on Facebook
on Twitter
on YouTube
RAND Review
Quick Fix
Drug Benefit Plans Driven by Short-Term Savings Could Be More Costly in the End
By Geoffrey F. Joyce and Dana P. Goldman
Geoffrey Joyce is a RAND senior economist, codirector of the UCLA/RAND Health Services Research Training Program, and a professor at the Pardee RAND Graduate School. Dana Goldman holds the distinguished chair in health economics at RAND, directs the Bing Center on Health Economics at RAND Health, and is a professor at the Pardee RAND Graduate School.
AP IMAGES/MORRY GASHLuillia Van Lanen looks at some of her prescription medications in her Madison, Wisconsin, apartment on March 29, 2007. She was so embarrassed about not being able to afford her $300 heart medicine that she lied to her doctor when he asked if she was taking it. |
In many developed countries, medical practice has changed dramatically in recent decades with the increased use of prescription drugs, the availability of more and better-quality drugs to prevent and manage chronic illness, and the ability of these drugs to reduce mortality, forestall complications, and make patients more productive. Ensuring access to prescription drugs has become a cornerstone of an efficient health care system.
At the same time, the resultant increase in drug spending has prompted pharmacy benefit managers and health plan administrators to seek either to reduce the use of prescription drugs or to steer patients toward less-expensive alternatives and distribution methods. The rapid proliferation of coinsurance plans, tiered formularies, mandatory generic substitution, mail-order pharmacies, and other cost-saving measures has transformed the drug benefit landscape.
Given the burgeoning demand for prescription drugs and the consequent efforts to rein in their costs, we undertook a systematic review of the published literature on how the salient cost-sharing features might affect drug access, medical spending, and health outcomes. We found several reasons for concern, based on 132 studies published in English between 1985 and 2006 on the effects of drug cost sharing in Australia, Britain, Canada, Germany, Sweden, the United States, and Taiwan.
We also conducted our own study of an extreme version of cost sharing: the use of an annual limit or “cap” on prescription drug benefits. A benefit cap represents cost sharing at the extreme because patients who reach the spending cap must pay all additional pharmacy costs out of pocket. Such a cap is an integral part of the new Medicare Part D drug benefit for elderly Americans.
We found that capping drug benefits for older people causes them to quit taking their medicine at higher rates than do elderly patients with noncapped benefits. The result is consistent with the growing body of evidence that higher drug cost sharing can reduce adherence to prescriptions, leading to adverse health outcomes and higher medical costs.
Higher drug cost sharing can reduce adherence to prescriptions, leading to adverse health outcomes and higher medical costs. |
Types of Cost Sharing
In today’s drug benefit market, most beneficiaries are covered by formularies in which drugs are assigned to price tiers based on cost, the availability of close substitutes, and other factors. Generic drugs, preferred brands, and nonpreferred brands might require copayments of $5, $15, and $35, respectively. In addition, plans may require beneficiaries to pay coinsurance, which is a percentage of the total cost of a dispensed prescription.
The purposes of tiered copayments and coinsurance are twofold: to encourage patients to use generic or low-cost brand-name medications and to prod manufacturers to offer discounts in exchange for having their brand-name products included in a preferred tier. By 2005, 74 percent of U.S. workers with employer-sponsored coverage were enrolled in drug plans with three or more copayment tiers, nearly three times the rate of 27 percent in 2000.
Some plans also impose benefit caps, which limit either the coverage amount or the number of covered prescriptions. For example, the standard Medicare Part D benefit offers beneficiaries coverage up to $2,400 in spending in 2007, at which point coverage stops until beneficiaries reach a catastrophic limit of $5,451 in total annual drug spending. Once the catastrophic limit is reached, coverage resumes with minimal cost sharing.
Prior to the introduction of Part D, benefit caps — with no catastrophic limit — were a standard feature of Medicare+Choice plans, now known as Medicare Advantage, and some retiree plans. As of 2002, 94 percent of Medicare+Choice plans that covered brand-name drugs had an annual dollar cap, ranging from $750 to $2,000.
