly to treat this as just “another case of those wacky Republicans grasping at straws” rather than a serious hypothesis with multiple competing explanations is equally depressing. The anti-empirical turn in America is real.


8 thoughts on “The Anti-Empirical Turn

  1. I assume you saw today’s NYT Upshot . “Counties in states with historically higher levels of uninsurance (and therefore greater subsequent growth in Medicaid) had lower growth in drug-related death rates from 2010 to 2015. This relationship holds within expansion and nonexpansion states separately.”

    Liked by 1 person

  2. I buy the correlation, but I question the direct link between the law and death.

    Perhaps the increase in opioid deaths has more to do with access to the medical community for treatments? If I get some insurance benefit, I go to the doctor for pain. If not, I eat Tylenol and deal with it. If I go to the doc in this timeframe, I get opioids.

    So, if the above is plausible, Obamacare is responsible for people going to the doctor, right? And in this timeframe, pain treatment at docs means hydrocodone.

    There seems some suggestion of this in the H-CUP graph: in 2008, Medicaid and Private insurance opioid related hospital utilization is about the same. PI and Medicaid docs may be treating a relatively homogenous population..??? Post 2008, Medicaid opioid related hospital utilization rises and equals the 2010 opioid related hospitalization rate among the uninsured, uncorrected for population. More Medicaid, fewer uninsured, but increasing opioid related hospitalization rates for the entire population, spiking in 2014. Is opiod-related hospitalization rate related to opioid deaths?…sure, makes sense..

    Anyway, I think the data indicates more access to healthcare and prescriptions among the target populations generated more opioid deaths.

    Similar to, but different than, an assertion that ACA led to death. There were more hospitalizations for penicillin allergies in the 1960’s than the 1930’s, but penicillin may still be a good thing…


    Thanks for your blog, and your analyses of things. I do share your alarm at the anti-empirical.

    Liked by 1 person

    1. These are interesting ideas, but I think it’s possible that the increase is driven by a small subset of doctors and patients who engage in a lot of deliberate fraud. Ron Johnson’s Senate report had a lot of examples, as did a recent WashPost report of two pharmacies across the street from each other in a tiny town in WV that were moving tens of millions of opioids a year.


    1. It doesn’t (I discuss in another post, “Off-Script.”) That’s a reasonable counterargument (tho I don’t think dispositive) but it is quite distinct from arguing that the correlation itself doesn’t exist.


Leave a Reply

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

You are commenting using your 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