Compliance, Costs Closely Linked in Schizophrenia

There appears to be a definitive relationship between compliance with medication regimens for schizophrenia and the economic costs of the disease: lower rates of compliance result in higher costs.

So concluded the authors of a report describing a literature search of studies published from 1995 to 2002 evaluating patient compliance with treatment for schizophrenia. The report appears in the April issue of the APA journal Psychiatric Services.

John Kane, M.D., a co-author of the report, said that the findings have implications both for clinicians and for public policymakers.

“There is a continuing concern about controlling health care costs, and if one wants to develop a more rational approach to treatment, one needs to look at the opportunities to spend the health care dollar more wisely,” Kane told Psychiatric News.

He is chair of the department of psychiatry at Zucker-Hillside Hospital and a professor of psychiatry, neurology, and neuroscience at the Albert Einstein College of Medicine in New York City.

“We know that rates of noncompliance are very, very high,” Kane said. “But we are not very good at predicting who will be nonadherent or even [knowing] when nonadherence occurs. As a result, there are many patients who stop taking their medication and relapse as a result. The cost of that is enormous because we are dealing not only with the direct costs of rehospitalization, but also with an increase in other costs, including those related to visits to emergency rooms, police courts, loss of job, and potentially aggressive or violent behavior.”

Kane and his colleagues reviewed studies published from 1995 to 2002 that evaluated compliance with treatment for schizophrenia. Literature searches were conducted for the period using MEDLINE, EMBASE, and Current Contents. The search terms were “schizophrenia,” “compliance,” “relapse,” and “economic costs.” The study was funded by Novartis.

Despite the clear relationship between compliance and cost, the authors stressed that the full economic implications of noncompliance were difficult to ascertain because of inadequacies in the reporting of compliance rates and outcomes of treatment over time.

They also said that better collection of data on longer-term clinical outcomes would improve future economic evaluations of schizophrenia.

Kane and colleagues noted, as well, that several countries have published guidelines to standardize methods for the production of high-quality economic analyses of medication use. These include Australia, Canada, and the Netherlands. In the United Kingdom, economic evidence has become a formal requirement in the approval process for new drugs, they noted.

Such analyses, which require long-term follow-up of patients, may be harder to achieve in the American health care system.

“Some countries have a better-integrated health care system looking across the population,” Kane noted. “Our system is so fragmented, with multiple payers involved. There is not a consistent or effective effort to prevent illness or recurrence over a long period of time. This might be associated with substantial impact both financially and from a public health perspective.”

Psychiatrist Robert Rosenheck, M.D., who reviewed the study, said the conclusions are sound, and he particularly highlighted the authors’ call for better models to assess the economic impact of noncompliance and associated factors.

“Noncompliance may be one of many mediating mechanisms that influence cost,” said Rosenheck, the director of the VA Northeast Program Evaluation Center in West Haven, Conn., and a professor of psychiatry and public health at Yale School of Medicine. “The most important issue is not so much compliance, but the cost-effectiveness of different regimens. We are increasingly doing cost-effectiveness studies of medications.”

Kane noted that a crucial clinical and public-policy issue related to compliance and cost is the development of long-acting injectable medications. “These are a very effective strategy for reducing the risk of nonadherence and, at a minimum, enabling clinicians and families to know immediately when someone does become nonadherent,” he said.

He said that when patients are on oral medication, it may be weeks or months before anyone becomes aware that they have stopped taking it. “Often, it is only when there is an exacerbation of symptoms or a relapse that it becomes apparent,” he said.

Kane said that a new generation of long-acting injectable antipsychotics is not available yet. “The currently available ones are conventional antipsychotics that are being replaced,” he said. ▪