Doubling copayments reduced the use of cholesterol-lowering therapies and antidiabetics by 25 percent — and the use of anti-inflammatories and antihistamines by as much as 45 percent. |
In addition to copayments, coinsurance, and benefit caps, other cost-saving measures include prior authorization (requiring permission before certain drugs can be dispensed), step therapy (requiring the use of lower-cost medications before providing coverage for higher-cost alternatives), closed formularies (offering zero coverage for nonpreferred brands), mandatory generic substitution, and reference pricing (capping the amount a plan will pay for a prescription within a specific therapeutic class).
Effects of Cost Sharing
Our review of 132 studies from around the world showed that for each 10 percent rise in cost sharing required of patients, either as copayment or coinsurance, their prescription drug use fell by 2–6 percent, depending on the therapeutic class and disease condition. For example, we found that doubling copayments reduced the use of cholesterol-lowering therapies and antidiabetics by 25 percent — and the use of anti-inflammatories and antihistamines by as much as 45 percent. Patients were less likely to reduce the use of these drugs if they were receiving ongoing care from a physician for the disorder. For example, doubling copayments decreased the use of antidepressants by 25 percent overall, but by just 8 percent among those currently in treatment for depression.
With respect to drug benefit caps, the most salient evidence comes from an analysis of medical and pharmacy claims from a single health maintenance organization. Members whose annual drug benefits were capped at $1,000 had 31 percent lower pharmacy costs than comparable enrollees who were not subject to a cap, but the capped enrollees also had higher rates of emergency department visits and nonelective hospitalizations.
The weight of evidence clearly demonstrates that increased cost sharing is associated with lower pharmaceutical use. These effects can be quite large, even for long-term medications, suggesting that there could also be long-term health consequences.
AP IMAGES/GREGORY SMITHJames Antonoff, 80, poses with his Medicare paperwork at home in Alpharetta, Georgia. Antonoff, who requires 13 prescriptions for a litany of ailments including glaucoma and gout, experienced difficulty selecting an affordable Medicare drug plan. He and his wife spent 40 hours comparing alternatives. |
Indeed, the weight of evidence suggests that increased patient cost sharing can adversely affect health outcomes. The results from studies that focused on the chronically ill were unambiguous: For patients with congestive heart failure, high cholesterol, diabetes, and schizophrenia, higher cost sharing for prescription drugs was associated with greater use of inpatient and emergency medical services. The impact of a benefit cap, in particular, fell disproportionately on beneficiaries with moderate-to-high drug expenses, many of whom had chronic illnesses that required ongoing drug therapy.
The results from studies that looked at prescription drug cost sharing more broadly were ambiguous. Some of these studies found that higher cost sharing was associated with adverse outcomes, especially among the elderly and poor. Other studies found that when the population was not limited to those with certain chronic illnesses, the outcomes were mostly benign.
In general, though, increased cost sharing for prescription drugs is associated with lower rates of drug treatment, worse adherence to prescriptions among patients, and more frequent discontinuation of therapy. For some chronic conditions — including congestive heart failure, high cholesterol, diabetes, and schizophrenia — increased cost sharing is also associated with an increased use of health care services other than prescription drugs, offsetting any potential cost savings from the reduced use of the drugs themselves.
The adverse effects of medication cost-containment policies may be magnified among low-income groups, whose high rates of chronic health problems, combined with low incomes, may result in more price-sensitive behavior. Low-income people with moderate-to-high drug expenses will likely pay a higher fraction of their incomes on medications and face longer gaps in coverage than will wealthier people.
Most evidence on this point comes from studies of Medicare and Medicaid, the joint federal and state health insurance plan for poor and disabled Americans. Medicaid enrollees in South Carolina used significantly fewer drugs after the imposition of just a 50-cent copayment. Elderly Medicaid recipients in states with copayments consumed fewer drugs and were less likely to fill any prescriptions during the year than were those in states without copayments. Among Medicare beneficiaries in Pennsylvania, those with annual incomes of more than $18,000 were 18 percent more likely to treat medical problems with prescription drugs than were those with incomes of less than $6,000.
Beware the “Doughnut Hole”
In our own study of benefit caps, we compared the behavior of U.S. retirees in capped and noncapped drug plans offered by a large private employer. The capped plan was comparable to the Part D drug benefit that Medicare now offers to all elderly Americans with one exception: Part D coverage resumes after a certain level of spending within the calendar year.
