Strategies to Improve Medication Compliance by Medicare Beneficiaries
In the article in this issue by Choudhry et al,1 the authors assess whether eliminating out-of-pocket costs would be an effective way to improve medication compliance in Medicare patients after they have experienced an acute myocardial infarction (MI). The specific questions the authors raise are whether full coverage would be cost-effective using the traditional values for such measurements and whether full coverage would actually save Medicare money by reducing health expenditures outside of the pharmacy area. For those of us who live in the policy world of Washington, this is an important issue, because if the additional coverage can be demonstrated to save Medicare money elsewhere, the change in coverage would not be regarded as an increase in spending but rather as a savings to the program. Under current Congressional rules, increased expenditures—even useful ones—need to be offset by either reduced expenditures elsewhere or by increased tax revenue. This means that being able to demonstrate the net effect of a change in coverage on overall Medicare spending (as well as on overall health spending) becomes an important budget matter as well as an important health policy issue.
Article p 1261
The more general problem that the article by Choudhry et al1 addresses is how best to improve the notoriously low compliance rates regarding the use of proper medications, along with other preventive strategies in the secondary prevention of coronary artery disease. One such strategy is to eliminate all out-of pocket costs; however, other strategies are also available, such as chronic disease management, education, incentive programs, etc. This point is addressed in greater detail below.
The study1 is a carefully done empirical analysis, particularly given the limited amount of data on Medicare Part D outpatient prescription drug spending that is currently available. The study’s authors used a Markov model to estimate the incremental cost effectiveness and the net effect on spending of providing full coverage for appropriate combination drug therapy after acute MI compared with the estimated average coverage under Medicare Part D programs.
They found that full coverage of post-MI combined drug therapy saves lives and money by reducing nondrug health expenditures, particularly costs related to subsequent hospitalizations, but only when costs are considered from society’s view; that is, all costs irrespective of whether Medicare actually pays them and using a very low discount rate of 3%, as is frequently done when estimating the effects from a societal point of view. Otherwise, discount rates are usually in the 5% to 6% real (inflation-adjusted) rate. They also found that if the cost of a single medication is limited to $90, full coverage becomes cost-saving from Medicare’s perspective as well, which they believed is plausible, given the availability of $4-per-month generics.
The authors1 conclude that their findings make a business case for Medicare to expand current levels of prescription drug coverage provided under Part D for post-MI secondary preventive therapies, but that is not obvious to this economist. If Medicare Part D policies, along with other insurance plans, were to begin adopting value-based insurance strategies, lower copayment rates would be expected for medications for particular populations that improve their health, particularly for those medications that can be shown to be broadly cost effective. Proper medications for secondary prevention of acute coronary artery disease would certainly seem to fall in this categorization. I am a proponent of such value-based insurance designs, which are consistent with my general interest in moving toward greater use of comparative clinical effectiveness measures in helping to set more appropriate reimbursement rates. It therefore makes sense to have lower, although not necessarily zero, copayments for the lowest-cost drugs like generics and also for the most efficacious drugs for patients with particular diagnoses.
All of this is very different than claiming that their findings provide a business case for Medicare to expand its current levels of drug coverage for post-MI preventive therapies. Although I am surprised by their finding that Medicare does not save money from the expanded use of these drugs, given their estimated effect on reducing the relative risk of coronary heart disease, a business case would require the cost savings from other Medicare expenditures to be at least equal to the added cost of expanding the drug benefit. Otherwise, it seems more like a prudent additional expenditure associated with improved clinical outcomes and meeting general criteria of cost effectiveness, which are actually not required for Medicare reimbursement, than a business case per se. More generally, it is also not clear why it would make sense to expand to full coverage for medications prescribed after acute MI but not for the medications that treat a variety of chronic diseases that plague the elderly as well.
Trying to assess why the results were cost-saving, however, raises some additional questions about both the way the study results were calculated and the policy implications that have been generated from those findings. Full coverage was found to be cost-saving from the perspective of a typical commercial insurance company but not Medicare, partly because Medicare provides somewhat lower hospitalization reimbursement rates than at least some private insurance and because the amount of patient cost sharing is assumed to be much greater for Medicare than for privately insured patients (63% versus 32% of drug costs). This means that the costs of “buying up” the current patient share is quite high, at least compared with the average employer-sponsored insurance policy. Nonetheless, the results are surprising given the relatively lower costs of drugs and the high costs of acute and post-acute care, even considering the relatively limited amount of nursing home care covered by Medicare, associated with future adverse coronary events and must reflect the relatively low likelihood of such an adverse secondary outcomes compared with the numbers of individuals who have such initial events.
