In the past six months, behavioral healthcare has seen broadly influential, highly publicized outcomes in both medical research and public health policy. These “real-world” studies have implications for improving behavioral healthcare nationwide.
In December 2005, Congress passed HR 4579, extending to December 2006 the Mental Health Parity Act. Originally passed in 1996, the parity law mandates that health insurers not place caps on mental health benefits without doing the same for medical and surgical benefits. From the late 1970s until passage of the parity law, health insurers had imposed increasingly stricter limits on benefits for behavioral healthcare needs.
Parity proponents argue that such restrictions place an undue financial burden on those in need of services and create a social climate of discrimination and stigmatization. Opponents argue that without these restrictions, healthcare costs would rise precipitously. These two sides recently clashed over the Health Insurance Marketplace Modernization and Affordability Act (HIMMA), which, among other things, would have overridden state health insurance laws that mandate mental health parity. HIMMA was defeated in the Senate in May.
Contrary to pessimism about the fiscal burden of parity, a recently published study conducted by the Parity Evaluation Research Team showed that parity implementation, when coupled with care management, did not lead to higher costs or increased use of services and that, when participants became ill, they paid less.1 These findings are compelling, as the study was a controlled evaluation of the Federal Employees Health Benefits Program, a large system that includes mandated parity requirements.
The general consensus among parity advocates is that we should remain cautiously optimistic. The parity law extension is a step in the right direction, but is limited in that it does not apply to benefits for substance abuse or chemical dependency, includes a small employer exemption (fewer than 50 employees), and has an increased cost exemption (if application of the parity provision results in an increased cost of at least 1% to a group health plan). The existing parity law also does not require full parity for mental health benefits.
CATIE is a novel contribution to the research on medication treatment for schizophrenia. Its design was meant to mimic actual clinical practice and included a real-world indicator of effectiveness—length of time taking a specific medication. Phase one2 and now phase two3,4 have been published.
The newest information to emerge continues to challenge prevailing wisdom. Phase one pitted perphenazine, a first-generation antipsychotic, against risperidone (Risperdal), olanzapine (Zyprexa), quetiapine (Seroquel), and ziprasidone (Geodon). The biggest surprise was that perphenazine's side-effect burden was modest, and perphenazine was considered at least as effective as the four atypical medications.
In phase two, patients who discontinued their phase one drug were given another drug. The remarkably high discontinuation rates in phases one and two stand as another humbling finding: 1,052 of 1,493 participants discontinued their phase one medication prematurely (before phase one ended).
Participants who stopped their phase one medication because of lack of response were offered entry into the “efficacy arm” of phase two, in which they were randomized to clozapine or another atypical. In the “tolerability arm” of phase two, participants who stopped their phase one medication for any other reason, plus those who refused clozapine, were randomized to one of the atypicals that they had not taken previously.
In phase two's efficacy arm, researchers found that people stayed on clozapine, with all its risks and requirements, longer than olanzapine, risperidone, and quetiapine. Ziprasidone was not included in this arm. In the tolerability arm, patients stayed on olanzapine and risperidone for significantly longer periods than patients stayed on quetiapine and ziprasidone.
Overall, the winners from the CATIE study are perphenazine and clozapine. Current practice largely ignores both of these effective and cheap drugs. Risperidone and olanzapine performed reasonably well. The practice implications are numerous. Most obviously, the findings inform an algorithm for deciding which medications to suggest, and the need to support an adequate trial while supporting consumer choice. But perhaps equally important is the opportunity for a new dialogue between prescribers and consumers around the topic of “effective” treatment.
STAR*5,6 is the nation's largest study of medication treatment for depression following the failure of an initial treatment attempt. The overall findings showed that 33% of depressed participants who were not responsive to a initial 14-week trial of an antidepressant (citalopram, Celexa) reached remission with augmentation with a second antidepressant (within 14 weeks). Twenty-five percent of initial nonresponders reached remission (within 14 weeks) when medication was switched for another.