|
Medicaid Policies To Contain Psychiatric Drug Costs
This study reviews Medicaid policies to restrict access to psychiatric medications. Policies on prior authorization, preferred drug lists, limitations on the number of prescriptions, fail-first requirements, and use of generics are reviewed. All states apply one or more of those policies to medications for mental illness, and many apply several. A large number of states have legislated exemptions from those policies for certain medications, particularly antipsychotics and antidepressants. Other psychiatric medications are less well protected. Some states appear to restrict access severely. Questions have been raised as to whether these policies actually save money in the long term.
Over the past few years, states have been forced to consider drastic reductions in spending as they have faced some of the largest deficits in half a century.1 Because of its size, Medicaid is one of the primary targets for cuts. In 2003, forty-six states made changes to reduce the costs of their Medicaid outpatient drug benefits.2 Although the total costs of outpatient drugs accounted for only 11.1 percent of Medicaid spending in 2000, there have been double-digit increases in yearly Medicaid drug spending since 1995 (and more than 18 percent per year since 1997).3 Recently, these increases have been twice as high as increases in total Medicaid spending.4 This has led states to adopt controversial policies to control these costs.5 Many advocates for low-income populations express concern about the impact of these policies on the health of the Medicaid population. The outpatient prescription drug benefit is an optional Medicaid benefit that all states have adopted. Federal law requires states to cover all drugs of manufacturers that have rebate agreements with the federal government if the state wishes to qualify for these rebates. However, states have great flexibility to manage the benefit. They may, for example, limit access through prior authorization (PA) requirements, require or encourage the use of generics, limit the number of prescriptions filled, require that enrollees fail on one medication before another is prescribed, and develop lists of preferred drugs to which there is ready access. Since states may not have closed formularies, any drug for which access is restricted must be available when shown to be medically necessary for a particular patient. Psychiatric medications are among Medicaid’s most costly and commonly prescribed drugs.6 In 1998 antidepressants and antipsychotics accounted for nearly 9 percent of Medicaid prescriptions and nearly 19 percent of Medicaid drug reimbursements.7 Limiting access to psychiatric medications may, however, result in poor-quality care and other costs. Although Medicaid law requires that all drugs be available when medically necessary, policies such as preferred drug lists (PDLs) and fail-first or PA requirements are similar in effect to formularies. Physicians have indicated that they believe formularies have a negative effect on the quality and efficiency of medical care.8 Moreover, a recent review of studies regarding restrictive drug policies in Medicaid found that targeting essential drug classes with heterogeneous patient responses and side effects could reduce appropriate care, adversely affect health status, and cause shifts to more costly types of care.9 For example, a three-prescription limit on psychiatric medications adopted in New Hampshire in 1990 resulted in reductions in the use of psychiatric medications but led to large increases in the use of emergency mental health services and partial hospitalization for people with schizophrenia. The average increase in service costs per patient was estimated to exceed the savings by a factor of 17.10 In the summer of 2003 the Bazelon Center for Mental Health Law reviewed state policies with respect to access for six classes of drugs prescribed for psychiatric conditions: antipsychotics (looking separately at both typicals and atypicals), antidepressants (looking separately at selective serotonin reuptake inhibitors, or SSRIs, and other antidepressants), anticonvulsants, anti-Parkinsons, stimulants, and sedatives/hypnotics/anxiolytics. We examined Medicaid agency Web sites for information on state policies and supplemented that data with information from the Kaiser Commission on Medicaid and the Uninsured and the National Conference of State Legislatures.11 We then surveyed Medicaid pharmacy directors in all states by mail, with follow-up telephone contacts, to verify our information and to collect new data regarding policies for covering medications in these six drug classes. Forty-seven states responded, for a 92 percent response rate. We were unable to verify the information collected from the remaining few states.
All states restrict access for one or more of the classes of drugs studied (Exhibit 1  ).12 The most widely adopted policies are prior authorization and encouraging or mandating the use of generics. The use of preferred drug lists (PDLs) is the third most frequently implemented policy.
Exhibit 2 shows how many states have adopted these policies and how many protect one or more groups of psychiatric drugs under them (note that the fail-first policies studied were those that applied only to psychiatric medications).
