Medicaid agencies are using clinical evidence to help manage drug costs

With state Medicaid agencies reporting that drug costs are a major driver of overall program spending growth, some states are looking for ways to use clinical evidence about the effectiveness of specific drugs or drug classes to curtail pharmacy costs while also ensuring beneficiary access to needed drugs.

A new Commonwealth Fund issue brief discusses 2004 site visits by the National Academy for State Health Policy and Georgetown University to determine how six state Medicaid agencies — California, Florida, Kansas, Michigan, Missouri, and Washington — manage their pharmacy benefit.

"There are lots of ways to manage the drug benefit," National Academy for State Health Policy senior program director Neva Kaye tells State Health Watch, "and they all can work."

The issue brief summarizes the states' experience in these areas of pharmacy benefit management: the role of pharmaceutical and therapeutics committees convened by Medicaid agencies in developing preferred drug lists; use of prior approval processes in managing the pharmacy benefit and enforcing preferred drug lists, the Drug Effectiveness Review Project subscription service providing states with comparative reviews of drug effectiveness, and management of behavioral health drugs.

  • Pharmaceutical and therapeutics committees. Many states use preferred drug lists to ensure Medicaid beneficiaries' access to needed medications, restrict inappropriate use of prescription drugs, and contain pharmacy costs. States rely on pharmaceutical and therapeutics (P&T) committees to produce actionable, evidence-based recommendations that can be used to develop the lists. Ms. Kaye says P&T committees use clinical evidence to evaluate drug merit on a class-by-class basis.

In some states, drugs that are judged to be superior to others in their therapeutic class are placed on the preferred list. In some states, if drugs are deemed therapeutically equivalent, the preferred drugs are identified based on cost, among other factors. In other states, clinical and cost factors are reviewed simultaneously.

Most Medicaid agencies create incentives for beneficiaries and providers to use preferred drugs by requiring them to obtain permission before prescriptions for nonpreferred drugs can be filled. Sometimes, copayments are imposed to encourage use of preferred drugs. States also have found that educating physicians, pharmacists, and beneficiaries about the preferred drug list, its purpose, and evidence base also is important in promoting compliance, especially in states not requiring prior approval or copayments. States also use the lists to negotiate supplemental rebates from drug companies.

Stakeholders in all six states and Medicaid agency officials in all states except Michigan emphasized to the researchers the importance of having practicing clinicians, ideally from a range of professional disciplines, on P&T committees. Most committees consist of doctors and pharmacists who make recommendations during formal meetings. California's committee, however, is an informal advisory group composed mainly of academic researchers who don't have in-person meetings. Washington's committee includes a nurse and a physician assistant, while Florida's has a consumer representative and requires that at least one member represent drug manufacturers.

The P&T committees in the six states don't dictate which drugs should be on the preferred lists. In four of the states, the committees make recommendations to state officials, while in Florida and Missouri they produce provisional lists that are reviewed and ratified by other committees. In five states, the committees make decisions about clinical equivalency of the drugs in a particular therapeutic class by member vote. In California, committee members provide their recommendations individually to state Medicaid staff who compile the recommendations and staff analysis of cost considerations to produce a preferred drug list.

  • Prior approval. Many states report using a prior approval process along with the preferred drug list. Ms. Kaye says a well-designed prior approval process can encourage providers to comply with the list, while also ensuring that beneficiaries can access nonpreferred drugs under clinically appropriate circumstances. To be most effective, prior approval should minimize administrative burdens on providers and the financial costs of administration for the state agency.

Providers who want to prescribe drugs not on the preferred list have to submit a formal request to the state Medicaid agency, which approves requests after determining that the drug in question is medically necessary. In practice, according to Ms. Kaye, most prior approval requests are approved.

While some have argued that the high approval rate for prior approval requests signals a lack of need for the technique, requiring approval "appears to produce a significant sentinel effect," according to the study. The site visits say the prior approval process aligns physician prescribing patterns with the preferred drug list and report compliance rates of 85% to 95%. States also require a high level of coordination between prescribing physicians and dispensing pharmacists in submitting prior approval requests.

In most cases, states establish clinical criteria to determine whether a specific prior approval request meets the definition of medical necessity. Thus, if there is evidence demonstrating that patients with a history of heart disease are more likely to suffer certain side effects from a preferred drug than from a nonpreferred drug, the state might establish criteria granting prior approval requests for the non-preferred drug for beneficiaries with a history of heart disease. In cases where clear evidence doesn't exist, decisions are made by Medicaid agency reviewers.

Two typical exceptions to a prior approval process are emergency provisions and "dispense as written" (DAW) instructions. Emergency provisions are intended to ensure that prior approval system interruptions don't stand in the way of dispensing necessary drugs. While site visit states reported emergency provisions are rarely used, advocates and pharmacists in the states said the emergency provisions don't always work as intended. Advocates said pharmacists on occasion don't dispense needed medications, while pharmacists said they are reluctant to dispense emergency medications for fear they won't be reimbursed or will be liable if anything happens to the beneficiary.

