Last week I put up two posts (Health-care costs and the ‘moral hazard’ problem; Relative-value health insurance) about my idea of relative value health insurance (“RVHI”).  RVHI could help control health-care cost inflation by providing consumers with the opportunity to purchase insurance that would cover only medical treatments that satisfy a specified cost effectiveness standard rather than all treatment with any expected medical benefit regardless of the size of that benefit or its cost.  The government would facilitate a market by rating medical interventions provided for different conditions on a scale of 1 (most cost effective) to 10 (least cost effective).  This would allow private insurers to offer, and consumers to purchase, insurance coverage of different “depths” at different prices.  Today I’m going to discuss some of the most significant challenges to implementing such a system.

The most basic problem is that the information necessary to creating relative value ratings is virtually nonexistent.  There is no scientific evidence even for the basic effectiveness much of the medical care that is provided every day (even clinical practice guidelines, which seem evidence based, are often based on consensus opinion rather than hard data), much less evidence that compares benefits of treatments to their costs.  A huge investment in developing the necessary information would be needed.

The good news, however, as that the Affordable Care Act has at least created a framework for this investment.  The ACA created the Patient-Centered Outcomes Research Institute (“PCORI”), a private non-profit organization charged with coordinating comparative effectiveness research (“CER”) to compare the efficacy of treatments for different medical conditions.  The health-care law also provides substantial funding for CER.  By combining cost data with comparative effectiveness data, research institute could provide the relative value ratings needed to facilitate a RVHI market.  Although the health-care law prohibits the government from using the results of CER as the basis for insurance coverage mandates – quite clearly a response to fears of government “death panels” – there is nothing in the law that prohibits the government from providing cost effectiveness ratings that private parties could then use as the basis for insurance contracts.

Even with a structure in place, and assuming the massive funding commitment that would be needed to create relative value ratings, it would surely take many years to create anything like a complete set of relative value ratings.  This problem, in itself, is not a substantial impediment to RVHI.  Currently accepted treatments would have to be grandfathered into the system with a rating of “1” until relative value ratings could be established, while new treatments and technologies would have to demonstrate their relative value before earning a rating and qualifying to be covered by RVHI policies.  Thus, the concept of RVHI would be phased in over time.

A second practical problem with RVHI is the risk of adverse selection.  Specifically, if only sick people choose deeper policies (which cover less cost effective treatments) and the young and healthy choose cheaper, shallow policies, the selection effect might make deeper policies unsustainable, even for customers who prefer to spend a large chunk of their wealth on health insurance rather than other goods and services.  To fight this problem, the law would have to have a mechanism to allow insurers to prevent customers who purchase shallow policies from switching to deeper policies immediately after they are diagnosed with a serious illness.

A third problem, and one that cannot be satisfactorily solved, is that relative value ratings, although scientific, would necessarily embody certain contestable value choices.  Relative value ratings would be based on the quality adjusted life years (QALYs) per dollar expected from a given medical treatment.  But there is no single, agreed upon way to measure QALYs.  Imagine two treatments with the same cost, one of which provides greater physical comfort for the patient, and the other which preservers greater functional ability.  Which one earns a higher relative value rating?  There will be many clear cases, of course, but there is simply no way  to devise a relative value metric so that every patient would agree that a treatment that earns a score of “4” is actually more cost effective than an alternative treatment that earns a score of “5”.

For a more in-depth analysis of the obstacles to making relative value health insurance a reality, see my Michigan Law Review article on the subject here.

In my next (and final) post on this idea, I’ll explain why I think that despite these obstacles and others, RVHI is a more promising approach to controlling health-care cost inflation than the customary proposals of (1) making patients pay more money out of pocket for medical treatments or (2) changing the payment system to compensate doctors and other health-care providers based on the quality of outcomes.