I. Executive Summary II.Background Introduction Pharmacy Costs Increasing Faster than Other Health Care Costs Medicaid Pharmacy Benefits Concerns about Access to Psychiatric Medications Prior Studies on Access to Antipsychotic Medications III. A Study Of Medication Coverage For Physical Health And Mental Health Study Design General Pharmacy Benefits and Management Prior Approval and Appeals Process Impact on Access of Fail-First and Prior Approval Requirements IV. Discussion Notes Appendix Fee-for-Service Questions Managed Care Questions List of Tables Table 1: Estimated Annual Cost of Adult Antipsychotic Treatment, 1998 Table 2: Pharmaceuticals Studied Table 3: Medicaid Programs Studied Table 4: General Pharmacy Benefits, 1998 Table 5: Pharmacy Management Table 6: New Product Presentation Information Requested by State Pharmacy Directors Table 7: Prior Approval and Appeals Process Table 8: Atypical Antipsychotics Prior Approval Table 9: Physical-Health Medications Prior Approval Table 10: Number of Integrated MCOs with Atypical Antipsychotics on Formulary and Prior-Approval List Table 11: Number of Integrated MCOs with Physical Health Medications on Formulary and Prior-Approval List

Medicaid Formulary Policies:
Access to High-Cost Mental Health Medications

(November 1999)

I. Executive Summary

Pharmacy expenditures constitute one of the fastest-growing healthcare costs. As a result, many health plans are adopting new policies to control use of medications—policies that raise questions about appropriate access to pharmaceuticals in private-sector health plans. Such concerns have led 29 states to enact laws requiring managed care organizations to disclose their drug formularies and/or the procedures their members can use to obtain nonformulary drugs. An important question for Medicaid recipients—particularly those now enrolled in managed care arrangements—is how good their access is to the new medications used in treating mental illness, which are more expensive. To date few studies have looked for answers. This study responds to expressions of concern by state policymakers, mental health care providers and consumers, and mental health advocates about the need for appropriate and timely access to and coverage of these newer pharmaceuticals, known as atypical antipsychotics, for the Medicaid population.

The National Institute of Mental Health provided resources to the Bazelon Center for Mental Health Law to: 1) examine general access issues in Medicaid formularies; 2) determine if there are any differences in access to newer, higher-cost medications between drugs used to treat physical problems and those prescribed for mental health conditions; and 3) determine if there are differences in access to medications, for both physical and mental health conditions, depending on the use of various Medicaid managed care arrangements.

In recent years, the combined effect of the increased use of antipsychotic medications and their higher cost has been a sharp rise in expenditures for these drugs. For instance, 1997 retail sales of antipsychotics increased 63 percent, faster than any other drug group. Two antipsychotics, Risperdal^®, with sales of $364 million, and Zyprexa^®, with $334 million, led this growth. States' Medicaid spending for drugs reflects this picture. Florida Medicaid, for example, spends 20 percent of its medication budget on 12 drugs, including five high-cost psychiatric drugs. In Massachusetts, Medicaid's highest drug expenditures are nearly all for psychiatric drugs: Zyprexa^® ($31.3 million), Prilosec^® ($19.6 million), Prozac^® ($18.9 million), Risperdal^® ($17.6 million), Zoloft^® ($15.6 million), Paxil^® ($10.7 million) and clozapine ($10.7 million). In general, the atypical antipsychotics are two to three times more expensive than older antipsychotics. For instance, the pharmaceutical cost of one year of treatment for an adult patient on Zyprexa^® is $1,700 more than treatment with thioridazine hydrochloride (known by the trade name Mellaril^®).

Generally speaking, however, access to pharmaceuticals is broader under Medicaid than under most private insurance. Changes in federal Medicaid policy have ensured fewer restrictions than in a typical private-sector plan. A 1993 amendment to the federal law allows states to exclude drugs from their Medicaid formulary only when based on official labeling or if the excluded drug lacks "significant, clinically meaningful therapeutic advantage in terms of safety, effectiveness or clinical outcome" compared to other available drugs. Barriers, however, still exist.

Access to more than one of the atypical antipsychotics is important for good patient care. An estimated 10-30 percent of individuals with schizophrenia experience limited or no therapeutic response to antipsychotic drugs. Therefore, the greater the range of antipsychotic medications offered, the greater the chance that a provider can find the right drug for that patient.

Nonetheless, all states and the District of Columbia still limit Medicaid recipients' access to medications in a variety of ways. Eleven states restrict access through caps on the number of prescriptions at one time; 42, through prior approval; and 26, through mandatory substitution of generics.

In 1998, the Health Care Financing Administration wrote to state Medicaid directors expressing concerns about access to antipsychotics. The letter suggested that states should update their formularies to reflect the advantages of the new atypical antipsychotics in reduced side effects and increased patient compliance. The letter also appended a summary of current research evidence from Dr. Steven Hyman, Director of the National Institute of Mental Health. Several states have now opened up their formularies and/or removed fail-first requirements. On the other hand, a few states, such as Florida, are known to be considering policies with the opposite effect, such as the adoption of prior approval for more expensive drugs.

Six pharmaceuticals were selected for this study—four atypical antipsychotics (clozapine, Zyprexa^®, Seroquel^® and Risperda^® and two pharmaceuticals for physical-health conditions (Azactam^® and Prilosec^®). Azactam^® and Prilosec^® were chosen based on their higher price and improved outcomes compared to older medications. The study reviewed Medicaid programs in nine states: Connecticut, Florida, Hawaii, Illinois, Iowa, New Mexico, Texas, Utah and Wyoming. This group of states has a range of prior approval use and different delivery systems, including fee-for-service, mental health managed care carve-outs and integrated managed care arrangements, such as HMOs. The response rate was 100 percent.

