Coverage and Reimbursement: Incentives and Disincentives for Product Development
As a person who needs to take the MOST expensive drug in the world to stay healthy … I am in my early 30s and looking to take a $400,000/year drug for the rest of my life…. I now have to pick jobs based off of their insurance plan, rather than the job itself.
Posted by Billiondollarwoman, 2010
Big companies are starting to get more interested in rare diseases, but the key issue is the high cost of developing a drug and the typically long time it takes to move it from a lab into a clinic as a treatment that gets prescribed. Before starting down this arduous path, a company needs to feel there is a reasonable chance of making a profit.
A small market is generally viewed as a disincentive for the development of drugs. Many of the costs of developing a new drug are incurred regardless of the size of the potential market. If, however, a company can set a price that is high enough to recover its costs and generate profits because enough public and private health insurance plans and patients and families will pay that price, then a manufacturer may not be deterred by a small target market. Some orphan drugs are among the most expensive drugs in the world, costing as much as $400,000 per year.
Orphan drugs can be very profitable. Wellman-Labadie and Zhou (2009) reported that 43 brand-name drugs with global sales of more than $1 billion had orphan drug designations, and 18 of these had been approved
solely as orphan drugs. Most had been approved for the relatively more prevalent rare conditions such as multiple myeloma, but one, imiglucerase (Cerezyme), was approved for Gaucher’s disease, which has a U.S. patient population estimated at 3,000 to 6,000.1 In 2008, about 1,500 patients in the United States were taking the drug, which was priced at more than $300,000 per year (Pollock, 2008).
Biotechnology companies have been prominent in the orphan drug market from the outset, and the Orphan Drug Act has been cited as a key factor in their growth (OIG, 2001b; Ariyanchira, 2008; Grant, 2008). Small- to medium-sized companies have also played a significant role in orphan drug development (Seoane-Vazquez et al., 2008; Villa et al., 2008). Perhaps influenced by challenges in developing traditional blockbuster drugs as well as by the market protection and other incentives provided by the Orphan Drug Act, some major pharmaceutical companies have recently announced that they are considering or pursuing orphan drug development; many already have at least one orphan drug on the market (Anand, 2005; Dimond, 2009; Pollack, 2009; Whalen, 2009).
In addition to incentives for developing orphan drugs provided by the Orphan Drug Act, the potential profitability of orphan versus nonorphan drugs may be affected by several other factors. One is that private health plans generally have little leverage in negotiating prices for expensive biotechnology drugs, many of which are orphan drugs. Even the Kaiser Permanente system in California, which is large and accustomed to negotiating prices, has noted that “opportunities are limited” in this arena (Monroe et al., 2006). In a Government Accountability Office (GAO, 2010b) survey, sponsors of Medicare prescription drug plans cited the lack of competitors in the market for a drug (which gives manufacturers little reason to offer discounts) and the limited volume of a drug used by the plan (which limits a plan’s negotiating power) as two primary reasons for a health plan’s lack of leverage on drug prices. Orphan drugs often have both of these characteristics.
Colchicine provides an example of the implications for health plans and patients of limited competition. The drug, which had never been approved by the Food and Drug Administration (FDA) because it predated the 1938 Food, Drug, and Cosmetic Act, has long been used to treat gout and has for some time been used to treat familial Mediterranean fever, a rare disease. A company recently received FDA approval for the product as a treatment for gout and, in addition, as an orphan drug to treat familial Mediterranean fever (NDA 022352). It then increased the price of the drug (which is taken two to three times daily) approximately fiftyfold, from about $0.10 a pill to
$5 a pill; it also initiated lawsuits to force unapproved competitor products off the market (ACR, 2010; Rockoff, 2010).2
When patents and other market protections no longer apply, orphan drugs may still face limited competition because they are less likely than nonorphan products to face competition from generic drugs (see Appendix B). Generic competition results in lower prices and also in large losses in market share for the original brand-name product within 6 to 12 months of generic entry (Grabowski and Vernon, 1992; Frank and Salkever, 1997; Reiffen and Ward, 2005). The small market anticipated by potential generic competitors is one explanation for the lower availability of generics. In addition, many orphan drugs are biologics, which historically have had no clear path for generic approval. The recent Patient Protection and Affordable Care Act of 2010 (P.L. 111-148, hereafter the Affordable Care Act) will change this (see Chapter 3). Nonetheless, it is not clear how much generic competition will arise given that biologics are harder to manufacture and more variable than small-molecule drugs, so FDA may require additional clinical and nonclinical data that are not required for small-molecule drugs.
Also, in markets for ordinary drugs, the initial generic entrant (which is granted a 180-day exclusivity period among generic manufacturers) typically “shadow-prices” (i.e., charges a lower but not a substantially lower price, e.g., 15 percent, than the original brand-name drug) (Reiffen and Ward, 2005). Consistent with this strategy, the chief executive officer of a company with a product that will compete with Cerezyme has said that the company plans to price the product at a 15 percent discount to gain market share (cited in Douglas, 2010). For nonorphan drugs, however, generic prices tend to fall sharply once additional generic competitors enter the market. For orphan products with a small potential market, the entry of multiple generic drugs is less likely; thus limited price competition can be expected to persist.
One factor that could moderate costs for orphan drugs is that manufacturers of orphan products have little need to invest heavily in marketing their drugs because the target populations of physicians and patients are so small. Manufacturers can also often expect that advocacy groups will be active in spreading information about new treatments.
Notwithstanding examples of profitable orphan drugs, companies considering the development of a drug for a disease that affects a small population must evaluate prospects for each potential product individu-
ally. In addition to market size and costs for research and development, an important consideration is the insurance status of target patients—not only whether they are covered at all but also the scope of coverage and the limits placed on it.
Individuals with serious rare conditions can face a number of problems with health insurance. Insurers, particularly companies that market products directly to individuals rather than indirectly through employer groups, have an understandable concern about covering individuals who do not seek insurance until they or a family member is diagnosed with a serious illness. Companies have therefore screened or underwritten individuals based on their health status and history. As a result, people with both common and rare diseases without access to employment-based health coverage have found it difficult to secure health insurance at an affordable price or at all. For example, an analysis of 2004 Medical Expenditure Panel Survey data found that only 13 of the 7,000 individuals under age 65 who had a disability (broadly defined) reported having nongroup insurance (cited in IOM, 2007). Many restrictive underwriting practices will change as a result of the Affordable Care Act, which should make it easier for some individuals with a rare disorder to obtain coverage in the future.
