David Cutler and Nikhil Sahni on the health care cost curve

Technological change has also been a key driver of spending increases for some time. From pharmaceuticals to imaging to cardiac procedures, markets have been saturated with new and expensive services and products in recent decades. But the adoption of new technology seems to have slowed. Major parts of imaging growth are down, some cardiac procedures are being performed less frequently as studies show they are overused, and the number of new molecular entities approved by the Food and Drug Administration has not kept up with research and development spending.

To be sure, there are many new drugs and imaging devices on the market, especially in fields like oncology. But sales of these new technologies have been more disappointing than robust. The therapeutic prostate cancer vaccine Provenge was not the hit it was expected to be; Zaltrap, a therapy for some cancers and macular degeneration, had to halve its price because it was losing out to Avastin.

Efficiency efforts are finally taking hold in the health care community. Recent news reports about delivery system changes in large health care organizations, declining rates of hospital-acquired infections, and new emphasis on reducing readmissions are indicative of changes going on across the country. These efforts have been facilitated by the ACA and state efforts to limit Medicaid, total health care spending, or both (as in Arkansas, Massachusetts, and Oregon).

The first part is “the great stagnation to the rescue,” the second part is good news, and in the third part the reference to Massachusetts is some kind of Straussian satire.  The article itself is here, and contains further analysis.

The initial pointer is from @JustinWolfers.


If only it was Straussian satire. Tyler... that is how people actually think.

It's certainly good news. But I fail to see any evidence this period of slower than expected growth in health care costs isn't directly related to the recession and a post-recession sluggish economy. There seems to be more wishful thinking in that post than hard evidence.

How can imaging be a major driver of healthcare cost? CAT/MRI/PET scanners are just widgets, largely electronic widgets at that, and widgets, especially electronic widgets, are supposed to be the high productivity growth sectors of the economy--unlike healthcare. And, as Tyler has observed before, more or less the same goes for new drugs. Drugs are super-cheap to supply per-unit compared to surgery or practitioner time, so if they have any measurable effect they're likely to reduce rather than increase costs.

So which is it--is the Baumol effect driving healthcare costs or something else entirely? If it's not the Baumol effect, this supports efforts to make U.S. healtcare policy more like Europe's, because we are certainly a lower-cost producer of most electronic widgets (and if we're not, we can just buy them from China like they do--widgets are tradeables, hence law of one price applies).

The article states that scans and drugs are major drivers of spending, not cost. Hospitals make big money by grossly overcharging for both of those. Related (and published yesterday): http://healthland.time.com/2013/02/20/bitter-pill-why-medical-bills-are-killing-us/print/

"How can imaging be a major driver of healthcare cost? CAT/MRI/PET scanners are just widgets, largely electronic widgets at that, and widgets, especially electronic widgets, are supposed to be the high productivity growth sectors of the economy–unlike healthcare."

...because healthcare productivity is measured, idiotically. Most health care economists (I'm exempting Alex here) are pretty stupid about this.
If they can find and or treat a disease that wasn't treatable before, it counts as increasing cost. If one year a disease kills you in 6 months and is untreatable, and the next year its treatable and you die of old age... that counts as cost increase.... not a productivity increase.

Its idiotic.

Well, it is a cost increase, because an untreatable disease that kills you doesn't cost money, while the treatment does (and also allows you to live longer and accumulate further healthcare costs).

I understand your point; the benefits of more and better care shouldn't be ignored. But when the discussion is "the amount of money we are spending on healthcare" then anything that costs money is a cost. There's nothing idiotic about that.

The terms of the discussion are idiotic and more people need to say this. Wouldn't it be better under current premises to just declare all treatments developed before say 1948? That would really lower costs. 1848 would be even better.

It is only good news if you aren't sick.


Even more to the point, stagnation is only good news if you don't plan to get sick in the future.

How is fewer useless treatment and technical "innovations" coming to market evidence of stagnation?


Quality of life is the issue with PCI (stents). If you have a choice between open heart bypass surgery and angioplasty, most would opt for the latter. That said, diet etc is important: http://www.nytimes.com/2004/03/21/us/new-heart-studies-question-the-value-of-opening-arteries.html (But, researchers say, most heart attacks do not occur because an artery is narrowed by plaque. Instead, they say, heart attacks occur when an area of plaque bursts, a clot forms over the area and blood flow is abruptly blocked. In 75 to 80 percent of cases, the plaque that erupts was not obstructing an artery and would not be stented or bypassed. The dangerous plaque is soft and fragile, produces no symptoms and would not be seen as an obstruction to blood flow.)

The recent Freedom trial showed that if you are diabetic, you would be better off opting for the open heart surgery.


John Goodman's book Priceless is itself priceless in its multitude of insights about what ails and what can be done about our health care system. In 2014 ACA is going to rock our world far more than it already has and not in a good way.

Technologically we are on the verge of a transformation to gene-based, individualized medicines, tissue/body part synthesis via additive manufacturing and in-body digital systems. Should be pretty amazing.

Hospitals charge for things like McDonald's. Foods are cheap, drinks are expensive.

Hospital drugs and rooms are expensive, procedures are cheap.

Here is a chart to make it easier to visualize the rising expenditure on healthcare 1989-2011:


In Massachusetts there is a huge problem with Partners Healthcare, a health system and hospital partnership, that has gained control of most of the major care outlets in the state. Insurers are nearly forced to pay whatever they charge. (Imagine: "Why, no, our insurance doesn't cover care at Mass General, Brigham and Women's or 80% of physician groups in Boston...") In a relatively small state where healthcare is a large industry, this is a major driver of insurance premiums, but such anti-competitive consolidations are happening all over the country. See Avik Roy's post on the subject: http://www.forbes.com/sites/aroy/2011/08/22/hospital-monopolies-the-biggest-driver-of-health-costs-that-nobody-talks-about/

Also, Massachusetts' health cost control legislation was only signed into law late last year. It is therefore unlikely that those measures are having any real impact on costs yet. MA explicitly put off cost containment when it passed its health care expansion, while the ACA explicitly included cost saving measures. So we have yet to see how it will shake out. Though I think a cartel of providers will continue to be the primary driver of costs if something isn't done about it.

Comments for this post are closed