Medicare Reform and Prescription Drugs: Ten Principles

Study | Health

Read Article as PDF
Get Adobe Reader
Bookmark and Share



No. 256

Wednesday, October 09, 2002

by John C. Goodman, Robert Goldberg, and Greg Scandlen

Principle No. 3: The Plan that Administers Drug Benefits Should Be the Same Plan that Administers Other Benefits.

Appreciating the opportunities to substitute between therapies also helps us understand why a unified health plan is so important

"The plan that controls access to drugs must be the same plan that controls access to other therapies."

A number of proposals would create an optional prescription drug benefit under Medicare. In most versions, seniors would pay an additional premium, and the benefit would be administered by a private pharmacy benefit manager (PBM) - a sort of managed care approach to prescription drugs. (See the discussion below.) In all versions, premiums would be the same for all - irrespective of preexisting conditions or expected health care costs.

One problem with this approach is that it encourages extreme adverse selection. Everyone whose expected drug costs exceed the premium being charged would quickly enroll, while those whose expected drug costs are below the premium would decline coverage and delay their enrollment. In other words, high-cost sick people would rush to join while low-cost healthy people demurred. The end result would be escalating deficits or, if the premium were intended to cover all costs, a death spiral in which the premium would rise so high that none but the most expensive enrollees would remain in the plan.

But even if this problem could be overcome, as some have attempted to do,17 a second problem is that if two health plans administered separate sets of benefits, the decision makers could avoid the full costs of their bad decisions and could not reap the full benefits of their good decisions. The result: perverse incentives to make bad decisions and avoid good ones. Suppose, for example, that one plan controls hospital benefits and another drug benefits. Given a fixed premium for each, the principal incentive of the first plan would be to reduce hospital costs. The principal incentive of the second plan would be to reduce drug costs. In neither case would the plan's administrator have an incentive to consider the overall picture and maximize the health benefits of a given money outlay.

There is also a third problem. Just as these proposals would distort incentives for plan administrators, so they would distort incentives for patients - as the next principle shows.