In response to health care cost increases that have persistently exceeded inflation in the economy overall, insurers have developed an array of strategies to control costs for those they do insure, including transferring more costs to health plan members and adding administrative mechanisms to identify and discourage inappropriate care. Thus, in addition to considering the likelihood that target patients will have insurance, manufacturers may consider
the processes that different payers use to determine, first, whether to cover a drug and, second, what to pay for it;
the ways in which payment methods and coverage levels may differ based on the site where the drug is administered;
the administrative controls that insurers, governments, or other third parties may place on coverage, for example, requirements for prior authorization of very expensive prescription drugs;
the amounts that insured patients will have to pay out of pocket, which may vary both across and within different categories of drugs; and
the existence of state or federal mandates that require coverage of certain classes of drugs.
Recent trends in the design and management of prescription drug benefits already include high patient cost sharing for some drug categories, especially expensive drugs. An extensive literature on the effects of patient cost sharing indicates that it reduces both needed and unneeded use of
services (see summary in Newhouse et al., 1993) and also that consumers show considerable price sensitivity in making decisions about prescription drug purchases and use, even for drugs that are critical to their health (see, e.g., Frakt and Pizer, 2009; Solomon et al., 2009). As described below, other practices—such as the use of tiered formularies that favor some drugs or drug classes over others—could also affect the use of orphan drugs.
In the next few years, the Affordable Care Act will, if successful, expand access to health insurance for people under age 65. This should benefit companies that develop drugs and biologics as well as patients and families. At the same time, given that health care costs continue to consume a growing share of the Gross Domestic Product and that financial projections for Medicare and Medicaid are alarming, pharmaceutical companies must consider the prospect that governments, employers, and insurers may in the future impose price controls, try to negotiate more vigorously on drug prices, transfer a much higher share of drug costs to patients, or add further administrative barriers to expensive drugs. Pharmaceutical companies may, in addition, contemplate the risks of some kind of backlash against very high prices for orphan drugs, especially if the drugs are also very profitable.
The rest of this chapter examines how the policies and decisions of public and private insurance programs may create incentives or disincentives for companies to develop drugs for small populations. (Chapter 7 discusses policies and practices involving medical devices.) It also examines Medicare and other health plan policies on coverage for certain costs incurred by insured individuals who participate in clinical trials. The chapter concludes with recommendations. Appendix C presents an analysis of coverage of orphan drugs by the private prescription drug plans for Medicare beneficiaries. The focus here is on drugs specifically developed or marketed for people with rare conditions rather than on drugs that are used to relieve pain, respiratory distress, and other symptoms of both common and rare conditions. (As used in this chapter, the term drug includes biologics unless otherwise specified.)
This chapter emphasizes Medicare policies, in part because many adults with debilitating rare conditions are covered by Medicare by virtue of age or qualification for Social Security Disability Insurance (SSDI). In addition, information about Medicare is more readily available than information on private health plans. Although variation is introduced by the contractors that administer various elements of the Medicare program, it is a single program in contrast to the 50-plus Medicaid programs and the thousands of private health plans for which systematic information is limited. The chapter includes brief discussions of Medicaid, private health plans, and company assistance programs and reviews some provisions of recent legislation that may make insurance more available and moderate some limits on coverage, for example, lifetime caps on benefits. The discussion does not examine
coverage of genetic testing.3 It also does not investigate health programs of the Veterans Health Administration or the Department of Defense.
For the most part, this chapter considers health plan coverage and reimbursement policies and practices from the perspective of companies that develop drugs, not from the perspective of patients and families or from perspective of public policy makers considering issues of equity and affordability in the context of escalating government budget deficits. Some of the patient and family stories in Chapter 2 illustrate the importance of insurance to individual and family security. The committee found no analyses of public or private expenditures specifically for orphan drugs.
MEDICARE AND MEDICAID COVERAGE OF FDA-APPROVED DRUGS
Responding to the growing availability of effective medical services and the difficulties that older people faced in paying for these services directly or obtaining health insurance, Congress created Medicare in 1965 to cover people age 65 or over, regardless of income or health status.
Today, Medicare also covers people who qualify for SSDI, although they must generally wait 2 years before they are eligible for Medicare. Congress has provided an exception to this waiting period for SSDI beneficiaries who have amyotrophic lateral sclerosis and has also provided that adults and children with end-stage renal disease are automatically eligible for Medicare whether or not they have qualified for SSDI (SSA, 2010b). In addition, under its compassionate allowances policy, the Social Security Administration has created a mechanism for quickly identifying and processing SSDI applications for individuals with specific conditions that invariably qualify them for benefits. The initial list of conditions included 25 rare diseases and 25 cancers; nearly all of the 38 conditions added in 2010 were rare diseases (SSA, 2010a). Although this procedure does not shorten the Medicare waiting time after SSDI qualification, it does reduce the total waiting time. In 2008, approximately 7.7 million Medicare beneficiaries (about 16 percent of all beneficiaries) had qualified for coverage by reason of disability rather than age (CMS, 2009d).
In 1965, Congress also created the federal-state Medicaid program to insure certain categories of low-income individuals (primarily low-income
A 2009 overview of Medicare policy on genetic testing noted that Medicare did not cover genetic testing based only on a family history of disease or in the absence of evidence that the test provided useful information to guide patient care (Straube, 2009). After determining that evidence was insufficient to support genetic testing to guide use of warfarin, the Centers for Medicare and Medicaid Services (CMS) recently announced it would support a clinical study of pharmacogenetic testing to predict patient responsiveness to the drug (Genomeweb.com, 2010).
mothers and children and low-income aged, blind, or disabled people). Certain people (“dual eligibles”) qualify for both Medicare and Medicaid. The federal government sets many of the basic rules for Medicaid and subsidizes state programs to varying degrees, but states have some leeway in deciding who and what to cover and how much to pay providers.
Following a model that had been established in private health insurance, Congress initially divided Medicare into two parts: hospitalization insurance (Part A) and supplementary medical insurance for physician and certain other services (Part B). Building on policies initiated in the 1970s, the Medicare Advantage program (Part C) provides Medicare beneficiaries opportunities to enroll in private health plans.4 In 2003, Congress created a Medicare outpatient prescription drug benefit (Part D), which was implemented in 2006 and is available only through private plans. Medicare is managed by the Centers for Medicare and Medicaid Services (CMS).