In these Medicaid programs, notwithstanding concerns, access to atypical antipsychotics appears to be greater than access to the selected physical-health medications. In the Medicaid carve-outs and integrated plans with open formularies, two of the programs require prior approval for Prilosec^®; one plan requires prior approval for Azactam^®. Only one plan requires prior approval for all atypical antipsychotics. In the two states that allow the integrated MCOs to employ restrictive formularies, only one plan required one or two failures before the studied physical and mental health medications can be prescribed. Three plans in one state exclude one atypical antipsychotic, Seroquel^®.

The finding is limited, however, by the nonrandom sampling, the small number of states studied, and whether the two physical-health medications are comparable to the atypical antipsychotics.

Yet Medicaid oversight of managed care was found to be very limited. States focus only on the formulary and prior-approval requirements. Connecticut, Hawaii and New Mexico, all states that include the pharmacy benefit under managed care arrangements, do not routinely collect information on whether the managed care organization lists a drug as a preferred drug, has fail-first requirements, or is using a pharmacy benefit manager. In addition, New Mexico and Hawaii, which allow managed care organizations to have closed formularies, track neither the appeals rate nor the rate at which the original decision is overturned on appeal. Such information is important to monitor whether a pharmacy benefit is too restrictive in practice. A critical study finding is that states' efforts to detect problems in access to pharmaceuticals are extremely limited.

Summary of Findings

General pharmacy benefit:

• Five of the nine states have restrictions on the number of refills per month and/or the number of prescriptions for different drugs that can be filled at one time.
• The percentage of the state pharmaceutical budget spent on psychotropics varies from a high of over 20 percent in Connecticut, Utah and Iowa to a low of 12 percent in Florida.
• None of the states pay for pharmacists to provide cognitive services, which could increase compliance with medication for people with serious mental illnesses.

Pharmacy benefit delivery:

• Medicaid programs include or exclude the pharmacy benefit in the benefit package of the managed care arrangements, depending on the type of managed care approach:
• States usually, but not always, include the pharmacy benefit in integrated managed care arrangements
• States are more likely to maintain the pharmacy benefit for managed care behavioral carve-outs within the fee-for-service program.
• Some states do not monitor to see whether managed care organizations are assuring that Medicaid recipients receive the Medicaid pharmacy benefit. Since often only a closed formulary is available to their commercial members, providers may not treat Medicaid recipients differently.

Prior approvals and appeals processes:

• Connecticut, Hawaii and New Mexico require a managed care organization to provide a process for members to appeal formulary decisions.
• The appeals process collects information on medical necessity, adverse reactions, subtherapeutic response, continuity of care, patient tolerance and compliance factors, and prescriber insistence.
• Only one of the Hawaii plans has enough appeals to calculate overturn rate. The rate of reversal of initial decisions is 50 percent.

Availability of atypical antipsychotics and selected physical-health medications:

• Physical-health medications are slightly more likely to require prior approval in all the plans studied and to be excluded from the closed formularies
• In the two states that allow managed care contractors to employ closed formularies, physical-health medications are slightly more likely to be excluded from the formulary.
• The fail-first requirement in one of the four Hawaii QUEST integrated plans applies equally to physical-health medications and atypical antipsychotics. Only Seroquel^®, the newest atypical, is excluded from the New Mexico integrated plans.

Conclusion

Despite the concerns about access to atypical antipsychotics, Medicaid recipients appear to have more open access to the atypical antipsychotics than to the selected physical-health medications. However, the movement in Medicaid toward managed care arrangements has the potential for expanding the use of closed formularies. Further, the managed care industry is also increasing its use of pharmacy benefit managers to control pharmaceutical treatment. Managed care organizations also have targeted high-cost drugs, including those used to treat psychiatric conditions. Medicaid programs will have powerful incentives to restrict access to atypical antipsychotics, given the likelihood of a continued increase in drug prices, the two- to three-times higher cost of atypical antipsychotics, and the 85 new medicines now under development to treat mental disorders. Finally, this study found that when states allow managed care organizations to develop their own formularies (e.g., Hawaii and New Mexico), current state oversight appears insufficient to identify access problems.

Given these trends, it will be important to track whether the openness of the pharmaceutical benefit for atypical antipsychotics this study found will continue, or whether Medicaid agencies will impose greater restrictions or intensify their monitoring of actual practice by the health plans with which they contract.

II. Background

Introduction

People who rely on Medicaid increasingly receive health services through managed care arrangements—54 percent of current Medicaid beneficiaries⁽¹⁾—although states have been relatively slow to move people who have severe disabilities into managed care. Still, 97 managed care programs in 47 states now provide some form of mental health and/or addiction services.⁽²⁾ Twenty-five percent of all non-elderly people with disabilities—an estimated 1.6 million people—are enrolled in managed care in 36 states.⁽³⁾ Ten of these states provide mental health and addiction services under managed care arrangements to most or all Medicaid beneficiaries, including people who receive federal SSI disability benefits.⁽⁴⁾

Like fee-for-service Medicaid, managed care arrangements often control the utilization of high-cost drugs through formularies and prior authorization. In addition, managed care arrangements are using new cost-control mechanisms, such as preferred drugs and pharmacy benefit managers.