About one-quarter of Part D enrollees are expected to fall into the doughnut hole in 2007. Many seniors with high drug expenses, especially the chronically ill, will then face higher out-of-pocket costs. |
The Part D coverage gap, which in 2007 falls between $2,400 and $5,451, is known as the “doughnut hole.” Under the standard benefit, beneficiaries must pay first a $265 annual deductible and then 25-percent coinsurance on annual drug costs up to $2,400. Beyond $2,400, the coverage stops completely until beneficiaries hit $5,450 in total drug spending — or total out-of-pocket expenses of about $3,850 for the year. At $5,451, coverage resumes with minimal patient cost sharing.
About one-quarter of Part D enrollees are expected to fall into the doughnut hole in 2007. Many seniors with high drug expenses, especially the chronically ill, will then face higher out-of-pocket costs. Although it is still too early to assess the effects of Part D, which just began in 2006, it is possible to gain insight into the likely effects by examining how retirees have fared under private insurance plans with similar benefit caps.
In our study of data from a large private employer, the annual prescription drug benefit for some retirees was capped at $2,500 in plan spending, after which the members had no drug coverage until benefits resumed in the next calendar year. The benefit for other retirees was not capped. Enrollees in both plans paid 35-percent coinsurance for both generic drugs and preferred brands and 60-percent coinsurance for nonpreferred brands, subject to identical out-of-pocket maximums per prescription.
We focused on enrollees whose annual prescription drug spending was at least $2,400 and who were thus most likely to be affected by a $2,500 cap. Such high-cost enrollees numbered about 7,200 in total, with those in the capped plan differing only modestly from those in the noncapped plan across several demographic and health measures (see Figure 1).
|
|
One of our most alarming discoveries was the roller-coaster nature of drug consumption among those in the capped plan relative to those in the noncapped plan over the course of nearly three years. |
One of our most alarming discoveries was the roller-coaster nature of drug consumption among those in the capped plan relative to those in the noncapped plan over the course of nearly three years. By December of each year, the use of antidepressant, antidiabetic, antihypertensive, and cholesterol-lowering agents was 15–28 percent lower among members in capped versus noncapped plans (see Figure 2). The relative reductions in use were even larger for two drug classes with broadly available over-the-counter substitutes: The use of anti-inflammatories and antiulcerants were 30 to 50 percent lower among members in capped versus noncapped plans by the end of each year.
However, these differences narrowed greatly once coverage resumed in the next year. For example, by March 2004, drug utilization differed by less than 10 percent for five of the six therapeutic classes. Put another way, the trends by therapeutic class clearly indicate relatively severe drops in use at the end of the year for those in the capped plan, followed by large upswings when coverage resumed.
We measured the annual rates of discontinuation and reinitiation of prescription drugs across seven therapeutic classes: antidepressants, antidiabetics, anti-hypertensives, anti-inflammatories, antiulcerants, cardiac drugs, and cholesterol-lowering drugs. We found that those in the capped plan who spent $2,400 or more on medications in a given year discontinued them at consistently higher rates across all drug classes than did those in the noncapped plan (see Figure 3). The drug classes with the greatest differences in discontinuation rates were antihypertensives (22 percent under the capped plan versus 14 percent in the noncapped plan) and cardiac drugs (28 versus 19 percent).
Figure 3 —Drug Benefit Caps Were Associated with Higher Rates of People Both Stopping Their Use of Medications and Reinitiating Use the Next Year |
SOURCE: A Look Inside the “Doughnut Hole,”, 2007.NOTES: Cardiac drugs include antihypertensive agents as well as anti-arrhythmic agents, cardiac glycosides, anticoagulants, and cardiac drugs not elsewhere classified. Findings for anti-inflammatories and antiulcerants, while consistent with the patterns for other drug classes, were confounded by the widespread availability of over-the-counter alternatives and other market factors and so are not depicted. |
Patients discontinue therapy for many reasons. But if the cap is playing a role, we would expect to see higher rates of resumption in the capped plan once the year ends and benefits resume. In fact, we do observe such a pattern, as also shown in Figure 3. Enrollees in the capped plan reinitiated drug use at higher rates when benefits resumed the next year, suggesting that decisions to discontinue had been strongly influenced by the benefit cap.