For beneficiaries not enrolled in Medicare Advantage plans, how Medicare pays for drugs and what controls it places on payments varies depending on where the drug is administered, for example, hospital, physician office, or home. Although the committee did not find any systematic analysis, a review of the list of approved orphan drugs suggests that most of them are administered in physician office or outpatient clinic settings or are taken by patients at home. Thus, for most drug companies as well as patients and families, Medicare policies related to Part B and Part D are of greater interest than Part A policies.
As discussed below in the section on coverage of certain costs in clinical trials, Medicare does not cover investigational drugs. CMS and FDA recently signed a memorandum of understanding to share data, including FDA data on drugs and medical devices that have not yet been authorized for marketing (75 Fed. Reg. 48699). The agencies are also considering a process of parallel review of products that would reduce the lag between FDA marketing authorization decisions and CMS national coverage determinations (75 Fed. Reg. 57045).
Medicare Part A
Medicare Part A covers inpatient services provided in hospitals. It also covers certain short stays in skilled nursing facilities, hospice services, and
certain home health services. Payment methods vary for each of these categories. Medicare pays for inpatient hospital care on the basis of prospectively determined rates for specified diagnosis-related, severity-adjusted bundles of services (OIG, 2001a). Oversimplified, each diagnosis-related group (DRG) payment has a weight assigned to it based on the average amount of resources used in caring for Medicare patients in that group. (In addition to diagnosis, assignment to a group takes into account other factors, including procedures, age, and comorbidities or complications.) Resources include facilities and services such as routine nursing care, laboratory tests, imaging services, intensive care, and all medications administered during the hospital stay. Within a payment category, payments generally will not vary based on which drugs are included in a hospital’s formulary or other details of how hospitals manage services to beneficiaries.5 The objective of this payment method is to control Medicare expenditures for hospital care by encouraging hospitals to provide care efficiently and economically.
CMS revises DRGs annually based on analysis of past inpatient claims data. In 2000, Congress provided that CMS could make an additional payment for certain new technologies on an interim basis until claims data were available to guide a normal revision in rates (Clyde et al., 2008). In general, Medicare pays half of the incremental cost to the hospital associated with the new technology on a case-by-case basis. To qualify for an add-on payment,
the technology must be new (generally meaning that it was approved by FDA within the preceding 2 to 3 years);
it must offer a substantial clinical benefit over existing options; and
it must not be adequately covered by the existing DRG payment.
In the first 9 years after the policy was implemented in 2001, CMS rejected more than 20 applications for add-on payments and approved 9 applications, 8 of which were for medical devices (Clyde et al., 2008; Berger, 2009; CMS, 2009c; Hill, 2010; see also Chapter 7). The one drug for which an add-on payment was approved was not for an orphan drug.
Medicare Part B
Medicare Part B covers physician services, hospital outpatient care, certain home health services, certain clinical laboratory services, some preventive services, durable medical equipment, and certain drugs. Covered
drugs include drugs administered in physician offices that usually are not self-administered, drugs used as part of durable medical equipment (e.g., nebulizers), and in some cases, immunosuppressive drugs used after organ transplants and other drugs as authorized by Congress (CMS, 2010; see also, MedPAC, 2008; Cassidy, 2009). Under these provisions, Part B covers certain orphan drugs that are administered by infusion or injection in a physician’s office or clinic, for example, galsulfase (Naglazyme), a treatment for Maroteaux-Lamy syndrome, a rare metabolic disorder. Likewise, Part B typically covers certain orphan drugs that are administered at home using equipment that is covered by Part B. An example is dornase alfa (Pulmozyme), a medication for cystic fibrosis that is inhaled using a neubulizer.
In addition, Part B also covers drugs provided as part of hospital out-patient services. These services are covered by a prospective payment system that includes inexpensive drugs and diagnostic radiopharmaceuticals as part of the packaged payment to an outpatient facility for a service such as a surgical procedure. More expensive drugs and biologics are reimbursed separately (MedPAC, 2008).
The Medicare Prescription Drug Improvement and Modernization Act of 2003 changed the way that Medicare pays physicians for Part B drugs and drug administration services. Before 2005, the Medicare payment rate for Part B-covered drugs was set at 95 percent of the average wholesale price, “which can be thought of as a manufacturer’s list price” (MedPAC, 2007, p. 4). Policy makers agreed that the payment rates for Part B drugs were too high, but some providers argued that the high rates for the drugs were needed to offset payment rates for administering the drugs that were lower than the costs of administration. Since 2005, physicians who provide Part B drugs to their patients are reimbursed for those drugs at 106 percent of the average sales price, which is computed as the average transaction price for all sales in the United States. The law provided that new biologics and single-source drugs (brand-name drugs with no generic version) would be paid based on an individually determined average sales price so that payment would not be coded or averaged with other products. At the same time that Congress reduced reimbursement for Part B drugs, it increased the payment to physicians for administering the drugs.
A 2007 report by the Medicare Payment Advisory Commission (Med-PAC) concluded that the change to the new payment system resulted in lower Part B expenditures for almost all covered drugs, largely due to lower prices ( MedPAC, 2007). For example, from 2004 to 2005, the drop in drug expenditures ranged from 1 percent for rheumatology Part B drugs to 52 percent for urology drugs.
Patients pay a general Part B deductible and then 20 percent of the Medicare-approved payment amount. A patient’s Part B coinsurance liability for medications is not capped and stays at 20 percent no matter how ex-
pensive the drug. However, as of 2006, just over 90 percent of beneficiaries had some form of public or private supplemental insurance that shielded them from Part B cost sharing requirements (MedPAC, 2010b).
Some drugs may be covered under Part B or Part D depending on the circumstances. The 2007 MedPAC report cited above stated that “physicians may work with patients to determine whether it makes more sense to use a drug that would [could] be covered under Part D or Part B,” depending on what out-of-pocket costs a particular patient might face (MedPAC, 2007, p. 17).
A MedPAC analysis (2009) found that the top six biologics covered by Medicare Part B accounted for 43 percent ($7 billion) of the total of $17 billion spent for Part B in 2007 for the approximately 650 drugs that are separately paid under Part B. Several of these products were approved as orphan drugs for at least one, generally several, indications. The analysis also found that biologics accounted for 6 percent ($3.9 billion) of Part D spending but that spending on such products was growing faster for Part D than for Part B.