Pharmacy Costs Increasing Faster than Other Health Care Costs

Total pharmacy expenditures constitute one of the fastest-growing areas of health care. Since 1995, national expenditures on prescription drugs increased at twice the rate of health care overall, with a 14.1-percent increase in 1997, compared to 4.8 percent for all health expenditures.⁽⁵⁾ The increase reflects several factors: more prescriptions filled; more new, higher-priced drugs; and rising consumer demand for brand-name drugs.⁽⁶⁾

Within prescription therapeutic classes, 1997 retail sales of antipsychotics increased faster than any other therapeutic class, a 63-percent increase for a total of $927 billion. A great many people with mental illnesses are treated with more than one of these drugs at a time, including medications for side effects.⁽⁷⁾

Our study includes several of the highest sellers among newer drugs. In 1997, Prilosec^®, included in our study, was the national leader in sales with $1.86 billion, an increase of 38 percent from 1996.⁽⁸⁾ Prozac's^® $1.75 billion in sales put it in second place.⁽⁹⁾ Two antipsychotics included in our study, Risperdal^®, with sales of $364 million, and Zyprexa^®, with $334 million, led this growth. Prilosec^® and Risperdal^®, both included in our study, were among the United States' top 200 prescriptions in 1997.⁽¹⁰⁾

Medicaid spending for drugs reflects overall national sales. Currently, 20 percent of Florida's Medicaid drug budget is spent on 12 high-cost medications. Five are psychiatric drugs, including three in our study: clozapine, Zyprexa^® and Risperdal^®.⁽¹¹⁾ Massachusetts Medicaid's highest drug expenditures are nearly all for psychiatric drugs: Zyprexa^® ($31.3 million), Prilosec^® ($19.6 million), Prozac^® ($18.9 million), Risperdal^® ($17.6 million), Zoloft^® ($15.6 million), Paxil^® ($10.7 million) and clozapine ($10.7 million).⁽¹²⁾ High sales of the new atypical antipsychotics reflect the higher cost when compared to older antipsychotics such as haldoperidol, thioridazine hydrochloride (known by the trade name Mellaril^®), fluphenazine hydrochloride (Prolixin^®), trifluoperazine hydrochloride (Stelazine^®) and chloropromazine (Thorazine^®).

The estimated prices are based on the average wholesale price, which is higher than the price a Medicaid program or managed care organization pays. Medicaid programs pay for a drug based on the average wholesale price minus a percentage that varies from state to state. In addition, under the Omnibus Budget Reconciliation Act of 1990, state Medicaid programs receive manufacturer rebates from the Health Care Financing Administration (HCFA) as part of a national program that divides drugs into innovator and non-innovator classes. Managed care organizations or pharmacy benefit managers also enter into their own rebate agreements with manufacturers. Table 1 illustrates the price differences between these drugs for one year of adult treatment. In general, the atypical antipsychotics are two to three times more expensive than the older antipsychotics. For instance, the pharmaceutical cost of one year of treatment for an adult patient on Zyprexa^® is $1,700 more than treatment with thioridazine hydrochloride ( Mellaril^®).

• Table 1: Estimated Annual Cost of Adult Antipsychotic Treatment, 1998

Medicaid Pharmacy Benefits

Under the Omnibus Budget Reconciliation Act (OBRA) of 1990, states must include in their formularies all prescription drugs produced by manufacturers signing rebate agreements. The law also allows states to require that physicians request prior authorization or prior approval for particular drugs. Drugs requiring such additional approval or authorization must be available within 24 hours after the request, and a 72-hour emergency supply of the medication must be made available.⁽¹⁴⁾

Prior to the 1990 changes, several successful lawsuits challenged the exclusion of certain drugs from formularies, restrictive requirements for their prior authorization, and the lack of adequate emergency procedures and of night and weekend access to authorization.⁽¹⁵⁾ Since enactment of the 1990 OBRA changes, however, there has been little litigation in this area.

In 1993, the federal law was amended again to allow states to exclude drugs from their formulary under two restricted conditions. First, exclusion may be based on labeling or information in official medical compendia, such as the manufacturer's designation of conditions for which the drug has been found effective. Second, the drug may be excluded if it lacks "significant, clinically meaningful therapeutic advantage in terms of safety, effectiveness or clinical outcome" compared to other available drugs.⁽¹⁶⁾

All states and the District of Columbia now limit Medicaid recipients' access to medications in a variety of ways. Eleven states restrict access through caps on the number of prescriptions at one time; 42 states, through prior approval; and 26 states, through mandatory substitution of generic drugs.⁽¹⁷⁾ A few states are known to be considering expanding the use of prior approval. Florida is considering a plan to develop two lists of drugs—one of drugs that may be readily prescribed and the other a list of high-cost drugs requiring prior authorization.⁽¹⁸⁾

On the other hand, several states have recently moved in the opposite direction by opening their formularies and removing fail-first requirements. In Tennessee, TennCare's guidelines promote first-line use of selective serotonin reuptake inhibitors (SRRIs)—prescribing this class of drugs first—for depression. Texas Medicaid's latest guidelines advocate first-line use of atypical antipsychotics. MediCal (in California) and Washington Medicaid have removed fail-first requirements before prescribing atypical antipsychotics. MediCal also removed restrictions on Prozac^®, Paxil^®, Wellbutrin^® and Effexor^®.⁽¹⁹⁾

In addition to adopting prior authorization, fail-first requirements, prescription caps and mandatory substitution, Medicaid programs may include or exclude the pharmacy benefit in the benefit package of the managed care arrangement, depending on the type of managed care approach. States usually, but not always, include pharmacy benefits in integrated managed physical and mental health care arrangements. For managed care behavioral health care carve-outs, states are more likely to maintain the pharmacy benefits within the fee-for-service program. A 13-state study of managed care behavioral carve-outs found that the pharmacy benefit remained in fee-for-service Medicaid in seven states (54 percent), was included in the carve-out in five states (39 percent) and was shared between the Medicaid agency and the managed care organizations in one state (7 percent).⁽²⁰⁾

Different issues are raised when states contract out pharmaceutical benefits to managed care organizations. An important recent review of 54 contracts found that only 33 percent set forth minimum specifications for formularies. Forty-one percent of the managed behavioral health agreements required coverage of specific drugs. None of the contracts established standards on therapeutic substitution of drugs.⁽²¹⁾

