This is the last of four posts on the idea of “relative value health insurance” that I’ve been promoting. (For my other posts on this issue, see here, here, and here). To recap, the idea behind RVHI is for a government-sponsored organization to rate medical treatments on a scale of 1-10, based on their relative cost effectiveness. This would enable private insurers to offer, and consumers to purchase, health insurance that would cover only medical interventions that satisfy a particular cost-benefit standard. Currently, unless a particular treatment is explicitly excluded by contract, private health insurance generally covers all treatments with an expected medical benefit, without regard to how costly the treatment or how substantial the expected benefit. This ensures that health care costs will continue to rise as new technology is developed, whether or not the benefits of technological advances justify the costs.
In this post, I want to compare RVHI to the two most common types of proposals for fighting heath care cost inflation: consumer directed health care (“CDHC”), and pay-for-performance.
The idea behind CDCH proposals, long a darling of the political right, is to substantially increase the cost- sharing requirements of health insurance. The Health Savings Account (“HSA”) program, instituted as part of the 2003 Medicare Modernization Act, is based on CDHC principles. The program permits individuals who purchase health insurance policies with high deductibles to established tax-advantaged savings accounts that can be used to pay the out-of-pocket costs. The idea behind the HSA program, and CDHC approaches more generally, is that consumers will have a greater incentive to balance the costs and benefits of interventions if they are directly responsible for a larger portion of the cost.
CDHC is undoubtedly a good way to encourage greater price competition in the healthcare marketplace, but it is unlikely to solve the problem of consumers demanding expensive new technologies even when the costs do not justify the benefits. CDHC has two problems in this regard.
First, the theory of CDHC assumes that consumers are hyper-rational actors who can make optimal cost-benefit trade-offs at the point of purchase of medical treatments, when they are often ill, scared and facing extremely complex alternatives. The vast literature in the field of behavioral decision-making suggest this is exactly the type of situation in which individuals would have the most difficulty making efficient choices. Empirical research on health care decision-making suggests that greater cost-sharing causes consumers to cut back on medical care usage, but that consumers are not very good at determining which interventions are cost effective and which are not. Better to encourage patients to choose generally how to allocate resources between health care and other goods and services ex ante, through their insurance purchase decision, as RVHI would do. This lets individuals make global resource allocation decisions consistent with their preferences, but leaves the very difficult determination of the whether any particular treatment satisfies a particular cost-effectiveness standard to the expert panels that provide the relative value ratings.
Second, health care costs are notably bi-polar, with five percent of the population accounting for about half of all costs in any given year. This means that, even with very high deductible levels, most heavy users of medical care in any given year will exceed their out-of-pocket maximums, or know that they inevitably will before the year is up, and thus have no more incentive to equate costs and benefits of care than do individuals with first dollar insurance coverage. The RVHI approach avoids this problem as well by pushing the critical choice of how much in the way of resources to allocate to medical care back to the insurance purchasing decision rather than to the specific treatment decision.
A second popular theory about how to control the rising cost of health care is to reform the way physicians and other providers are compensated, moving from a system of payment for services (which creates its own moral hazard problem), to a system of “pay-for-performance,” or pay based on the health outcomes the physician obtains. Assuming providers would be paid based on the health outcomes of their patients net of treatment costs incurred, this approach would seem to incentivize cost-effective care while discincentivizing interventions with little benefit compared to its cost.
The problem with pay-for-performance is that it would create a conflict of interest between physicians and their patients. Patients who pay for insurance that promises to cover all “medically necessary” care rightly expect their physician to recommend treatments with a positive expected medical benefit, regardless of cost. In this situation, providers who refuse to recommend, or to provide, treatments with an expected benefit because they are not cost effective would breach their moral, and possibly their legal, obligations to their patients. Furthermore, patients would seek out the physicians with a reputation for recommended all medically effective care and avoid those with a reputation for recommending only cost-effective care.
In a world of RVHI, this conflict of interest would not exist. Physicians could recommend to their patients any treatments they believe would have a medical benefit, regardless of cost. Such physician recommendations would then be mediated by patient preferences concerning the allocation of resources between medical care and other goods and services. Whether a particular patient would follow the physician’s recommendation would depend, at least in part, on how “deep” and expensive the health insurance policy purchased. If a physician believed that the best treatment option, for a purely medical perspective, is one that has a very high cost and relatively modest expected benefit, a patient who paid more for a deep insurance policy might well find that treatment option is fully covered, while a patient who chose to pay much for a shallower insurance policy might find that the treatment option is not covered, or requires a substantially higher copayment.