Medicare Part D
Medicare Part D adds an outpatient prescription drug benefit to the Medicare program. All Medicare beneficiaries have access to the benefit. Although they are not required to enroll, beneficiaries who do not enroll during their initial eligibility period and who do not have equivalent alternative coverage will pay more if they enroll later. As of February 2009, more than 26 million beneficiaries were enrolled in a Part D plan (KFF, 2009b).
The Part D benefit is administered by private plans approved by CMS. Part D benefits are offered through stand-alone prescription drug plans and through Medicare Advantage plans that cover all Medicare benefits including medications. Congress also specified that drug coverage for all individuals dually eligible for Medicare and Medicaid would shift from the Medicaid program to the Part D benefit. As a result, all dual eligibles are now enrolled in private Part D plans.
Beneficiary Financial Responsibilities
In 2010, the average adjusted monthly Part D premium was $38.94 (Q1Medicare.com, 2009). Beneficiaries with low incomes and modest assets receive substantial financial assistance with Part D premiums and cost sharing. For example, full-benefit dual eligibles pay no premiums, pay relatively nominal fixed copayments per prescription, and are not subject to the deductible or the coverage gap described below.
Under the standard Part D benefit offered in 2010, most beneficiaries
pay a $310 deductible and then 25 percent coinsurance up to the initial coverage limit ($2,830 in total drug spending during the calendar year). In addition, beneficiaries are currently responsible for 100 percent of expenditures between $2,831 and $6,440 in total drug spending. This gap in coverage is sometimes referred to as the “doughnut hole.” Once his or her total drug spending exceeds $6,440, a beneficiary pays the higher of 5 percent of a drug’s cost or a copayment of $2.50 for a generic prescription or $6.30 for a brand-name prescription for the rest of the calendar year.
Plans have the option to deviate from this standard benefit by offering an actuarially equivalent benefit or by offering enhanced benefits. However, the vast majority of drug plans (80 percent) offered no benefits in the coverage gap as of January 2010, and most plans that did offer such benefits covered only generic drugs in the coverage gap (KFF, 2009b).
Under the Affordable Care Act, beginning January 1, 2011, Part D beneficiaries in the coverage gap will receive a 50 percent discount (absorbed by the manufacturer) off their plan’s negotiated price for brand-name drugs and biologics covered by the plan. Despite the discount, 100 percent of the negotiated price will count toward the beneficiary’s out-of-pocket expenses and thus toward the catastrophic coverage threshold ($6,440 in 2010). Over the subsequent 10 years, the beneficiary’s coinsurance for brand-name drugs in the coverage gap will drop annually or biannually, down to 25 percent by 2020, and the coinsurance for generic drugs will likewise drop to 25 percent.
Structuring Coverage: General
Within limits specified by the federal government, Part D plans have considerable flexibility in structuring formularies, imposing cost sharing requirements, and establishing procedures for managing drug utilization. Plans are required to cover at least one drug in each therapeutic class, although plans must cover “all or substantially all” medications in six protected classes: anticancer drugs, antidepressants, anticonvulsants, antipsychotics, HIV/AIDS drugs, and immunosuppressants.6 If a product is the only one available for treating a particular condition, Medicare generally requires plans to cover it.
For companies making orphan drugs or considering development of an orphan drug, several features of Part D plans could significantly affect beneficiary access and costs. These features—which include “tiered” cost
sharing, requirements for prior authorization of coverage and step therapy, and quantity limits—are explained below. More details about these features as they affect orphan drugs are included in Appendix C.
Structuring Coverage: Tiered Cost Sharing
Most Part D plans use tiered formularies. Generic drugs in tier 1 require the lowest copayment; brand-name drugs preferred by the plan in tier 2 require a somewhat higher copayment; nonpreferred brand-name drugs in tier 3 require a still higher copayment.7 Virtually all plans with tiered cost sharing also include a additional specialty tier to help manage spending for expensive specialty medications (KFF, 2009d). CMS guidelines specify that all drugs listed on the specialty tier must cost at least $600 a month (CMS, 2009a). Rather than paying a fixed copayment per prescription (e.g., $20) as is typical for less expensive drugs, beneficiaries must typically pay a percentage of the cost of medications in the specialty tier as coinsurance.8 For the 2010 plan year, the median coinsurance rate for medications in the specialty tier across plans is 30 percent (KFF, 2009d; GAO, 2010b). Although the incentives in tiered formularies for beneficiaries to use generic or preferred drugs can provide leverage for plans to negotiate discounts with drug manufacturers, both the lack of competition for many orphan drugs and the small number of users for these drugs, as noted above, weaken the negotiating position of plans.
For a drug with a price to patients of $4,000 or more per month, coinsurance payments at 30 percent (plus the deductible) would reach the “no-coverage” threshold of $2,831 in total drug spending in 2 to 3 months and would exceed the upper end of the coverage gap ($6,440 in total drug spending) in about 6 months. (This assumes that the patient has no other applicable out-of-pocket costs.) At that point, costs for each prescription would drop to approximately 5 percent (which, if a prescription was for a month’s worth of medication, would be approximately $200 in this example). Depending on how the initiation of treatment matched the start of a coverage year, that period of more generous coverage could last as long as 6 months before the coverage cycle started anew.
A recent GAO (2010b) study reported that high-cost drugs (i.e., those
eligible for the specialty tier) accounted for 10 percent of total Part D costs. It also reported that 55 percent of beneficiaries who used at least one specialty-tier drug exceeded the upper threshold of the coverage gap. In addition, more than 75 percent of prescriptions for specialty tier-eligible drugs were for subsidized beneficiaries such as dual eligibles who qualify for reduced cost sharing for these and other drugs and who are not subject to the coverage gap. To illustrate the impact of increasing drug prices, the report cited the 46 percent price increase for the drug imatinib (Gleevec) from approximately $31,200 per year in 2006 to $45,500 per year in 2009. The increase meant a corresponding increase in out-of-pocket costs for a nonsubsidized beneficiary (taking the drug for the entire year) from $4,900 to $6,300.
Structuring Coverage: Utilization Management
In addition to patient cost sharing features, Part D plans also employ a variety of utilization management strategies to control the use of drugs and overall costs as well as to promote medication safety in some instances. These include prior authorization, step therapy, and quantity limits. Other things being equal, a pharmaceutical company would expect less use of an orphan drug and lower profits if the drug were a target of the most stringent of these utilization controls.