The movement in Medicaid toward managed care arrangements has the potential to increase the use of formularies. The managed care industry frequently uses formularies to control pharmacy costs and to encourage the use of more effective pharmaceutical treatments. The Annual Industry Survey of the American Association of Health Plans found that 96 percent of health plans use a formulary—58.5 percent have an open formulary and 38.3 percent have a closed formulary.⁽²²⁾ Most HMOs use open formularies, which allow doctors to prescribe non-formulary drugs without prior authorization or penalty. Fewer than three in 10 HMOs use closed formularies.⁽²³⁾ In 1997, 92 percent of HMOs used pharmacy benefit managers.⁽²⁴⁾

Managed care organizations have targeted high-cost drugs, including medications for psychiatric conditions, to control their use. A 1998 study of HMOs and pharmacy benefit managers found an increase in prior-authorization requirements and in drugs designated as "restricted use." Wellpoint Health Networks and Aetna U.S. Healthcare have restricted the use of Celebrex^®, a new anti-inflammatory drug, by increasing co-payments and restricting use to a class of patients.⁽²⁵⁾ Antidepressants are included in the 10 therapeutic classes for which the largest number of plans report a restriction.⁽²⁶⁾ Major insurers in Pennsylvania ⁽²⁷⁾ and New Hampshire ⁽²⁸⁾ have restricted access to Prozac.^®

Managed care arrangements have also introduced new strategies to reduce drug expenditures, such as increased paperwork, lists of preferred drugs, incentives for pharmacists and physicians, and pharmacy benefit managers.⁽²⁹⁾ Pharmacy benefit managers (PBMs) negotiate discounts with pharmacists and manufacturers in exchange for an increased volume of members and increased volume of prescriptions for particular drugs. In 1995, the Federal Employees Health Benefits Plan saved $505 million through the use of pharmacy benefit managers: 73.5 percent from discounts, 14.3 percent from paying only the maximum allowable cost, 7.2 percent from requiring prior approval, 2.0 percent from drug-utilization review, and 2.7 percent from PBMs' encouraging patients and physicians to use less expensive brand-name drugs.⁽³⁰⁾

The Congressional Budget Office estimates that, as a result of these strategies, HMOs pay 18 percent less for prescription drugs than retail pharmacies. Several of the largest PBMs are owned by pharmaceutical manufacturers, attracted to the business by the advantage it gives them to channel patients to their drugs. Quality of care concerns have been raised regarding access to pharmaceuticals that are not made by the PBM's owner.⁽³¹⁾

Concerns about Access to Psychiatric Medications

States, providers, mental health consumers and pharmaceutical companies are expressing concerns about managed care consumers' access to and coverage of pharmaceuticals.⁽³²⁾ Managed care organizations have responded to the disproportionate rise in pharmacy costs by finding ways to save money.

A total of 29 states have laws that require managed care organizations to disclose the drug formulary and/or the procedure necessary to obtain nonformulary drugs. Five of these laws (in Iowa, Maryland, North Dakota, South Dakota and Virginia) were passed in 1999. In addition, 26 states are currently considering enactment of similar statutes or amendment of their current laws to require such disclosure.⁽³³⁾

Responding to the rising number of formulary restrictions and prior-approval requirements, the Pharmaceutical Research and Manufacturers of America Association has supported state legislation mandating open formularies.⁽³⁴⁾ Psychiatrists have been concerned enough about access to new treatments to challenge restrictions on prescriptions. The Massachusetts Psychiatric Society was successful in its effort to have one of the state's largest health plans include SSRIs, such as Prozac^®, as a first-line drug in its treatment guidelines.⁽³⁵⁾

In 1998, the Health Care Financing Administration wrote to state Medicaid directors to express concerns about access to antipsychotics. The letter suggested that states update their Medicaid formularies to reflect the advantages of the new atypical antipsychotics in reduced side effects and increased patient compliance. The letter also appended a summary of current research evidence from Dr. Steven Hyman, Director of the National Institute of Mental Health.⁽³⁶⁾ Since the letter's distribution, HCFA's Technical Director at the Center for Medicaid and State Operations has heard fewer concerns regarding access to psychiatric medications and had fewer reports of prior authorization programs' becoming increasingly burdensome or restrictive.⁽³⁷⁾

The leading mental health advocacy organizations have supported consumer access to the latest pharmaceuticals. The National Alliance for the Mentally Ill (NAMI), in its public policy platform, strongly opposes any limitations on the availability of new medications such as atypical antipsychotics and SSRIs. NAMI has emphasized these treatments because "of ongoing efforts by various individuals and organizations to limit availability due to cost, ideology, or the infringement of the rights of persons with brain disorders to receive them."⁽³⁸⁾ The National Mental Health Association (NMHA) opposes restrictive formularies that limit available drugs or provide a list of preferred drugs. NMHA is concerned with increased hospital stays and reduced productivity due to use of less effective medication and the more severe and painful side-effects of the older medications.⁽³⁹⁾

Prior Studies on Access to Antipsychotic Medications

A few studies have assessed ease of access to medications for mental health conditions, finding, among other things, that access to pharmaceuticals appears broader under the Medicaid program than under private insurance. The National Alliance for the Mentally Ill surveyed nine of the major behavioral health managed care organizations nationally on prompt access to new atypical antipsychotics and fail-first requirements. None of the five responding organizations provided prompt access or offered a no-failure requirement.⁽⁴⁰⁾

The Public Advocate for the State of New York examined coverage of antidepressants and antipsychotics in the formularies of the state's 14 major commercial HMOs. The study considered whether a brand-name drug is on the formulary, whether a generic form of a multi-source drug is covered, and whether a drug is a preferred drug. Access was measured by creating a level of coverage for each therapeutic category—the percentage of drugs each PBM covers or prefers out of all drugs listed on any of the 15 formularies. The study found restrictions on mental health medications available through private arrangements. For antidepressants, HMO formulary coverage ranged from a high of 85 percent of all antidepressants to a low of 48 percent. For antipsychotics, the HMOs ranged 94 percent to 53 percent. HMOs provided comparable and more extensive coverage of antidepressants and antipsychotics compared to some other categories of drugs. Coverage of antidepressants and antipsychotics was lower than antiulcer coverage, about the same as coverage for steroid/anti-inflammatory inhalers, ACE inhibitors for cardiovascular conditions and otitis media antibiotics, and higher than coverage for beta blockers, calcium channel blockers, bronchodilator inhalers, beta agonists and vasodilators.⁽⁴¹⁾