A few states exempt certain groups of Medicaid beneficiaries from their policies. For example, Kansas exempts beneficiaries who need medications to treat schizophrenia, depression, and bipolar disorders from all restrictions examined by this study. Colorado exempts medications used to treat "biologically based" mental disorders from its use of generics rule, and New Mexico exempts use of antipsychotics from PDL requirements for beneficiaries with "severe mental illness." Pharmacy restrictions. States tend to adopt several different pharmacy restrictions. Forty-three states adopted two or more of four policies studied (those policies that relate to all drug groups: PA, PDL, generic mandate, and prescription limit).13 Four states adopted all four of these restrictions, and eighteen states adopted three. All states had at least one restriction. Some policies apply to all psychiatric medications in a group, while others are targeted to individual drugs. Exemptions from PDL and prior authorization. The drug groups most frequently exempted from PDL and PA policies are antipsychotics (including atypicals) and SSRIs. These are less often exempted from policies encouraging or mandating use of generics or requiring prescription limits. Stimulants and sedatives/ hypnotics/anxiolytics are somewhat less likely to be exempted from most policies. Fourteen states exclude all six psychiatric drug classes from PA policies. Sixteen exclude some psychiatric medications, while nineteen states have no mental health drug exclusions. Virtually all of the states with exemptions exempt antipsychotics (both typicals and atypicals) and anticonvulsants; a substantial number exempt anti-Parkinsons, SSRIs, and other antidepressants. Stimulants and sedatives/hypnotics/anxiolytics are somewhat less likely to be exempted.
Some states with PDLs have created specific exemptions for some classes of drugs (Exhibit 3
Generic mandates. Policies to mandate or encourage the use of generics are widespread. States may encourage the use of generics (such as by requiring physicians to write "brand necessary" on the prescription) or require it (although PA must permit some exceptions). States are more likely to exempt anticonvulsants, anti-Parkinsons, typical and atypical antipsychotics, SSRIs, and other antidepressants from the generic mandate. They are less likely to exempt stimulants. Prescription limits. Monthly limits on prescriptions range from three to eight; one state has a yearly limit of forty-three prescriptions. Sometimes limits apply to brand-name drugs only. Three-quarters of states with prescription limits have processes for requesting prescriptions above the cap. Six states with limits exclude one or more psychiatric medications from the limit. Two states have special limits on specific psychiatric medications: Utah limits beneficiaries to two atypical antipsychotics per month, and Maine limits them to five anticonvulsants per month. Fail-first policies. States have adopted fail-first policies either for a specific medication or for an entire medication group. Seven states require failure on the generic before any brand-name drug can be used. Three states require failure on another antipsychotic before Clorazil may be used, and two have a similar policy for Zyprexa. Two states have fail-first policies with respect to SSRIs. "Grandfather" policies. Five states have adopted "grandfather" policies that exempt established users of specific psychiatric medications from newly adopted PA policies, and eight states have such clauses for PDLs. Policy adoption processes. Policies have been adopted through varying processes in different states. We found that in some states certain policies (particularly PDLs) were mandated by the legislature, while in others they were demanded by senior executive branch or Medicaid agency staff. Although costs were a concern, some Medicaid officials reported concerns also about the effects of restricting access to drugs for those with serious mental illnesses, such as schizophrenia.
Missouri. Some states have approached the use of psychiatric medications in Medicaid with a different set of policies. Missouri, for example, has a program to improve physicians’ prescribing patterns. Missouri Medicaid worked with the mental health authority and an advisory council comprising physicians, other providers, and representatives of consumer groups. In Missouri, Medicaid pharmacy claims data are reviewed to identify prescribing and usage trends; claims that show possible deviation from recommended best practices are identified, and these prescribers are sent a letter and educational materials about best practices; claims data are monitored to determine whether the provider makes changes; and additional interventions (such as peer-to-peer consultation) are provided as needed. Areas for review include multiple medications in the same therapeutic class; excessive dosing and inadequate dosing; no timely refill (early discontinuation); rapid switching of medications; and patients with two or more physicians prescribing the same medication. According to Missouri officials, preliminary data show that changes in practice are occurring. In FY 2004, the Missouri Medicaid program saved $7.7 million in psychiatric medication costs, as compared with costs expected to be incurred. According to an independent audit, the changes in practice that resulted from this program significantly contributed to these savings.14 Oregon. Oregon has adopted a PDL that allows practitioners much greater discretion, encouraging but not mandating prescribing practices. The goal of the practitioner-managed Prescription Drug Plan for the Oregon Health Plan (Medicaid) is to allow enrollees to receive the most effective prescription drug available at the best possible price. The formulary developed under this plan is voluntary, and practitioners are allowed to prescribe any drug that they indicate is medically necessary and the most effective available. Oregon requires use of generics unless the practitioner prescribes otherwise and an exception is granted by Medicaid. The state legislature eliminated the restriction on the type of legend drugs that a practitioner may prescribe. Psychiatric drugs are excluded from the voluntary PDL. Other examples. Other states have more directly addressed the cost of medications by increasing their Medicaid program’s purchasing power. Some states (such as Texas) have combined Medicaid with other state agency or state employee drug purchasing programs or included a larger population of uninsured people within the Medicaid drug benefit (Maine). Others have joined to create interstate purchasing pools. The federal government recently approved a purchasing pool involving eight states (Alaska, Hawaii, Maryland, Michigan, Minnesota, Nevada, New Hampshire, and Vermont) to negotiate discounts with drug manufacturers.15