Plans generally require enrollees to obtain prior authorization from the plan to secure coverage for certain medications, particularly higher-cost medications or drugs with particular safety concerns. The committee found no data on the extent to which plans approve or deny requests for prior authorization.
Plans also may employ what are termed step-therapy requirements for certain medications for which alternatives are available. Under step therapy, enrollees are required to try a lower-cost medication and document a poor response to that drug before coverage of a higher-cost medication would be granted.
Quantity limits (i.e., a limit on the amount of a drug that is covered during a fixed time period) are another commonly used utilization management tool. In general, more frequent prescriptions to obtain the same quantity of a drug mean more costs shifted to patients.
Coverage and Off-Label Use
FDA approval of a drug is for a specific indication or indications based on evidence of safety and effectiveness for each indication. These indications are described in the FDA-approved labeling for the product. FDA regulations restrict companies from promoting unapproved or “off-label” uses,
but FDA does not regulate the practice of medicine. Physicians may legally prescribe drugs for off-label indications. For example, a drug approved for use with a common disease may be used off-label for a rare condition, and physicians likewise may prescribe an orphan drug for either a common indication or a rare indication other than the indication(s) for which it has been approved. Some have expressed concern that some companies seeking orphan drug approval and marketing protection are really aiming at off-label use for a common indication and are inappropriately benefiting from the marketing protections attached to orphan drug approval (Fugh-Berman and Melnick, 2008; Wellman-Labadie and Zhou, 2009).
In addition to cost concerns, off-label use raises concerns about patient exposure to drugs that have not been determined to be safe and effective for uses other than those approved by FDA. At the same time, such use may provide options for patients for whom FDA-approved products are limited or nonexistent because physicians can try a medically plausible treatment approach to a disease (Kocs and Fendrick, 2003; Gillick, 2009). If case reports suggest promise from such an off-label use, then this may prompt companies or others to undertake controlled studies to support approval by FDA of a new indication. Alternatively, because systematic research could contradict the promising case reports and thereby curtail off-label sales, companies may choose not to pursue further studies.
Studies have indicated that off-label use is common in oncology (see, e.g., Kocs and Fendrick, 2003; Eastman, 2005; Abernethy et al., 2009). Off-label use is likewise common in pediatrics. Indeed, the lack of testing of drugs for use with children prompted legislation to encourage and in some cases require such testing (see Chapter 3). One study of outpatient prescribing patterns reported that of more than 500 medications covered in the study, 21 percent of uses were off-label and most of these off-label uses (73 percent) lacked clinical evidence of efficacy (Radley et al., 2006). Even when off-label uses are backed by research, companies may not wish to incur the costs of pursuing FDA approval unless they expect such approval to encourage further use that will offset those costs.
The initial CMS regulations for the Part D program denied coverage of medications for off-label uses unless the prescribed use was supported by one of three specific medical compendia (Le Masurier and Edgar, 2009).9 Currently, off-label drug treatments for cancer are covered if they are listed in any CMS-approved compendium that is used to determine coverage of
off-label uses of cancer drugs under Part B. One concern is that questions have been raised about the independence of compendia compilers, the degree to which they cite current evidence (or any evidence), the quality of their methods and their assessments of evidence, and the potential for official acceptance of such compendia to discourage research aimed at FDA approval of off-label uses (Tillman and Gardner, 2004; Abernethy et al., 2009; Butcher, 2009; Mitka, 2009; Sox, 2009; see also ASHP, 1992; Gillick, 2009).
To inform future off-label coverage determinations, CMS commissioned a technology assessment from the Agency for Healthcare Research and Quality (AHRQ) to summarize the process by which anticancer drugs are added to various compendia and to identify methods used to collect evidence for listed drugs and biologics and their indicated uses (Abernethy et al., 2009). The assessment covered six compendia and a sample of 14 anticancer combinations that were selected to include newer and older agents, common and rare cancers, and biologics and drugs. Among the findings was that there was little agreement in the evidence regarding efficacy cited by the compendia and that the compendia were discordant on whether they discussed adverse effects among patients with specific cancers. Moreover, when compendia did not include off-label indications, the analysts could not determine whether a particular omission reflected a conscious editorial decision following the evaluation of available evidence or whether the available evidence was not identified and evaluated. The authors observed that although they could not generalize to other disease areas, the compendia’s performance might be expected to be highest in oncology, given their importance for reimbursement. The authors also pointed out the major challenges of managing a near-continuous systematic review of large numbers of drug uses not approved by FDA. They also noted that FDA itself was not authorized or prepared to undertake such a review.
For rare diseases, the volume of drugs and uses is obviously much smaller but so is the research to support evaluations of off-label use. The growing databases from Part D claims could, when linked to Medicare hospital and physician claims data, be a resource for studying the nature and outcomes of some off-label use of orphan and other drugs for patients with rare conditions.
Part D Plans and Drug Prices
Congress has not provided CMS itself with the authority to negotiate prices with pharmaceutical manufacturers. However, as noted above, to the extent that private health plans, including Part D plans, are able to “move market share” across drugs in a class using such financial incentives, then plans have the potential to negotiate sizable rebates or discounts from
manufacturers. For many expensive drugs, including many orphan drugs, insurers may have little leverage in negotiating price discounts.
Given the introduction of Part D just 4 years ago, only limited empirical evidence has accumulated on its impact on drug prices. A recent study by Duggan and Morton found that Part D led to a decrease in the average price for brand-name drugs and an increase in overall utilization of Part D drugs among Medicare recipients (Duggan and Morton, 2006). They estimate that each percentage point increase in the pre-Part D Medicare market share for a given drug is associated with a 1.2 to 1.4 percent decrease in a drug’s average price relative to other drugs. However, Frank and Newhouse (2008) found some evidence that the shift from Medicaid to Part D of drug coverage for dual eligibles resulted in higher drug prices for this population. (These analyses did not specifically consider orphan or specialty-tier drugs.)
An analysis commissioned by MedPAC reported that prices for Part D drugs rose 11 percent between January 2006 and December 2008 (based on national drug codes as the unit of analysis) (MedPAC, 2010a). However, after taking into account the substitution of generic for brand-name drugs (which is encouraged by Part D plans), the analysis found Part D prices declined by 3 percent over the same period.