The SAMHSA Tracking System conducted the only study of access to pharmaceuticals for mental health and substance abuse under Medicaid formularies. Six states with Medicaid behavioral carve-out programs were surveyed for access to four atypical antipsychotics: clozapine (known by the trade name Clozaril^®), Risperdal^®, Seroquel^®and Zyprexa^®. In all six states, prior authorization was required only for clozapine.⁽⁴²⁾

III. Study of Medication Coverage for Physical Health And Mental Health

Study Design

This study was designed to: 1) examine general access issues in Medicaid formularies and 2) determine if any differences exist between physical and mental health conditions in access (a) to newer, higher-cost medications and (b) depending on managed care arrangements. Study issues were based on a literature search, a review of previous studies and input solicited from experts in the field. Data were obtained from state Medicaid programs and existing sources such as the National Pharmaceutical Council, and from HCFA, when available.

The review was separated in two sections. Part I includes general issues regarding the pharmacy benefit, such as who manages it, what mechanisms are in place and how consumers can appeal. Part II includes specific issues related to the six drugs studied, such as whether they are listed in the formulary, require pre-approval or have any fail-first requirements. The separate documents to review fee-for-service Medicaid and managed care arrangements are included in the Appendix.

Six pharmaceuticals were selected for the study. The four selected atypical antipsychotics were first prescribed as early as 1989 (clozapine) and as late as 1997 (Seroquel^®). Clozapine is the only drug studied that is currently available in a generic version. The two pharmaceuticals for physical-health conditions, Azactam^® and Prilosec^®, were selected based on their higher price and improved outcomes compared to older medications. These two drugs were introduced 13 and 10 years ago, respectively, several years before three of the atypical antipsychotics. Table 2 provides detailed information for the selected pharmaceuticals regarding conditions treated, year introduced, generic alternative and patent expiration.

• Table 2: Pharmaceuticals Studied

Nine states were selected to represent a range of experience and of delivery systems including fee-for service, mental health managed care carve-outs and integrated managed care arrangements such as HMOs. Two states (22 percent of those reviewed) employ prior authorization; this is comparable to the 16 percent of all states with prior authorization.

Illinois and Wyoming were selected for their fee-for-service arrangements, which serve most beneficiaries. Four states—Hawaii, Iowa, Texas and Utah—were selected for their contracts with mental health carve-out companies. Florida, Connecticut and New Mexico were selected because they contract with integrated physical and mental health managed care entities. Hawaii and Texas have more than one arrangement included in the study. The oldest program is the Florida HMO program, established in 1982, and Texas NorthSTAR, which began in 1999, is the newest.

All nine states responded, for a response rate of 100 percent. Information was provided by the Medicaid director or the pharmacy director in each state identified for study, except for Hawaii, which requested that each organization contracting with the state provide the information. Connecticut, Hawaii and New Mexico provided varying degrees of information on their integrated managed care arrangements. Connecticut was unable to provide any information on the managed care organizations' formulary, prior approval, fail-first requirements or preferred-drug status. New Mexico provided information only on the formulary, but did not provide any information on the appeals process, fail-first requirements or preferred-drug status.

States with behavioral health carve-outs have chosen two different approaches. The Iowa Plan for Behavioral Health and Texas NorthSTAR carve-out arrangements do not include a pharmacy benefit. Their pharmacy benefit remains under the state's Medicaid fee-for-service program. On the other hand, Hawaii QUEST for Serious Mentally Ill and Utah's program include the pharmacy benefit in the carve-out arrangement. Table 3 describes in further detail the Medicaid pharmaceutical programs included in the study.

• Table 3: Medicaid Programs Studied

General Pharmacy Benefits and Management

Five of the states have restrictions either on the number of refills per month or the number of prescriptions for different drugs that can be filled at one time. Wyoming limits Medicaid recipients to a maximum of three prescriptions for different drugs per month. Florida authorizes prescription limits of six to 12 prescriptions per month. None of the states pay for pharmacists to provide cognitive services, which could increase compliance with medication for people with serious mental illnesses.

Laboratory testing to monitor the blood level of clozapine patients usually is not counted as part of the mental health visit benefit, except in New Mexico, where it may be, depending on who provides the testing. Whether testing is counted as a visit is an important issue in the plans that limit the number of visits for mental health care.

The average Medicaid drug payments per recipient varies from a high of $1,383 in Connecticut to a low of $343 in New Mexico. The two states with closed formularies do not present a pattern regarding drug payments per recipient, with Hawaii having the second highest payments and New Mexico the lowest. The percentage of the state pharmaceutical budget spent on psychotherapeutics varies from a high of over 20 percent in Connecticut, Utah and Iowa to a low of 12 percent in Florida. See Table 4 for each state's general pharmacy benefits.

• Table 4: General Pharmacy Benefits, 1998

None of the state Medicaid plans that include their pharmacy benefit in a managed care arrangement have dollar limits for pharmaceuticals. These programs employ at least one if not all four mechanisms to reduce pharmacy costs and increase efficiency. See Table 5 for greater detail.

• Table 5: Pharmacy Management

In selecting new products for their formulary, almost all the state pharmacy directors request information on pricing, efficacy, side effects, drug interactions, safety, pharmacoeconomic data and outcomes data. See Table 6 for more information.

