The Eternal Problem
Have you read Alan Garber and Jon Skinner’s new NBER paper, “Is American Health Care Uniquely Inefficient?” The title is provocative enough, but the answer – yes – and the research that they critique – mainly Cutler’s work on innovation – make for a fabulous read.
You know the basic statistics. The U.S. outspends every nation on medical care by a huge amount. Even if you adjust for the recent weakness of the dollar, we still outspend Europe and Canada by 75 percent. A few decades ago, the gap was more like 20 percent. And what do we have to show for our profligate ways? Basically zilch. We have average life expectancy, average infant mortality, and above average discrepancies in health. If we measure efficiency by the bang we get for our health care bucks, then how can we disagree with Garber and Skinner’s conclusions?
To what do we owe this dubious distinction? Garber and Skinner mention high physician incomes, but would do well to include the high earnings of other professionals. I think it was Victor Fuchs who first pointed out that our medical professionals earn a lot more than their Canadian and European counterparts. We can’t realistically scale back the wages of our medical professionals, not without sending droves of talented young men and women into other careers. So let’s look elsewhere for something we can fix.
It is easy, but incorrect, to blame administrative costs (high, but not growing in percentage terms) or malpractice (ditto). What is left, of course, is technology. Expensive, new technology. Implantable defibrillators, transplants, hip replacements, chemo, you name it, the rate at which we use these technologies dwarfs just about anywhere else. If you want to cut U.S. health spending, you have to cut spending on technology.
And here is where Garber and Skinner take their shot at Cutler. Cutler’s seminal work on technology claims that the health benefits more than outweigh the costs. It is too easy to conclude that we should avoid scaling back on technology. But if Cutler is right, why don’t these benefits show up in our aggregate health statistics? Yes, we are living a lot longer than we used to, and these technologies seem to be responsible. But Canadians, Europeans and Asians are also living a lot longer, and they aren’t such rabid consumers of technology. Garber and Skinner try to pooh-pooh the role of technology, suggesting (tongue in cheek I hope) that exercise guru Richard Simmons has more to do with our improved health than all the cardiac surgeries and heart drugs. But that is going too far, as Garber and Skinner would surely acknowledge.
Could it be that new technology is responsible for much of our improved health, but also that it is possible to have too much of a good thing? (In economics parlance, the average benefit of technology exceeds cost, but the marginal benefit of the last unit of technology consumed does not.) New technology lures patients and their providers with the Siren’s promise of hope. And insured patients (and their providers who get paid more for doing more) cannot resist. And so we spend, spend, spend.
But what happens if tie ourselves to the mast of technology rationing, as they do elsewhere? (Obama has some inklings in this direction, despite receiving advice from Cutler.) I fear that we are in an odd sort of equilibrium. The rest of the world free rides on the innovations whose profits depend on the largesse of American consumers. If we ration technology, future innovations may disappear and the U.S. would be the biggest loser. So we don’t regulate and the rest of the world continues to enjoy the free ride.
Maybe we should threaten the rest of the world. Pony up more for technology so we don’t have to. And if you don’t, we graduate another generation of finance MBAs to screw up the global economy.
Any way you cut it, I’d agree it’s hard to tell a story that we are getting great value for money for our marginal healthcare dollars. This said, a couple of thoughts to consider. First, I am not sure we should simply accept high compensation rates as a necessary evil to assure a quality workforce. It is true American physicians are the highest paid in the world. But surveys typically indicate they also have among the lowest satisfaction levels. Some oft cited reasons are high administrative burdens imposed by our fragmented system of oversight and issues with malpractice. It seems likes there could be a trade-off—less pay in exchange for a less contentious and burdensome work environment. Add to this that we are strongly tilted towards high paid specialists versus lower compensated primary care physicians compared to other OECD nations and it seems like there could be some substantial savings.
This leads to a second point. It’s easy to cite our eager adoption of technology as a central driver of cost differentials. But the effects of technology are often measured as a residual after making adjustments for factors like inflation, demographic changes and growth in GDP etc. Simply saying we should ration technology does not provide a lot of guidance on an agenda. An alternative lens for looking at costs is by disease category. The CDC estimates chronic illness accounts for over 75% of healthcare costs in the U.S. Further, Ken Thorpe and colleagues (Health Affairs 2007) argue that there are substantially higher rates of prevalence and treatment of chronic disease in the U.S. compared to other OECD nations. Higher prevalence rates may explain both greater resource use and poorer outcomes. This also suggests why Cutler may find significant benefits from medical interventions without necessarily seeing an improvement in aggregate rates compared to other countries- we may just be playing catch up.
Of course, differences in prevalence rates could partly be an artifact of more aggressive practice styles (e.g. the old medical school adage that a healthy patient is one that has not been adequately studied). Higher treatment rates for diagnosed disease seem consistent with this. However, markedly higher rates of problems like obesity in the U.S. are consistent with a higher underlying level of morbidity. A takeaway which is hardly new, but perhaps more compelling in this context, is that in terms of getting the biggest bang for our marginal bucks, we may be best off focusing innovation and policy efforts on getting prevalence rates down versus addressing the consequences. This may be associated more with public health measures then technological wizardry. But I wouldn’t underestimate the role of new technology in this area if we turn our minds to it. Examples include enhanced IT systems and new informational, monitoring and feedback tools to support disease management and wellness programs. Meanwhile, an interesting natural experiment in progress is how growing rates of obesity etc. in the OECD and elsewhere may impact these nations’ health costs.
Finally, in terms of punishing technological free riding, turning another generation of finance MBAs lose on the global economy is a fearsome threat. But I’m having another sort of worry. With jobs drying up on Wall Street, what happens if instead they move over to healthcare?