We health wonks have been accused (and in some cases convicted) of endless arguments similar to the oft-cited “how-many-angels-can-dance-on-the-head-of-a-pin”.
But every so often an intellectual debate erupts over a core concept, a foundational understanding that underlies what passes for generally-accepted wisdom in the health care world.
One recent example is the ‘does making people share cost reduce unnecessary utilization?’ question, with the latest foundation-shaking brought on by an article questioning the seminal study on the issue, the RAND Health Insurance Experiment.
Fortunately, we can turn to an objective economist who actually understands both health care and the awkward science (my term of endearment for econ) to referee the verbal volleying – Jason Shafrin. Jason has posted his thoughts on the controversy over the RAND HIE, and notes that, despite some methodological concerns, the central takeaway is still valid – people who have to pay for some portion of their health care tend to use less health care than those who get it for free.
OK, so that’s settled.
The next question is the tough one, and it is not how much care we are getting, it is are we getting the care we should?
And on that subject, the RAND study is silent.
Nov
15
The RAND study drew two conclusions, as I understand it. Conclusion #1 was that people use less health services when incented to do so.
Nyman’s challenge is primarily directed at conclusion #2 – that this reduced use of medical services has no long term adverse health impact.
I believe that Jason’s post makes a convincing argument regarding Conclusion #1, but fails to do so for Conclusion #2. Thus, I feel he has failed to refute Nyman’s primary point.
More to follow, if I can find the time.
Actually, the key take-aways from the RAND HIE are stated right on its first page.
You can get it in HTML or PDF for yourself here:
http://www.rand.org/pubs/reprints/RP1114/index.html
Richard has Conclusion #1 right, but the RAND researchers’ Conclusion #2 was that cost-sharing made people reduce the use of both unnecessary and necessary medical treatments in equal amounts.
The RAND researchers’ Conclusion #3 was that reduced use of medical services had no long-term adverse health impact, with a few key exceptions. They noted free care led to improvements in hypertension, vision, dental, and something they called “selected serious symptoms”. And also that the improvements were concentrated most among the sickest and poorest patients.
The recent questions about the methodology of the RAND HIE are valid, but even in a worst-case scenario (i.e. those patients who dropped out of the study ALL would have worked against the eventual findings), were not statistically significant enough to change the core conclusions.