Skip Navigation

August 10, 2001
 

Honorable Michael Bilirakis
Chairman
Subcommittee on Health
Committee on Energy and Commerce
U.S. House of Representatives
Washington, DC 20515

Dear Mr. Chairman:

During the Subcommittee's May 16, 2001, hearing on a Medicare prescription drug benefit, you asked me to explain the Congressional Budget Office's (CBO's) rationale for not adjusting its cost estimates to account for savings from fewer hospital admissions or less use of other health services resulting from improved access to outpatient prescription drugs.

We would expect some improvements in beneficiaries' health from expanding access to outpatient prescription drugs. Better health is, after all, one of the main reasons to enact such a benefit, and improvements in health should lead to savings in costs for hospital admissions and other services for some people. However, it is not clear that in total such costs would decline.

If the alternatives were "all" or "nothing"--either full access to the complete array of outpatient prescription drugs available today or no access to any outpatient drugs--the net savings in the costs of other health care services provided to Medicare beneficiaries might be easier to predict. Few Americans would disagree with the premise that, on the whole, the prescription drugs available today have not only contributed to declining mortality but have also helped keep people out of hospitals.

But our task is not as simple as estimating all-or-nothing options. Today, Medicare beneficiaries without any drug coverage already consume a large number of prescription drugs. (On average, they filled 17 prescriptions in 1997, compared with about 24 for Medicare beneficiaries with good employer-sponsored coverage.) The additional or more expensive drugs that beneficiaries without current coverage might use as a result of gaining coverage would probably provide less-dramatic improvements in their health than do the drugs that they are already taking. At the same time, greater use of drugs, especially in an older population, would increase the chances of side effects, allergic reactions, medication errors, and other adverse drug events--which could increase the use of hospitals, emergency rooms, and other health care services. To help limit the frequency of adverse drug events, physicians often require periodic visits and lab tests for people on chronic therapies, and those precautions also raise costs.

Surprisingly, research has not provided clear evidence about the balance between improved health for some people and more adverse events that might be expected from increased spending on prescription drugs under a Medicare drug benefit. Although recent evidence suggests that the net effect may be to lower the cost of other services, the studies are difficult to interpret, especially in the context of a Medicare drug benefit. More evidence is expected in the next few months from evaluations of state-level prescription drug programs for low-income elderly people. We are continuing to investigate both existing studies and new evidence as it emerges.

It is worth noting that if our continuing review of the evidence convinced us at some point that a drug benefit would, on balance, lead to savings in other health care costs, the magnitude of those savings would probably be quite small. Moreover, savings would accumulate gradually, over a number of years. For one thing, only about 30 percent of all Medicare beneficiaries are currently without prescription drug coverage. Those with Medicaid and employer-sponsored insurance already have better coverage than would probably be available through a Medicare drug benefit. And, as noted earlier, beneficiaries without drug coverage still consume prescription drugs, so the increased spending would be expected to result in only modest improvements in health. Finally, a large part of those improvements would probably delay rather than fully prevent the use of expensive health care services. Delaying the onset of disability or end-of-life care is unquestionably a good thing, but CBO's 10-year time frame for cost estimates would require us to recognize a large part of the delayed costs. Moreover, if a drug benefit helps people live longer, they may consume more health care over their remaining lifetime than they would without the benefit.

The details of a Medicare drug benefit's design might also affect our estimate. Proposals that include active management of costs might control utilization or steer patients to less expensive or older drugs. Such designs might result in lower savings in other costs than would designs with less control over patterns of drug use.

Sincerely,

Dan L. Crippen
Director
 

cc:	Honorable Sherrod Brown Ranking Member, Subcommittee on Health
	Honorable W. J. (Billy) Tauzin Chairman, Committee on Energy and Commerce
	Honorable John D. Dingell Ranking Member, Committee on Energy and Commerce