On Monday, I posted about how new hepatitis C drugs priced as high as $1,000 per pill are a symptom of a fundamental problem with our health-care system. When most private insurance has relatively low deductibles and co-payments and covers all “medically necessary” treatments without regard to their cost (except for items that are specifically excluded by contract), patients have an incentive to demand all treatments with a positive expected benefit, no matter how small the benefit and how high the cost. As medical technology continues to improve and becomes more expensive, this “moral hazard” problem becomes more and more severe, and the price of health care spirals ever higher. The benefits might be worth the increased costs, but they might not be.
My proposal to confront this problem, which I’ve articulated in academic journal articles here (Mich. L. Rev.) and here (JHPPL), is to promote what I call “relative value health insurance” (“RVHI”). The basic idea of RVHI is that insurance policies should be available that would cover medical care that satisfies a higher or lower cost-effectiveness standard, thus enabling customers to determine, through their purchasing decisions, just how much of their resources they wish to devote to health insurance as opposed to other goods and services.
One example I use to illustrate how RVHI would work is the new drug Procysbi, which treats juvenile kidney disease more conveniently and with fewer serious side effects than the current treatment but, according an article in the New York Times, retails for $250,000 per year — far more than the $8,000 per year cost of the current treatment.
It is impossible to say in any objective way whether the benefits of Procysbi are “worth” the addition cost. It depends on the value each individual places on the marginal benefits of the drug. RVHI would enable customers to determine (before become ill) whether they wish to purchase what I call a “deep” insurance policy, which would cover treatments that offer even relatively little marginal health benefit compared to its cost, or a “shallow” insurance policy, which would only cover treatment options that offer substantial expected benefits relative to their cost. A deep policy would cover Procysbi, and other medical treatments with a similar cost-effectiveness profile. A shallow policy would cover the established treatment, but not Procysbi. Deeper policies would cost more, of course, than shallower policies. The market would determine the precise difference in price.
How would customers decide what depth of policy they would prefer? After all, before the customer or a dependent contracted kidney diseases, they would have little reason to even know what treatments are available for different conditions.
My proposal is for the government to provide ratings of different available treatments for different medical conditions on a 1-10 scale based on their relative cost-effectiveness. The most cost-effective treatments would earn a score of “1,” and treatments that provide very marginal benefits at a high cost would earn a score of “10.” The benefit scores would be based on QALY (quality adjusted life year) ratings; the cost scores would be based on retail charged for the drugs, devices, or services in question. Notice that these ratings would in no way prescribe what level of coverage could be bought or sold; it would only provide a relative measure of what medical treatments are more or less cost effective. Insurance companies then could sell policies that cover treatments rated “3 or better,” “5 or better,” “9 or better,” etc.
Even though few healthy customers would investigate the rating of a specific drug like Procysbi at the time of purchase, they could glance at lists of relative value ratings and obtain a general sense of what they would be sacrificing, in terms of coverage, if they chose to buy a shallow insurance policy that covers only treatments rated, say, 3 or better, rather than a relatively deep policy that covers treatments rated, say, 9 or better. Those customers could then compare the prices of policies of various depths and determine what depth of coverage they wish to buy. Some customers would choose to spend more money for deeper coverage; others would choose to save money by purchasing shallower coverage and have more resources to devote to other valuable goods and services. If customers prove too risk averse to go without any coverage at all for treatments with ratings that suggest they are not very cost effective, insurance companies might sell policies that require higher co-payments for treatments ratings below a particular cut-off point rather than providing no coverage at all for such treatments.
A system of RVHI would mean that patients would sometimes find, after becoming ill, that their insurance would not cover (or only provide limited coverage for) some potentially efficacious treatments when those treatments fail to satisfy the cost-effectiveness cut-off point specified by the insurance policy. But compare this admittedly unfortunate consequence to the problem with the current system: everyone with private insurance is effectively required to buy extremely deep insurance coverage, even when they might wish to spend their marginal dollars on other goods and services instead. A consequence is ever-increasing costs of medical care. The system works tolerably well for the very wealthy, the very anxious, and the very risk averse, but not as well for the rest of the insured population that might prefer to consume relatively less health care and relatively more of other goods and services. The availability of RVHI would not force anyone to purchase shallower health insurance than they have now; it would just afford the opportunity to do so.
If RVHI would provide valuable options to consumers, why has the market not already offered it? The simple answer is that to facilitate such a market, some entity must provide the relative value ratings, which private parties would then use as the basis for their insurance contracts. Because we lack good information about the effectiveness of most medical treatments, creating the necessary information is a massive undertaking, and no single private entity is likely to have the incentive to create such an expensive public good. The government needs to provide the information necessary to facilitate the market.
If you’ve read this far, you have probably identified several problems with RVHI. (If you haven’t, I’m sure the comments to this post will do so in the near future!) In future posts, I’ll try to describe some of these problems and explain why I think that they are not fatal to the proposal. I will also explain why RVHI, with all its flaws, is a better alternative than the leading proposals to control moral hazard in health care, such as “consumer-directed health care” and “pay for performance” for health-care providers. I’ll also discuss how RVHI can fit with current law under the Affordable Care Act. If you don’t want to wait for the blog posts, feel free to read more here (Mich. L. Rev.) or here (JHPPL).