Given the empirical findings of Choudry et al,1 one possible policy conclusion is that full coverage of the combined pharmacotherapy after acute MI is cost-saving from a societal point of view and incrementally cost effective under traditional measures and should therefore be enacted, even though it is not cost-saving to Medicare because the relative low coverage under Medicare makes it not cost-saving, which is what the authors conclude.
A second approach is to assess whether, given the current program design, more help is needed in guiding people with coronary heart disease or other known diseases to the right plans in which it would be unlikely that their actual coverage would be as limited as the average suggests. The current structure of Part D actually offers better coverage than the analysis by Choudhry et al1 predicts, assuming people pick their plans according to the medicines they expect to use, which of course is what Medicare tries to help people do on their Web site. Seniors who specify the drugs they are taking are provided with the plans in their area that have the best coverage for those drugs.
Under the current structure, beneficiaries pay a deductible and then typically pay 25% of the first block of costs until they hit the so-called “doughnut hole” where they have no coverage for about $3000, after which they have approximately 95% coverage. The doughnut hole for 2008 begins after drug expenditures of $2510 and continues until expenditures of $5726 have been reached. The low average coverage reflects the effect of the lack of coverage in the doughnut hole. Even that exposure can be moderated, however. Almost a third of the plans offer some coverage for the doughnut hole, usually generics.
It would be useful to know what percentage of the people on Medicare would be able to choose at least one plan that provides coverage for at least one drug in each of the 4 categories, including the desired combination pharmacotherapy plus generic coverage of each of the categories. My guess is that most or all seniors would have access to such plans. Assisting individuals who experience major health events during the year with choosing the appropriate prescription drug plan in the following year could and should be part of an additional educational strategy for the Centers for Medicare & Medicaid Services to consider. With the right assistance, it should be possible for seniors to choose plans that provide the coverage they need, although not necessarily full coverage, without a change in the program.
An alternative strategy would be to encourage individuals who experience major health events such as an acute MI to join Medicare Advantage plans, particularly those that involve integrated delivery systems like Kaiser. Medicare Advantage plans not only frequently provide more drug coverage for a given expenditure of money than a separately purchased Part D plan but many also include disease management programs as well. They are also better suited to provide the kind of patient education that is needed to convince or persuade people to alter their life styles after a major health event and to take the proper medications.
The policy challenge for Medicare raised by the authors1 is far bigger than determining whether full coverage for a combined set of medications after an acute MI is justified or appropriate. Most seniors suffer from one or more chronic diseases, and the majority of Medicare expenditures are for people with multiple chronic diseases. The program, however, is designed on an acute-care model. Physicians are reimbursed for specific services on the basis of a disaggregated fee schedule that encourages the provision of more and more complex services. It should not come as a surprise that the volume and intensity of services under Part B of Medicare, which reimburses physicians, continues to grow, putting downward pressure on individual fees even when the most appropriate treatment may be education and counseling of patients along with periodic checks on their follow-up care.
Fixing Medicare to better reflect the medical conditions confronting seniors and providing incentives for the proper care of these patients is going to require much more change than just pressing for the full coverage of specific medications on a disease-by-disease basis. The Medicare physician fee schedule needs to be reconstructed so that payments are bundled together for the treatment of chronic diseases. Bundling payments to include all of the charges (that is, hospital and physician charges) for caring for patients with high-cost–high-volume diagnoses is another strategy that would encourage a more coordinated provision of services. Establishing a medical home for seniors who remain outside of a Medicare Advantage plan and paying for the coordination of care for these individuals is another strategy that is receiving increasing attention. Encouraging, providing incentive for, or even requiring case management for patients who experience major medical events may also provide better-quality cost-effective care.
Low copayments for medically appropriate therapeutics are a fundamental part of value-based insurance. I support the concept. However, pressing for full coverage for specific pharmacotherapy classes on a disease-by-disease basis is not likely to be an effective way to fix the current program. The question is how to best move the current, fragmented, stove-piped system that characterizes Medicare to one that rewards physicians and institutions that provide high-quality cost-effective care to patients. This will require fundamental changes to Medicare as we now know it. In the interim, individuals who need specific medications after acute medical events or in response to chronic diseases need to understand the choices of drug programs available, get help if needed in using the Medicare Web site that can identify plans best-suited for particular medication needs, and be encouraged to consider the advantages of more integrated care where it is available in their communities.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.