This study found widespread use of policies to restrict access to psychiatric medications in state Medicaid programs, although it also found a large number of exemptions from those policies for certain drugs or drug classes. Our findings suggest that states’ broad flexibility to restrict access to medications has led to many restrictions on physician’s’ ability to prescribe psychiatric medications. Although we did not explore these policies’ specific impact on quality of care, focusing only on a drug’s expense—as some of these policies do—is "as biased as to focus only on its efficacy." Instead, recognizing the impact of medications on social and economic measures may lead to a more comprehensive and accurate understanding of their value.16 Limited information about treatment effectiveness for individual patients and considerable heterogeneity in the population make it difficult to predict which psychiatric medication will be effective for any one person.17 For example, limits on the use of some antipsychotics are contrary to practice guidelines that support the use of atypical antipsychotics as the first choice for treatment of schizophrenia.18 Moreover, many people tolerate some side effects more easily than others do. When prescribing atypical antipsychotics, physicians are advised to monitor closely for side effects and to screen for specific conditions resulting from patients’ medical/metabolic profiles.19 Side effects can include seizures, low blood pressure, irregular heartbeat, diabetes, tardive dyskinesia (a neurological disorder characterized by involuntary movement), and other extrapyramidal symptoms (neurological side effects). Some state policies might be particularly problematic. As demonstrated in New Hampshire, denying access to a full array of antipsychotics can increase other Medicaid service costs resulting from ineffective treatment. Also, limitations on stimulants may be harmful to children and youth with attention deficit hyperactivity disorder (ADHD), while others may be harmed by restrictions on barbiturates or anticonvulsants, which are often prescribed as supplements to other psychiatric medications. In efforts to circumvent these adverse outcomes, some states have crafted exemptions that are applicable in certain cases—for example, stimulants for people under age eighteen or established users of antipsychotic drugs. It is also important to note that the PA mechanisms are different from those of a drug cap, and, therefore, the effects of any one restriction (or the combination of restrictions) may be quite different. The policies discussed in this paper generally rest on an assumption that patients in critical need of a particular medication will nonetheless have access through PA (which can be a stand-alone policy or part of a PDL, generic mandate, or prescription limit). However, there are concerns about PA’s impact on quality of care and patient outcomes. For example, prescribers may be reluctant to engage in a PA process viewed as a "hassle" and thus do not prescribe the appropriate medication for a specific patient. Or low-income patients, unaware of PA requirements, may arrive at the pharmacy to find that they cannot obtain their drug and leave empty-handed, thus contributing to problems of nonadherence to treatment. Available studies on the effect of PA in Medicaid have generally failed to demonstrate cost-effectiveness or an offset in other spending.20 Severely limiting the number of prescriptions may be particularly problematic for people with serious mental illnesses, many of whom require treatment with multiple psychiatric medications. Moreover, these people have high rates of co-occurring serious and often chronic physical disorders such as diabetes, hypertension, cancer, and HIV, as well as other behavioral disorders. Fail-first policies, particularly when they do not comport with professional practice guidelines, could result in an extended episode of active psychosis before symptoms are controlled, leading to the loss of employment, housing, and social supports. Fail-first policies require trial and failure on one drug (usually cheaper) before approval to use a more costly drug can be obtained. The Texas Medication Algorithm Project (TMAP) presents evidence-based algorithms for managing schizophrenia. TMAP strongly recommends that prescribers initiate psychopharmacologic treatment with a second-generation antipsychotic (SGA) versus a first-generation antipsychotic (FGA). A meta-analysis of the efficacy of SGAs found that the effect sizes of clozapine, risperidone, and olanzapine were significantly greater than those of FGAs, supporting their use as the first line of intervention.21 State policies reported in this study clearly reduce access to certain psychiatric drugs for the Medicaid population. Earlier studies on such approaches suggest that this may have adverse affects on the quality of care and overall cost of treating mental illnesses. As such, they have the potential to harm beneficiaries with mental disorders. Because such people have little assistance to help them negotiate their rights and do not have the resources to pay for noncovered drugs themselves, it is critical that states consider the impact on beneficiaries’ health as they design and adopt policies to control their spending on psychiatric drugs.
Chris Koyanagi is policy director at the Bazelon Center for Mental Health Law, a legal advocacy organization in Washington, D.C. Sandra Forquer is a clinical associate professor in the Department of Psychiatry, University of Colorado Health and Sciences, and vice president for education services at Comprehensive Neuroscience Inc. in Colorado Springs. Elaine Alfano is a policy analyst at the Bazelon Center. This research was supported by Comprehensive Neuroscience Inc.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Study Findings
Different Approaches