Analysis of Part D Plan Coverage of Orphan Drugs
Appendix C presents the results of a commissioned analysis of Part D plan coverage of orphan drugs as reported in the January 2010 CMS Prescription Drug Plan, Pharmacy Network, and Pricing Information Files. For drugs that are not covered by Medicare Part B (or, rarely, Part A) and that thus are eligible for Part D coverage, the analysis found that the great majority are covered by more than half of Part D plans (Table 6-1). Of the handful of orphan drugs that were not covered by any plan as of January 2010, several of these have now been added to the formularies of at least one plan. In addition, several drugs that were included in the analysis because they were not in the CMS price list for Part B-covered drugs have since been added to that price list.
There are some differences in coverage of orphan drugs across different types of Part D plans. For example, overall, orphan drug coverage seems to be somewhat more generous among national stand-alone Part D plans than among nonnational stand-alone plans. For orphan drugs, 27 percent are covered by fewer than half of nonnational plans, while only 9 percent are covered by fewer than half of national plans. Still, Medicare beneficiaries who rely on an orphan drug should be able to find a Part D plan that covers it.
Although nearly all orphan drugs are covered by at least half of Part
TABLE 6-1 Coverage of Part D-Eligible Drugs by Type of Medicare Prescription Drug Plan
D plans, significant limits of the kinds described above typically apply. Almost half (46 percent) of orphan drugs are included in specialty tiers by 50 percent or more of stand-alone Part D plans. One-third of orphan drugs were subject to prior authorization requirements before coverage is granted by 50 percent or more of stand-alone plans.
The Medicaid program, which is jointly financed and administered by the federal and state governments, covers health and long-term care services for approximately 20 percent of Americans under age 65, including eligible low-income children, parents, and individuals with disabilities (EPI, 2009). It also covers eligible low-income individuals over 65 who are covered by Medicare. Under the Affordable Care Act, state Medicaid programs will
be required to extend eligibility to all individuals with income up to 133 percent of the federal poverty level. One estimate is that this eligibility expansion could increase Medicaid enrollment by approximately 16 million individuals (Holahan and Headen, 2010).
Although states are not required to cover prescription drugs in their Medicaid programs, all 50 states and the District of Columbia offer a drug benefit (KFF, 2008). As noted above, prescription drug coverage for individuals who are dually eligible for Medicare and Medicaid shifted from Medicaid to the Medicare Part D program as of January 2006. Other Medicaid beneficiaries still receive drug coverage from their state Medicaid program. Medicaid programs typically require copayments for each prescription, which generally range from $0.50 to $5 per prescription depending on the type of drug (generic versus brand name, preferred versus nonpreferred) and the state (KFF, 2008).
Under the Medicaid Rebate Program, manufacturers are required to have a rebate agreement with the Secretary of Health and Human Services in order for states to receive federal Medicaid funding for outpatient prescription drugs dispensed to Medicaid patients. Until 2010, the rebate had been 15.1 percent of the average manufacturer price (AMP, or the average price at which the manufacturer sold it) or the difference between AMP and the best price offered by the manufacturer within the United States, whichever is greater (CMS, 2009b). Under the Affordable Care Act, the minimum rebate for innovator drugs will increase in 2010 to 23.1 percent of AMP, with the exception of clotting factors and drugs with only pediatric indications for which the minimum rebate will be 17.1 percent of AMP. (The rebate for noninnovator drugs will increase from 11 to 13 percent of AMP.) Under the new law, Medicaid rebates must be paid on outpatient drugs dispensed to enrollees of Medicaid managed care plans (close to 70 percent of all Medicaid enrollment), which was not the case before 2010.
Overall, the rebate provisions make orphan drugs more affordable for state Medicaid programs, although very expensive drugs remain very expensive. The committee found no analysis specific to orphan drugs, but some evidence suggests that this rebate approach results in much lower prices for Medicaid than for other payers in the market. For example, the House Committee on Oversight and Government Reform estimated that Medicaid pays prices that are about 30 percent lower than prices paid by Medicare Part D (Outterson and Kesselheim, 2009). However, a study by Duggan and Morton (2006) found that drugs sold disproportionately to Medicaid beneficiaries have higher prices than otherwise similar drugs. Because the Medicaid rebate is based on prices paid for these drugs in the private sector, manufacturers have an incentive to increase prices charged in the private sector, thereby distorting both the private market price and the Medicaid price.
In response to increasing prescription drug utilization and expenditures, states have adopted a variety of cost containment approaches over the past decade. For example, according to an analysis of the 50 states and the District of Columbia, 44 states have state maximum allowable cost programs that set maximum reimbursement levels for generic and multisource brand-name drugs, and 26 states were members of multistate purchasing coalitions intended to increase negotiating power over price with pharmaceutical manufacturers (Smith et al., 2009). Forty-four states negotiate supplemental rebates in addition to rebates negotiated through the national drug rebate program for Medicaid. Almost one-third (16 states) limit the number of prescriptions that are covered per enrollee, and 46 require prior authorization before granting coverage of specific medications as of 2009. Although not specific to orphan products, these policies would affect orphan drugs and the patients who use them.
PRIVATE HEALTH PLAN COVERAGE OF FDA-APPROVED DRUGS
As of 2008, approximately 65 percent of nonelderly individuals had private health insurance (KFF-SHF, 2010). This percentage is expected to increase substantially in the next few years when subsidies for the purchase of private insurance plans become available for lower-income individuals under the Affordable Care Act.
Virtually all individuals (98 percent) with employer-sponsored insurance currently have a prescription drug benefit (KFF, 2009c). Employer-sponsored plans vary substantially with respect to cost sharing requirements, formulary breadth, and utilization management requirements. Many of the practices now found in Medicare Part D plans were initially devised for employment-based plans.
According to a survey for the Kaiser Family Foundation, tiered formularies are common among employer-sponsored plans. More than three-quarters (78 percent) of individuals enrolled in these plans face a formulary with three or more tiers (KFF, 2009c). Average cost sharing requirements per prescription have increased steadily over the past few years. The majority of plans require copayments for each prescription filled rather than coinsurance payments per prescription. In 2009, average copayments were $10 for generic drugs in tier 1, $27 for preferred brand-name drugs in tier 2, $46 for nonpreferred brand-name drugs in tier 3, and $85 for drugs in tier 4. A minority of plans required coinsurance rather than copayments for one or more tiers. In 2009, 29 percent required coinsurance for tier 4 drugs, and the average coinsurance rate was 31 percent. A much smaller subset of plans (6 to 10 percent) required coinsurance for medications in tiers 1, 2, and 3.