• Table 6: New Product Presentation Information Requested by State Pharmacy Directors

Prior Approval and Appeals Process

Connecticut, Hawaii and New Mexico require a managed care organization to provide a process for members to appeal formulary decisions. Connecticut has a unified process, in which consumers simultaneously file a grievance with the plan and request a fair hearing by the Department of Social Services.

The four Hawaii managed care organizations illustrate how managed care organizations create different prior-approval and appeals processes under closed formularies. In these plans, the prior-approval process ranges from an informal service request to a more formal one-page questionnaire. The average time for prior approval ranges from one hour to five days following the first prescription. The appeals process collects information on medical necessity, adverse reactions, subtherapeutic response, continuity of care, patient tolerance and compliance factors, and prescriber insistence. It is noteworthy that only one of the Hawaii plans receives enough appeals to calculate an overturn rate; in this plan the rate of reversal of initial decisions is 50 percent. See Table 7 for detailed information.

• Table 7: Prior Approval and Appeals Process

Impact on Access of Fail-First and Prior-Approval Requirements

Fail-first requirements are rarely required for physical health and atypical antipsychotics. Only one plan (one of the Hawaii QUEST plans) requires one or two failures before the studied medications can be prescribed. This requirement applies equally to physical health and atypical antipsychotics.

In both the Medicaid carve-outs and integrated plans, physical-health medications are more likely to require prior approval than atypical antipsychotics. See Tables 8-11 for further detail. Two of the programs require prior approval for Prilosec^®; one plan also requires prior approval for Azactam^®. Only the Hawaii QUEST plan that provides services for people with serious mental illnesses requires prior approval for all the atypical antipsychotics. Most plans in the two states that allow managed care organizations to have closed formularies require prior approval for the physical-health medications studied. Of the atypical antipsychotics, Seroque^® is the only one excluded from the closed formulary and only for the New Mexico managed care plans. Differences in availability between the atypical antipsychotics and the physical-health medications in the states that allow managed care plans to have closed formularies are shown in tables 10 and 11.

• Table 8: Atypical Antipsychotics Prior Approval
• Table 9: Physical-Health Medications Prior Approval
• Table 10: Number of Integrated MCOs with Atypical Antipsychotics on Formulary and Prior-Approval List
• Table 11: Number of Integrated MCOs with Physical-Health Medications on Formulary and Prior-Approval List

IV. Discussion

With pharmacy costs rising at more than twice the rate of inflation since 1995, Florida could be the first of a wave of states moving to restrict access through prior approval. Whether states will require all, a few, or only one atypical antipsychotic to be available in managed care arrangements is a matter of concern, because providing a range of pharmaceutical treatment is critical for the effective treatment of severe mental illnesses. An estimated 10 to 30 percent of people with schizophrenia have limited or no therapeutic response to antipsychotic drugs.⁽⁵¹⁾ Therefore, the greater the range of antipsychotic medications offered in a state Medicaid program, the greater the chance that a provider can find the right drug for that patient. Eighty-five new medicines are under development to treat mental disorders—15 for schizophrenia, 18 for depression and 19 for substance abuse disorders.⁽⁵²⁾ Accordingly, access to pharmaceuticals for mental health conditions will continue to be an issue.

With a large proportion of the pharmacy budget devoted to psychopharmaceuticals, state Medicaid programs and contracting managed care organizations may target the atypical antipsychotics for reduced access by allowing contracted managed care organizations to use closed formularies, prior approval, preferred drug status, fail-first requirements and pharmacy benefit managers.

However, although costs are two to three times greater for the new atypical antipsychotics than for older antipsychotic drugs, the new drugs have fewer side effects, provide greater improvement in negative symptoms, increase patient compliance, and reduce overall use of medical resources.⁽⁵³⁾ The question remains whether pharmaceutical restrictions result in true savings or merely shift costs to other areas of health care. One study found that when a limit on the number of prescriptions at one time reduced use of antipsychotic drugs, the use of outpatient services increased. With removal of the prescription limit, the use of medications and most mental health services returned to initial levels.⁽⁵⁴⁾

The findings of this study of Medicaid formulary policies are limited by the sample method and the sample size. The study targeted a sample of nine states to represent the various managed care arrangements. Other states not sampled may also use prior approval and allow managed care arrangements to exclude certain drugs from the benefit package. In addition, only two physical-health medications were selected, compared to four atypical antipsychotics. The two physical-health medications have been on the market from four to 10 years longer than most of the atypical antipsychotics. The two physical-health medications may not have been comparable in terms of efficacy, alternative treatments and duration of treatment. In addition, the study examined only antipsychotics, excluding other mental health medications such as antidepressants, anti-anxiety and substance abuse medications.

As more states consider moving people with serious mental illnesses into managed care, it is important to track continued openness of the pharmaceutical benefit regarding atypical antipsychotics. Currently, only eight states have closed formularies. States can require that managed care contractors offer a certain number of drugs in each therapeutic class. For instance, New Mexico requires its contractors to offer at least one (but not necessarily more than one) drug in each therapeutic class.

The following findings of this study are particularly important:

• Some state Medicaid agencies focus on formulary issues and prior approval. They have not expanded their oversight to new cost-control practices, such as preferred drugs, fail-first requirements and manufacturer price reductions, which create incentives for patients to receive one drug over another.
• In Medicaid managed care arrangements (both the carve-outs and the integrated plans), physical-health medications are slightly more likely than atypical antipsychotics to require prior approval.
• In managed care arrangements in the two states that allow managed care organizations to exclude drugs from their formularies, physical-health medications were more likely to be excluded than atypical antipsychotics. The latest atypical antipsychotic, Seroquel^®, was more likely not to be on the formulary and to require prior approval in one of the states.
• The study indicates a broad range of approaches and criteria for considering the approval of a nonformulary drug.
• State Medicaid agencies lack requirements on the content and criteria for prior approval and appeal processes implemented by the managed care contractors.