DAW is a privilege granted to physicians by states; of the six site visit states, Florida, Kansas, and Washington have experience with it. Kansas eliminated the privilege after finding that doctors used it to override the preferred drug list 70% of the time. Florida also reported problems. In Washington, doctors were given the privilege after agreeing not to oppose creation of a preferred drug list during legislative debate. But it is available only to doctors participating in the Therapeutic Interchange Program, which permits pharmacists to substitute a preferred drug when a physician prescribes a nonpreferred drug (without also writing DAW). Although the state agency has been concerned that the DAW privilege might be used to circumvent the prior approval process and the Therapeutic Interchange Program, Ms. Kaye reported early evidence suggested it was being used modestly and thus linking DAW to a Therapeutic Interchange Program might be a key to success of such programs.

  • Drug Effectiveness Review Project (DERP). DERP is a collaborative partnership between states and other government and non-profit entities that conducts systematic, evidence-based reviews of drugs. Based at the Oregon Health & Science University in Portland, OR, it currently has 19 members, mostly state Medicaid programs.

Since its inception, DERP has developed 12 drug class reviews commissioned by Oregon, Washington, and Idaho, and 13 drug class reviews commissioned by the larger DERP collaborative. Each of the reviews is updated based on new evidence every six months. Together, the drug classes that have been reviewed account for more than half of all drug utilization.

DERP is a self-governing project, with member organizations setting priorities, determining which drug class reviews to conduct, and developing key questions and inclusion criteria for each review. The reviews involve a comprehensive search of the literature for all relevant articles, including citations received as a result of soliciting input from drug companies. Articles are selected according to inclusion criteria (based on factors such as the patient populations, treatments, and outcomes studied). Articles then are rated for their methodological quality by at least two independent reviewers. Data from the included studies are abstracted by two reviewers, allowing the research team to synthesize the results in various ways.

Ms. Kaye reported that among the six site visit states, two made DERP their sole source of evidence to support decisions, two had at least one alternate source, and two did not participate in DERP.

Several states have indicated that some stakeholders including doctors and pharmacists questioned whether preferred drug lists and prior approval policies would really be evidence-driven. Participation in DERP has helped to convince some skeptics that pharmacy management decisions are based on clinical evidence. Other stakeholders have needed to be reassured that while DERP reports and other products inform states' decision-making processes, they don't take the final decision-making out of the hands of officials in their own states.

To date, according to Ms. Kaye, most DERP reviews have focused on whether there are differences among a set of competing products in classes such as statins for managing cholesterol, proton-pump inhibitors for gastrointestinal disorders, or angiotensin-converting enzyme inhibitors for lowering blood pressure. To the extent that reviews show few, if any, significant differences among the competing products, states can negotiate with manufacturers. Where there is evidence that certain drugs result in better outcomes, states can help to ensure that these products are available to beneficiaries. Ms. Kaye said future DERP reviews might evaluate the effectiveness of combination drugs, such as a new drug that combines a cholesterol medication with a treatment for hypertension, or the effectiveness of different dosage forms.

  • Behavioral health pharmaceuticals. In recent years, according to Ms. Kaye, states have begun to consider ways to manage behavioral health pharmaceuticals because of increases in cost as well as shifts in utilization. Medicaid agencies in the site visit states reported that drugs that treat mental illnesses command a greater proportion of their pharmaceutical expenditures than drugs for any other disease category.

While behavioral health pharmaceuticals include many drug classes, states report that atypical antipsychotics and second-generation antidepressants are of greatest concern because they are among the most expensive and widely used drugs. Yet the site visit states report they also view these drugs as efficacious and cost-effective.

States said prescribing for psychotropic drugs must consider individual response and tolerance for side effects. As a result, site visit states were concerned that managing access through preferred drug lists or other means could inadvertently lead to psychiatric destabilization and more costly acute or inpatient care, which can be significantly more expensive for states since federal Medicaid funds can't be used for services provided in institutions for mental disease to beneficiaries between the ages of 22 and 65.

Each site visit state also reported that strong advocacy coalitions for behavioral health issues make it politically difficult to manage access. It is anticipated that the new Medicare Part D prescription drug benefit might cause states to reconsider strategies to manage use of behavioral health pharmaceuticals. Because Medicaid agencies no longer manage the drug benefit for dual eligibles, state officials anticipate that behavioral health drugs will represent a large share of their remaining market. As a result, they might consider implementing strategies to manage the remaining high-cost therapeutic classes, including those drugs used to treat behavioral health conditions.

Among the six site visit states, four strategies to control access to behavioral health drugs were in use, though none widely. Two of the strategies, preferred drug lists and caps on the number of prescriptions a beneficiary can have filled each month, are not specific to behavioral health pharmaceuticals. Four states reviewed physician prescribing patterns for behavioral health pharmaceuticals. Kansas, Missouri, Michigan, and California use their federally mandated drug utilization review programs to manage behavioral health pharmaceuticals. In particular, those states work to change the practices of physicians who have a history of suboptimal behavioral health pharmaceutical prescribing patterns.

Lessons learned

Ms. Kaye says that based on the experiences of the site visit states, it is important that states seeking to apply clinical evidence to coverage decisions 1) base coverage decisions on comprehensive, high-quality clinical evidence; 2) involve practicing clinicians and local opinion leaders in applying the evidence to coverage decisions; 3) offer those affected by the policies an opportunity to provide input; and 4) ensure that beneficiaries who need access to nonpreferred drugs are able to access those drugs when medically necessary.

Call Ms. Kaye at (207) 874-6524 or by e-mail: