In early April on The Incidental Economist (TIE), I posted an analogy to illustrate reference pricing.
Let’s make a deal. You shell out just $500 and I’ll pick up the tab for any automobile you care to buy. I’d better protect myself a little, so the deal is only good for Honda, Toyota, and Hyundai. Are you going to get the base model Yaris ($14k)? I seriously doubt it. You’ll probably get something nicer, maybe a souped up Land Cruiser ($80k+). I’m going to pay an awful lot.
Having learned my lesson, let me make a smarter deal. I’ll give you $15k toward a car, any car on the planet. If you want something more expensive, you pay the difference.
Which deal will lead to more prudent shopping, less wasteful car spending? Which will incentivize the market to be more efficient and consumer friendly?
The first deal above is akin to copayments in health care. To be sure, copayments can dissuade patients from seeking care in the first place (buying a car in the analogy), but once they do, the payer (insurer) is at financial risk for the balance.
The second deal above is (or can be) a version of reference pricing. The payer’s liability is capped (at $15k in the example). If the value of that cap is set according (in reference) to the price for one version of the product, it’s a reference price. Though I didn’t say so, I deliberately set the price so that it was just above that of a Yaris. The implication is that everyone taking the deal would have the means to buy at least some type of decent-quality car. (I am stipulating that a Yaris is of decent quality, but we can differ on that and adjust the reference price accordingly.)
The post was titled provocatively as “Copayments are stupid.” Perhaps for that reason it garnered a considerable number of comments: 53 as of this writing, which is a lot for my posts, though a pittance compared to some of Aaron’s. Most were very good, as is typical for TIE, causing me to think more deeply about the subject and to do some searching through the literature. First, let’s consider some critiques from readers:
Readers reminded me that people can’t shop for health care as easily as they do for other products like cars. The market is opaque, with poor access to information on prices and quality. The products and services are complicated, requiring more years of training to understand than even car repair. Patients are at an informational disadvantage relative to providers. Some are incapacitated or have low health literacy.
Moreover, readers instructed me that, for cars, higher prices generally accompany more amenities that people can see and appreciate (a nicer stereo, more room, leather seats). In health care, higher prices don’t necessarily correlate to quality. In other words, when people face a higher price for a car, they can usually assess why and whether that’s worth it to them. In health care, they can’t. Even if it comes with more tests, it is hard to know the value.
Finally, I was reminded that health care prices are high and unaffordable for many people.
All true. Analogies can be imperfect, and shopping for health care is not easy. And yet, these critiques lose their punch provided the reference price is sufficient for good quality care. That’s a big condition. Will Americans accept someone else deciding what’s sufficient quality?
Let me agree with my critics that patients cannot determine what good quality care is or whether a higher price corresponds to better value. If so, then one has two options: either accept that someone else needs to establish a floor on quality or that we have to give up on assessing quality altogether. If someone or some body can establish minimally accepted quality, and if the reference price is set according to that floor, everybody would have access to at least that.
One thing that was not raised in the comments was whether reference pricing reduces spending and/or quality by some standard. For this question, we can turn to a recent paper in Health Affairs by Sarah Thomson, Laura Schang and Michael Chernew.
A Cochrane review finds that reference pricing can induce a switch toward the use of less expensive drugs, with mixed evidence of cost shifting to other health care sectors and no evidence of adverse effects on health (mortality). These results have been confirmed by a more recent systematic review. [Hyperlinks added.]
These are both reviews of the application of reference pricing to prescription drugs. James Robinson and Kimberly MacPherson have described cost-lowering use of reference pricing variants by CalPERS for knee and hip replacement surgery and by Safeway for imaging and laboratory tests. Steven Pearson and Peter Bach have illustrated how reference pricing could reduce expenditures for Medicare-covered services. So, while reference pricing is an atypical feature of current-day insurance plans, it’s hardly unknown or untested.