Traditionally, a substantial proportion of health plans have limited the
total amount the plan would pay for a given enrollee over the course of his or her lifetime, often referred to as a lifetime spending maximum. In 2009, 16 percent of those with employer-sponsored coverage had a lifetime maximum between $1 million and $2 million, and another 43 percent had a lifetime maximum of $2 million or more (KFF, 2009c). An adult who receives twice-monthly injections of alglucosidase alfa (Myozyme) for Pompe disease could run up costs of $300,000 a year just for the drug, and the drug can have serious side effects that require hospitalization and additional expenses. Such a patient could reach a $1 million cap fairly quickly. (By way of comparison, total first-year costs for a heart transplant for the year of the transplant could run $800,000, but costs in subsequent years—assuming no serious complications—would be much lower, perhaps $20,000 to $40,000 [UNOS, 2010].) Only a minority (10 percent) of private health plans had an out-of-pocket maximum specifically for prescription drugs in 2009, which limits an enrollee’s financial risk for medication costs.
Effective in 2010, the Affordable Care Act prohibits individual and employer health plans from setting lifetime limits on the dollar value of coverage, and it permits annual caps on coverage only as allowed by the Department of Health and Human Services. The law also prohibits plans from canceling coverage because an individual develops health problems. Effective in 2014, the law provides an array of measures to expand access to insurance, one of which will prohibit insurers participating in newly created insurance exchanges from refusing coverage to people with medical problems and varying premiums based on health status. These and other provisions should benefit individuals who use high-cost orphan drugs, although many details remain unclear. For example, private plans could restrict coverage of drugs used by high-cost patients, unless regulations restrict that strategy.
Private health plans vary in their policies and practices with respect to off-label use of prescription drugs. Some conduct evidence reviews for certain drugs (see, e.g., Monroe et al., 2006), and some have set forth criteria for when off-label use will be considered (RegenceRx, 2010; Wellmark Blue Cross Blue Shield, 2010). An informal review of plan policies for a few orphan drugs likewise showed variation. Some excluded one or more of the drugs on the basis that other alternatives are preferable, some required prior authorization, and a few covered the drugs without restriction except for specialty-tier listing.
OTHER MEANS OF FINANCIAL ACCESS TO ORPHAN DRUGS
Even if drug companies expect that they can set profitable prices for a drug for a rare condition and anticipate that they will have a sufficient market of mostly insured patients able to pay those prices, they may also judge
it desirable or prudent to provide assistance in some form to patients who cannot afford the drug. A number of companies that have set high prices for orphan products have established some kind of assistance program for patients without insurance (e.g., without Medicare Part D coverage) or for individuals with insurance who face high out-of-pocket costs.10 Some companies have also established programs to help patients and families understand and navigate health plan requirements and procedures to secure payment for an expensive drug. Companies presumably factor the cost of assistance programs into their economic projections for a drug and then into the price of an approved drug. In this way, public and private health plans and insured individuals who pay for the drug support some of the cost of company assistance.
Company assistance programs may require considerable financial information from individuals seeking assistance, for example, tax returns, bank statements, and W-2 forms. Assistance may be restricted to people who have no insurance, and programs typically set income and asset limits (e.g., income up to some percentage above the federal poverty level). Types of company assistance may include
providing a supply of the drug at no or reduced cost for 3 months or some other defined period, after which time patients and families must seek a means of continued access now that use of the product has been initiated;
assisting with the cost of copayments or other cost sharing requirements for patients with insurance coverage; and
supplying information to patients and families about Medicaid eligibility, private charities, and other possible routes of financial aid.
A survey by Choudry and colleagues (2009) of 165 company assistance programs (not limited to orphan drugs) found considerable variability across programs. They reported that half the programs would not disclose their income eligibility criteria, and very few (4 percent) disclosed how many patients the programs had helped.11
The National Organization for Rare Disorders administers financial
The American Society of Health-System Pharmacists includes on its website a list of programs that may help patients get assistance from drug manufacturers (ASHP, 2010). In addition to other resources, an organization called NeedyMeds provides a list of programs and companies, some of which are explicitly identified as having no program (http://www.needymeds.org/program_list.taf).
A 2001 study by the Office of the Inspector General at the Department of Health and Human Services examined the implementation of the Orphan Drug Act (OIG, 2001b). It reported that roughly 3 out of 4 of the 36 companies that it contacted said they either had some kind of assistance program or planned to have one if FDA approved their products.
assistance programs for several medical product companies in connection with at least one of their products (information available at http://www.rarediseases.org/programs/medication). It also has several other programs of assistance for a number of mostly rare conditions, including infantile spasms, Hunter syndrome, and paroxysmal nocturnal hemoglobinuria.
In addition to company programs, advocacy groups for rare diseases and other nonprofit programs may assist some patients and families who lack insurance or cannot afford the cost sharing requirements of their health plan. The smaller the group, the more difficult it is likely to be for it to provide assistance. Another option for some individuals is the Patient Advocate Foundation Co-Pay Relief Program (http://www.copays.org/). It offers financial support to qualified insured patients, including Medicare Part D beneficiaries, who are being treated for one of 21 conditions, a few of which (e.g., pancreatic cancer and multiple myeloma) are rare.
Some families themselves or their relatives and friends create fund-raising efforts, for example, to raise enough money for a transplant or to help with costs for a child being treated for a brain tumor. It is doubtful that these kinds of activities factor into company decisions about product development.
In the future, an expansion of access to health insurance and the removal of certain limitations on coverage may reduce but are unlikely to eliminate the role of company assistance programs. Some individuals will remain uninsured, and some of those with insurance will continue to have difficulty with out-of-pocket payments.
