A post by M.S. on The Economist’s website makes some of the same points I’ve made about the fact that we shouldn’t expect to save a lot of money by squeezing health insurers or increasing competition in that market. But M.S. devotes considerable attention to the profit and administrative costs associated with providers, which is not something I’ve explicitly addressed.
If M.S. is reading the literature correctly (and if that literature is itself correct), then provider profit and administrative costs are higher than those of the insurance industry. M.S. quotes the Physicians for a National Health Program (PNHP),
The estimate that total administrative costs consume 31% of U.S. health spending is from research by Drs. David Himmelstein and Steffie Woolhandler and published in the New England Journal of Medicine in 2003. The figure would undoubtedly be higher today. Insurance overhead accounts for a minority of the overhead. Much more occurs in physicians’ offices, hospitals, and nursing homes—driven by our current fragmented payment system.
Sensibly, M.S. is looking for confirmation of PNHP’s assessment of administrative costs in the health care system and its allocation to insurers and providers. He hasn’t been able to find anything, and he isn’t sure he buys the 31% figure or the notion that most of it can be attributed to providers.
And there are a lot of grounds on which you might argue that the Himmelstein-Woolhandler figure of 31% administrative costs is exaggerated. You might critique their decision to allocate one-third of doctors’ office rent as an administrative cost. Are American doctors’ offices commensurately larger than Canadian ones? Are physicians’ self-reports of time spent on administrative tasks accurate? But the curious thing is, I’ve hunted around for critiques of the Himmelstein-Woolhandler numbers, and I can’t seem to find any. I also can’t seem to find any alternative studies that also tried to measure all of the administration costs incurred by providers, to get a sense of how much the fractured private insurance system really costs.
Note there are two issues here. One is the size of U.S. providers’ administrative costs. The other is that size relative to that of a single payer or national health care system (e.g. Canada’s). At the moment I’m more interested in the former than the latter. We’re not going to a Canada-style system anytime soon. But perhaps other more politically feasible reforms could reduce provider overhead. How big is that overhead and what are its components?
In fact M.S. contacted me before publishing his/her post looking for some other evidence, papers, or reports on this topic. I’m not aware of any. But maybe you are. If so, please let me know.
Later: A reader suggests that the 1992 Health Affairs paper by Danzon serves as a response to Himmelstein and Woolhandler. Clearly it isn’t a direct response to their 2003 paper. But it does cite earlier work by Himmelstein and Woolhandler that may be similar or use similar methodology and assumptions (I’m speculating). I gave the Danzon paper a quick skim (so take the following is my initial impression and not necessarily my final opinion). It seems to me that it suggests that U.S. provider overhead is greater than insurer overhead. So, while it may differ from Himmelstein and Woolhandler on some points, it might also corroborate what M.S. was seeking to confirm.
And later still: Another reader suggests taking a look at the response article by Henry Aaron in the same NEJM issue as the Himmelstein and Woolhandler paper cited above.