These results suggest that capping benefits is disruptive to drug therapy for many chronically ill patients. At the same time, high-cost enrollees who reach the cap also switched to other plans at disproportionately higher rates, suggesting that a benefit cap’s long-term ability to contain costs is limited by the fact that some users who hit the ceiling will migrate to less-restrictive plans.
Unlike the private benefit plan we studied, of course, Medicare’s Part D drug benefit resumes when spending reaches $5,451 within a calendar year. Whether this additional benefit, which might affect 10 to 15 percent of the enrolled population, will result in different beneficiary responses remains unclear.
The policy challenge for Medicare today is to motivate prescription drug users to spend prudently without leading them to forgo needed and cost-effective medication. One promising trend is that several private plans have begun covering most generic drugs within the Medicare coverage gap at little additional cost.
The question now is whether Part D itself can be redesigned to offer continuous coverage at the same overall cost. As one idea, requiring the elderly to make higher copayments that remain constant throughout the year — in exchange for uninterrupted coverage — might result in better long-term clinical outcomes.
Prescription Regimen
Many questions remain regarding drug cost sharing in general and drug benefit caps in particular. The remainder of this essay prescribes a course of future inquiry. Five key research issues remain unresolved.
The policy challenge for Medicare today is to motivate prescription drug users to spend prudently without leading them to forgo needed and cost-effective medication. |
First, while greater cost sharing is certainly associated with reduced access to drugs, the precise mechanisms are obscure. Reduced access could be triggered by reduced initiation of drug treatment, worse adherence among existing users, or more frequent discontinuation of therapy. Distinguishing among these three is important to help physicians and benefit plan managers provide proper monitoring and advice to counteract any adverse consequences. We found evidence that all three triggers may be complicit when cost sharing rises; however, worse adherence seems to be the primary culprit.
Second, the time horizon over which higher cost sharing is associated with adverse medical events — such as hospitalizations and worsening clinical outcomes among patients with congestive heart failure, high cholesterol, diabetes, and schizophrenia — is longer than one to two years. There might be adverse consequences for asthmatics as well. Because patients leave employers and plans frequently, and because benefits change rapidly, it is difficult to isolate any long-term consequences using only short-term studies.
Third, there is little reliable evidence to support our supposition that the worst effects of cost-containment policies are likely to be suffered among low-income groups, whose high rates of chronic health problems and low incomes could diminish their adherence to prescriptions. One of the severe limitations of claims data is that they do not include information on race, ethnicity, income, education, and wealth. Whenever economic status is included, as derived from national survey data, there is substantial bias in its measurement.
Fourth, the introduction of Medicare Part D has initiated a bold experiment with benefit caps. The dynamics of discontinuation and reinitiation of therapy from year to year under Part D, as well as the effects of the catastrophic limit, have yet to be assessed. Understanding how patients respond to a benefit cap is also important because it serves as a counterpoint to consumer-directed health plans with high-deductible catastrophic coverage. In these latter plans, patients must pay all the costs until a cap is reached, beyond which they pay nothing. Benefit caps, in contrast, provide coverage up to a specified limit. A comparison of these financing alternatives is needed, especially with regard to how they might affect those with chronic illnesses.
Fifth, there has been a marked increase in the use of costly specialty drugs, which are often used to treat complex chronic conditions, such as anemia, cancer, growth hormone deficiency, and multiple sclerosis. Spending on specialty drugs is expected to soar in the near future as new drugs enter the market for the treatment of diabetes, osteoporosis, and rheumatoid arthritis — diseases that affect much larger populations. Many insurers are contemplating a variety of cost-sharing strategies to control the use and cost of these drugs. This area may be the next frontier of dramatic changes in drug benefit design. It will be important to assess the consequences for spending and health.
To date, the evidence suggests that more cost sharing is associated with more hospitalizations and emergency department visits, particularly among the chronically ill. In some cases, cost sharing can steer patients to more cost-effective treatments without harming health. But for drugs that are known to be cost-effective in preventing expensive medical care, greater cost sharing appears to be cost-ineffective in the long run.
The challenge for public and private prescription drug plans is to make patients more sensitive to the cost of drug treatment without inducing them to forgo cost-effective care. This requires knowing how patients respond to different incentives and evaluating the net benefits of alternative drug therapies, not only for the immediate effects of the drugs but also for future health care costs, patient productivity, and patient utility. 
Top