PUBLIC AND PRIVATE HEALTH PLAN COVERAGE OF CERTAIN COSTS IN CLINICAL TRIALS
In some cases, health plans may cover certain costs of care for patients involved in clinical trials, thus reducing the burden on participants in the trial and potentially easing recruitment challenges for sponsors, including sponsors of trials of orphan drugs. The legislation that created Medicare provided generally that payment was to be limited to items or services that were “reasonable and necessary” for the diagnosis and treatment of illness or injury or to improve the functioning of a malformed body member” (42 USC 1395y). Historically, those administering the Medicare program interpreted the terms “reasonable and necessary” to mean that a service or item must be safe and effective, medically necessary and appropriate, and not experimental in order to qualify for reimbursement. Medicare coverage was typically denied for drugs or devices being studied under an investigational device exemption (see Chapter 7) or an investigational new drug application (see Chapter 3) that had not yet been approved or cleared by FDA.
Based on directives in an Executive Memorandum in 2000, Medicare
now covers, subject to certain qualifications, routine care costs for patients in therapeutic trials to support FDA approval of a drug or device as well as costs of medical complications arising from participation in such trials (CMS, 2000). The cost of the investigational product is usually not covered for these trials (but see Chapter 7 for a discussion of policy on medical devices). In addition, beginning in 2004, CMS may pay for some specific new items or services for which evidence is inadequate on the condition that additional patient data be provided to supplement standard claims data, a process termed “coverage with evidence development” (CMS, 2006, 2007). Many state Medicaid programs cover routine patient care costs in clinical trials under policies similar to those for Medicare (ACS, 2010). In addition, approximately 30 states have mandated that private health plans cover such costs (NCI, 2009b).
The Affordable Care Act of 2010 provides that health plans may not, in general, deny coverage for routine patient costs for items and services provided in connection with participation in a clinical trial for cancer and life-threatening conditions and may not discriminate against individuals based on their participation in clinical trials. The law explicitly mentions coverage for trials conducted under an Investigational New Drug application reviewed by FDA. It does not mention trials of devices under investigational device exemptions, although no language otherwise excludes coverage of routine costs associated with medical device trials. Health plans are not required to cover the cost of the investigational item.
As discussed in Chapter 3, the Orphan Drug Act provides significant incentives to companies to develop drugs for rare diseases. Translating these advantages into sales and profits depends on the availability of buyers. Fortunately for companies and individual patients, many of those in need of expensive orphan drugs have found that if they are insured, their health plans have, in general, been willing to cover the drug, usually at a price unilaterally established by the company. The great majority of orphan drugs eligible for Part D coverage are covered by more than half of Part D plans, and very few are covered by no plan. Plans, however, typically require that patients pay a significant share of the cost of expensive drugs, and many plans impose prior authorization requirements that could limit access for certain orphan medications.
Notwithstanding the generally positive picture for orphan drug coverage by Medicare Part D plans, companies keep an attentive eye on public and private health plan policies and coverage trends, recognizing that escalating health care costs put Medicare, Medicaid, and other health programs under pressure to take actions to manage costs, including costs for prescrip-
tion medications. Although a number of provisions of the Affordable Care Act of 2010 should reduce the burden of high medication costs for patients (if the provisions are not repealed), the escalating costs of health care will keep or increase the pressure on public and private health plans to transfer more costs to patients within the boundaries of the law. To the extent that personalized medicine and other developments lead to an increase in the proportion of drugs that win approval as orphan drugs and an increase in the share of insured patients who use orphan drugs, concerns specific to orphan drugs can be expected to expand to more drugs and more patients.
Better information about some aspects of health plan policies could help decision makers assess how policies may affect access to orphan drugs. In particular, little is known about the application of prior authorization requirements to orphan drugs. Such requirements could have negative effects by restricting or delaying access to needed drugs. They could also have benefits if they improve physician adherence to evidence-based guidelines for effective drug use and reduce unnecessary or even harmful use of expensive drugs. It would be helpful to have some evidence of what is actually occurring with these requirements in Medicare Part D plans. The recommendation below focuses on Medicare and Medicaid. More information on the policies and procedures of other health plans would also be desirable to gauge the effects on people who depend on orphan drugs.
RECOMMENDATION 6-1: The Centers for Medicare and Medicaid Services or the Medicare Payment Advisory Commission should study how the implementation of prior authorization requirements by Medicare Part D and state Medicaid plans affects beneficiary access to orphan drugs. The findings should guide recommendations and actions to improve policies and practices for the Part D program.
A second area the committee identified for attention involves payment for off-label use of orphan drugs or for off-label use of common drugs for patients with rare conditions. Indiscriminate coverage of off-label uses has the potential to harm patients as well as waste scarce resources. Indiscriminate exclusion of such uses likewise has the potential to harm patients and produce a backlash by policy makers, who have already required selective coverage of off-label uses under Medicare Part B and Part D. This coverage has been linked to information provided in compendia prepared by private companies with little public oversight or evaluation of their practices or analyses. The extent to which these compendia cover off-label uses relevant to patients with rare conditions is unknown, and the quality of such discussions as exist is likewise unknown.
Recommendations about compendia generally are beyond the scope of this report. However, the committee believes that the creation of an
evidence-based compendium focused specifically on off-label uses of drugs for rare diseases could inform clinicians, health plans, and potentially patients and families. Such a compendium is not likely to be feasible for commercial publishers but could be undertaken by a public agency such as AHRQ that has experience in similar analyses. Based on experience with the U.S. Pharmacopeia, which developed one of the early compendia but discontinued it because sales did not cover expenses, the estimated potential cost of a pilot project to develop and update a pilot compendium is within the range of current AHRQ grants (Dr. Roger Williams, CEO, U.S. Pharmacopeia, June 2010, personal communication to Carolyn Asbury [Committee member].)
RECOMMENDATION 6-2: The Agency for Healthcare Research and Quality or a similar appropriate agency should undertake a pilot project to develop an evidence-based compendium to inform health plan decisions on both orphan and nonorphan drugs that may have indications for rare conditions that have not been evaluated or approved by FDA.
Some of the issues that such a pilot effort would confront include determining a focus (e.g., rare cancers, rare metabolic disorders), establishing criteria for evaluating research that involves small numbers of participants and nontraditional research designs, and exploring the use of Medicare or Part D claims data for analyses to supplement the review of published studies. Depending on its experience, AHRQ or another agency could propose a strategy for updating or expanding the compendium.
In the process of developing the compendium, analysts may also identify directions for future research on specific drugs to demonstrate efficacy, side effects, or optimum dosage. If not undertaken by industry, such studies might be supported through National Institutes of Health awards, FDA orphan products grants, AHRQ grants to its Centers for Education and Research on Therapeutics, or other grant programs.