The New York Times-Upshot ran a critical article about the Obamacare-Opioid connection, and were nice enough to link to my first post in the series (“Obamacare Didn’t Help.”) Since I’ve done a lot more research on the issue since then, I’d like to summarize some of the posts here, especially for New York Times readers who would like my best argument:
a) “Some Stronger Evidence that the Affordable Care Act Worsened the Opioid Crisis“: This presents my main OLS regression and propensity score matching evidence that, conditioning on characteristics of counties in 2010, counties in states that fully implemented the Affordable Care Act had larger increases in overdoses than counties in states that did not. Data and code to reproduce my findings are also here.
b) “Good and Bad Arguments Against the Obamacare-Opiate Effect“: This is a direct response to the Goodman-Bacon and Sandoe Health Affairs post that the New York Times bases their rebuttal on.
c) “The (Older) Kids and their Parents Are Not Alright“: This looks at the divergence across age groups between Medicaid Expansion and non-Expansion States.
d) “It’s Opioids, not ‘Dying from Despair,’ that’s driving the Obamacare Effect.” This looks at suicides and alcohol-related deaths to show that the increase in deaths in Medicaid expansion states relative to holdouts is not due to a generalized social collapse but is specific to overdoses.
e) “Obesity is important for mortality, but it’s not the reason for the Obamacare Effect“: This is a similar post that uses obesity data to argue that the Obamacare effect is not due to worse general health in Medicaid expansion states.
f) “They’re Bringing Drugs“: This is my review of Sam Quinones’s Dreamland, about the pre-ACA opioid crisis. One of the ways the Obamacare hypothesis has been misrepresented is to say that it is arguing Obamacare was responsible for the crisis as a whole, which is clearly false: the opioid crisis has been a worsening disaster for two decades now. When I refer to the Obamacare effect, I’m referring to the marginal increase in Medicaid expansion states over what occurred in holdouts, which is statistically significant and substantively important but is clearly not the only story, given that this is an epidemic that is on track to kill more Americans than World War II.
g) “Regional Differences in Personality and the Obamacare Effect” and “Off-Script“: These are looking at the two stronger arguments against the Obamacare effect- that it is a regional phenomenon and that prescriptions do not appear closely tied to overdoses and in some cases seem to go in opposite directions. I recommend that future journalists who want to rebut the Obamacare hypothesis overall rely more on these arguments rather than the silly idea that legislation that was passed in 2010 could not have affected insurance and overdose rates from 2010 through 2013.
h) “Marriage and the Obamacare Effect“: This was an earlier attempt at a summary post that also looked at whether divergences in marriage rates (or related social/cultural divergences between Medicaid expansion states and holdouts) could be responsible for the Obamacare effect.
i) “Incentives and the Iron Law“: About why HHS didn’t publish its findings until Ron Johnson requested them.
j) “Health Affairs Blog Confirms Health Insurance is Bad for You“- A more flippant response to the Goodman-Bacon and Sandoe piece, focusing on the negative correlation they identify between uninsurance rates in 2013 and increases in overdoses 2010-2015.
k) “One of these Things is Not Like the Other“: This is a more speculative piece about why a relatively small increase in insurance rates, once most people are already insured, could produce an outsized increase in overdoses.
m) “2016 County-Level Estimates of the Obamacare Overdose Effect“: This updates the county-level estimates in (a) above to use 2016 overdoses as the outcome.
n) “Every man is a piece of the continent“: This discusses how the Obamacare effect interacts with the association between overdoses and income and between overdoses and local academic achievement.
Disclaimer time: I am not a public health expert. I am a former school-teacher and present public policy researcher in areas not closely related to Medicaid or opioids. While I have done my best to get the analyses above right, I have only a superficial grasp of the broader research literature on addiction and recovery and only an “informed outsider’s” understanding of health insurance markets and health economics. Although I think it is likely that I (along with Gabriel Rossman) started this debate off last Spring (or at least changed the conversation from the ACA making the opioid crisis better to debating how much it was making it worse), I mostly would like to see more well-informed and technically competent people than myself do thorough investigations of the issue. Frankly, I have not seen that yet; the attempts to debunk the effect are generally speaking superficial and attempt to paint a he said, she said picture rather than dealing seriously with a serious issue.