Managed care organizations limit access in ways other than through closed formularies. Many are implementing new cost-control mechanisms, such as preferred drugs, fail-first requirements and pharmacy benefit managers. However, state Medicaid programs have focused on formularies and appear unconcerned about these other approaches. This conclusion is supported by findings that, while many Medicaid managed care contracts set forth minimum specifications for formularies, few require coverage of specific drugs and none had established standards on therapeutic substitution of drugs.⁽⁵⁵⁾

Connecticut, Hawaii and New Mexico, all states that include the pharmacy benefit under managed care arrangements, do not routinely collect information on whether the managed care organization lists a drug as a preferred drug, has fail-first requirements, or is using a pharmacy benefit manager. A follow-up to this study would be to examine additional states that allow Medicaid managed care contractors to exclude drugs from their formularies, in order to discover how individual managed care organizations are designing their formularies, designating drugs as preferred and requiring fail-first and prior authorization.

In addition, New Mexico and Hawaii, which allow managed care organizations to develop their own formularies, track neither the appeal rate nor the rate at which the original decision is overturned on appeal. Such information would be useful for a state to monitor whether a formulary benefit was too restrictive.

Another concern is whether managed care organizations are assuring that Medicaid recipients receive the Medicaid pharmacy benefit. Usually a more restrictive closed formulary is available to commercial members, and providers may not treat Medicaid recipients differently from the commercial population in the same managed care plan. To test this, one approach would be to conduct a case study that would interview managed care organizations, psychiatrists, pharmacists and consumers to look at the educational efforts of managed care organizations and the level of awareness among both providers and consumers regarding the Medicaid open formulary pharmacy benefit.

Given the rise in spending on psychiatric medications and the expansion of managed care arrangements for people with serious mental illnesses in Medicaid, it will be important to track whether the openness of the pharmaceutical benefit for atypical antipsychotics found in this study continues, or whether Medicaid agencies impose greater restrictions or intensify their monitoring of actual practice by the health plans with which they contract.

Acknowledgments

The Bazelon Center wishes to thank all who assisted in the preparation of this publication. In particular, many state Medicaid directors and Medicaid pharmacy program administrators gave time in support of our survey of state Medicaid pharmacy benefits.

This report was prepared by the Bazelon Center for the National Institutes of Mental Health, Service Systems Research Program, 5600 Fishers Lane, Rockville, MD 20857 (Contract No. QVP-80076). The content of this publication does not necessarily reflect the views or policies of the funding agency and should not be regarded as such.

Robert Bernstein, Executive Director, Bazelon Center for Mental Health Law, 1101 Fifteenth Street N.W., Suite 1212, Washington D.C. 20005, (202) 467-5730; TDD (202) 467-4232, Fax: (202) 223-0409, HN1660@handsnet.org, www.bazelon.org

Credits: Report written by: Rafael M. Semansky and Chris Koyanagi; design by: Lee A. Carty

Notes

1. Health Care Financing Administration. (1998). Medicaid managed care enrollment report. Baltimore: Author.

2. SAMHSA Managed Care Tracking System. (1998). State profiles on public sector managed behavioral healthcare and other reforms. Rockville, MD: Author.

3. Kaiser Commission on Medicaid and the Uninsured. (1999). Medicaid's disabled population and managed care. Washington, D.C.: Author.

4. SAMHSA Managed Care Tracking System, note 2, above.

5. Levit, K., et al. (1998, November/December). National health expenditures in 1997: more slow growth. Health Affairs, 99-110.

6. Ibid.

7. West, J.C., et al. (1996, July). Treatment provided to psychiatric patients. Psychiatric Services, 47 (7), 693.

8. NDC Health Information Services. US pharmaceutical market 1997 year in review. (Online, http://www.simatics.com/insights/market.htm).

9. Ibid.

10. Top 200 US prescriptions. (Online, http://www.rxlist.com/top200.htm).

11. Florida Governor seeks Medicaid drug cuts; MH advocates worry about impact. (1999). Mental Health Weekly, 9 (10), 1, 5.

12. Massachusets' Medicaid review supports use of SRRIs over older antidepressants. (1999, April 12). The Pink Sheet, 16(15).

13. Sources: Physicians' Desk Reference, 1997 (Montvale, NJ: Medical Economics Company, 1997, 1999), provided the suggested maintenance dosage or, if a range was provided, an average was calculated. The 1998 Red Book (Montvale, NJ: Medical Economics Company, 1998) provided the average wholesale price. The price of the generic was used, if available.

14. Omnibus Reconciliation Act of 1990, Pub. L. No. 101-508.

15. Dodson v. Parham, 427 F. Supp. 97 (N.D. Ga. 1977); Visser v. Taylor, 756 F. Supp. 501 (D. Kan.1990); National Ass'n of Chain Drug Stores, Inc. v. Bowen, No. 87-2911, 1989 WL 43948 (D.D.C. Apr. 12, 1989); Jeneski v. Myers, 209 Cal. Rptr. 178 (Cal. Ct. App. 1984), cert. denied, 471 U.S. 1136 (1985).

16. Omnibus Budget Reconciliation Act of 1993, Pub. L. No. 103-66.

17. National Pharmaceutical Council, (1997, 1998) National pharmaceutical benefits under medical assistance programs. Reston, VA.:Author.

18. Becker, J. (1999, February 28). Plan pushes cheaper medicine. St. Petersburg Times. (Online, http://www.sptimes.com/News/22899/State/Plan_pushes_cheaper_m.html.)

19. Gregorian, R.S., and Hughes, C. (1999). Access to psychotropic medications and substance abuse agents. SAMHSA Managed Care Tracking Report, 1 (3).

20. Ibid.

21. Rosenbaum, S., et al. (1998). Special report: Mental illness and addiction disorder treatment and prevention. Washington D.C.: Center for Health Policy, George Washington University Medical Center.

22. Personal communication, Peter Fitzgerald, Director of Quality Management and Health Management Research, American Association of Health Plans, October 1999.

23. Drug formularies: Is their growth among HMOs leveling off? (1997, February 10). Medical Economics 39.

24. Trend of the month: More HMOs are using drug formularies. (1998). Drug Benefit Trends, 11, (2), 8-9.

25. Morrow, D. (1999, January 8). Some HMOs plan limited coverage for new arthritis drug. The New York Times. (Online, http:// www.nytimes.com.)

26. Formulary restrictions on the rise. (1999). Newtown, Pennsylvania: Scott-Levin.

27. Snowbeck, C. Prozac squeeze: Major insurers restricting coverage for popular antidepressant. (1999, February 2). Pittsburgh Post-Gazette. (Online, http://www.post-gazette.com/healthscience/1999ozozdrugs3asp.)

28. NH-based health plan restricts use of Prozac. (1998). Mental Health Weekly, 8.

29. Sources: Congressional Budget Office. (1998). How increased competition from generic drugs has affected prices and returns in the pharmaceutical industry. Washington D.C. Public Advocate for the City of New York. (1996). Compromising your drug choice: How HMOs are dictating your next prescription. New York, NY.

30. U.S. General Accounting Office. (1996, September). Blue Cross and Blue Shield: Change in pharmacy benefits affects federal enrolles. GAO/T-HEHS-96-206. Washington D.C.

31. Congressional Budget Office, note 29, above.

32. National Pharmaceutical Council, note 17, above.

33. Herstek, J. (1999, April).Managed care drug formularies. Washington D.C.: Health Policy Tracking Service.

34. Special report: Medicaid, prior-approval policies hinder access to newer dugs, groups say. (1998). Washington: Bureau of National Affairs, 6 (135), 634-637.

35. Psychiatrists speak out against managed care formularies-and win. (1996). Psychiatric News, 9, 15.

36. S.K. Richardson, Director, Center for Medicaid and State Operations, Health Care Financing Administration, February 12, 1998.

37. Personal communication, Larry Reed, Technical Director at the Center for Medicaid and State Operations, Health Care Financing Administration, September 1999.

38. National Association for the Mentally Ill. (1999). Public policy platform. (Online, http://www.nami.org/update/platform.htm.)

39. National Mental Health Association. (1998, August). NMHA policy position on restrictive formularies—revised (online, http://www.nmha.org/position/ps03.26980.cmm); NMHA policy position on restrictive formularies (online, http://www.nmha.org/state/form/index.cfm).

40. Hall, L.L., Edgar, E., and L.M. Flynn. (1997). Stand and deliver: Action call to a failing industry, the NAMI managed care report card. Arlington, VA: National Alliance for the Mentally Ill.

41. Public Advocate for the City of New York, note 29, above.

42. Gregorian & Hughes, note 19, above.

43. Sources: Conditions treated from Physicians Desk Reference (Montvale, New Jersey: Medical Economics Company, 1997, 1999). Year introduced from Electronic Orange Book, (Rockville, MD: Food and Drug Administration, 1998, Online, http://www.fda.gov/cder/od/defaul.htm). Note selected earliest year if multiple patent dates. Patent expires from personal communication, Mary Krimsner, Food and Drug Administration, January 25, 1999, except for the patent expiration of Azactam^®, which was provided in personal communication, Regulatory Office, Durable Pharmaceuticals, March 1998.

44. Sources: Medicaid Enrollees 1998 from Medicaid Managed Care State Enrollment (Baltimore, Md: Health Care Financing Administration, 1998). Available online at http://www.hcfa.gov/medicaid/mcsten97.htm Year Program Began from State Profiles on Public Sector Managed Behavioral Healthcare and Other Reforms (Rockville, MD: SAMHSA Managed Care Tracking System, 1998) and Office of Managed Care, Florida Agency for Health Care Administration.

45. 1999 data.

46. National Pharmaceutical Council, note 17, above.

47. 1996 data.

48. Three of the four plans require physician compliance.

49. National Pharmaceutical Council, note 17, above.

50. One plan requires prior authorization only after 60 capsules.

51. Corrigan, P., Lieberman, R., and Engel, I. (1990). From noncompliance to collaboration in the treatment of schizophrenia. Hospital and Community Psychiatry. 41 (11), 1203-1211.

52. Pharmacy Research and Manufacturers of America. (1998, June). PhRMA survey finds 85 medicines in testing for mental health illness. (Online, http://www.pharma.org.)

53. Sources: Nightengale, B.S., et al. (1998). Economic outcomes of antipsychotic agents in a Medicaid population: Traditional agents v. risperidone. Psychopharmacology Bulletin, 34, (3), 373-382. Reid, W.H., Mason, M., and Toprac, M. (1994). Savings in hospital bed-days related to treatment with clozapine. Hospital and Community Psychiatry, 45 (3), 261-264. Honigfeld, G., and Patin, J. (1990). A two-year clinical and economic follow-up of patients on clozapine. Hospital and Community Psychiatry, 41 (8), 882-885. Ghaemi, S.N., et al. (1998). Cost-effectiveness of clozapine therapy for severe psychosis. Psychiatric Annals, 49 (6), 829-831. Fichtner, C.G., Hanrahan, P., and Luchins, D.J. (1998). Pharmacoeconomic studies of atypical antipsychotics: review and perspective. Psychiatric Annals, 28 (7), 381-396.

54. Soumerai, S.B., et al. (1994). Effects of limiting Medicaid drug reimbursement benefits on the use of psychotropic agents and acute mental health services by patients with schizophrenia. New England Journal of Medicine, 331 (10), 650-655.

55. Rosenbaum, note 21, above.