How Much Should Psychiatrists Weigh Costs When Prescribing?
Abstract
A study in Psychiatric Services in Advance finds the long-acting injectable (LAI) haloperidol decanoate to be more cost-effective than a second-generation LAI. This raises difficult questions about clinical choice when one intervention is marginally more effective, but significantly more costly, than another.
To what extent should clinicians consider “cost-effectiveness” in the choice of treatment?
It’s a question Psychiatric News put to several experts in light of a new report in Psychiatric Services in Advance that found the long-acting injectable (LAI) paliperidone palmitate (PP), a second-generation antipsychotic, is not as cost-effective as the first-generation LAI haliperidone decanoate (HD), despite having a slight advantage in terms of clinical effectiveness.
Robert Rosenheck, M.D., a professor of psychiatry and public health at Yale Medical School, and colleagues conducted a double-blind, randomized 18-month clinical trial at 22 clinical research sites in the United States to compare the cost-effectiveness of HD and PP. A total of 311 adults with schizophrenia or schizoaffective disorder who had been clinically determined to be likely to benefit from an LAI antipsychotic were randomly assigned to monthly intramuscular injections of HD (25 mg to 200 mg) or PP (39 mg to 234 mg) for up to 24 months.
What Is a Quality-Adjusted Life Year?
Cost-effectiveness research is a highly technical and somewhat rarified area of health services research that looks at the relative costs of a “quality-adjusted life year” (QALY).
Eric Slade, Ph.D., an associate professor of psychiatry at the University of Maryland and an economist who specializes in mental health services research, explained that the QALY is a metric of health and morbidity developed by medical "decision scientists" and economists. “The concept is that all states of morbidity can be valued on a 0 to 1 scale, where 0 represents death and 1 represents ‘perfect health,’” he said. “One quality-adjusted life year is interpreted as one year of perfect health.”
Slade explained that QALY ratings are usually based on a “preference-weighted” quality-of-life scale, of which there are several. Preference weighting means that a sample of individuals is asked a series of questions about the relative values they would place on different health limitations—for instance, vision impairment compared with mobility impairment. The answers to these questions are then used in an analysis to value different states of health.
In the Rosenheck study, the authors used a special version of the standard methodology to capture the value of changes in psychotic symptoms, translating scores on the Positive and Negative Symptom Scale (PANSS) scores into QALYs.
Rosenheck and colleagues found that compared with haloperidol decanoate, peripalidone palmitate was associated with .0297 greater QALYs over 18 months, or approximately .02 QALYs over 12 months—an improvement of 2 percent of one year of perfect health or approximately one week of perfect health during a full year.
That’s a relatively small benefit. And the cost of that small benefit for paliperidone is very high. The added cost of producing one additional year of perfect health using paliperidone versus haloperidol was over $500,000, Slade said.
According to Slade, $100,000 per QALY is often used by policymakers as a benchmark. “If something costs more than $100,000 per QALY, it is not considered cost-effective,” he said. “If something costs less than $100,000, it is considered cost-effective. This threshold is somewhat arbitrary and is frequently debated. However, many prevalent health care services have been shown to cost less than $100,000 per QALY. Interventions like vaccinations are among the most cost-effective in health care, costing generally less than $100 per QALY.”
Results showed that PP was associated with a small but statistically significant health advantage over HD, as measured by “quality-adjusted life years” (QALY), drawing on scores on the Positive and Negative Symptom Scale (PANSS). The cost of PP ran on average $2,100 more per quarter for inpatient and outpatient services and medication compared with HD. Statistical analysis that divides incremental costs by incremental benefits generated an incremental cost-estimate ratio of $508,241 per QALY for PP compared with HD—well above the $100,000 per QALY benchmark often used by policymakers to determine if something is cost-effective (see sidebar).
“The results of this study should encourage consideration of older, less expensive drugs, such as HD,” Rosenheck and colleagues wrote. “Used at moderate dosages in this study, HD’s overall effectiveness and tolerability were only slightly worse, as reported here, than those of PP, and it had clear advantages in cost-effectiveness. … A rational policy for treatment of chronic schizophrenia might limit use of the more expensive LAIs to patients who do not benefit from or cannot tolerate HD.”
In an interview with Psychiatric News, Rosenheck said the results should be useful to payers and policymakers. However, he emphasized that individual clinicians also have a responsibility to pay attention to cost-effectiveness.
“It’s true that research on effectiveness and cost-effectiveness tells us about averages, not about individuals, and this kind of study does not tell the individual practitioner how to prescribe individualized therapy for every patient,” he said. “On the other hand, I think psychiatrists tend to be more sensitive to pharmacologic risks than to economic risks. Many patients with serious mental illness are poor and insurance coverage is uneven. A study like ours allows clinicians to say, ‘I can prescribe you an effective medicine that fits your budget.’”
He added, “If psychiatrists don’t play a role in developing a scientific basis for cost-effectiveness, then the only people who are setting the agenda are those whose main interest is profit. Cost- effectiveness becomes especially important when there is a treatment that is marginally more effective but significantly more costly. Then we have to face the question of the monetary value of health benefits. That is extremely painful and extremely hard to do scientifically.”
William Carpenter, M.D., a professor of psychiatry at the University of Maryland and editor of Schizophrenia Bulletin, has been critical of pharmaceutical companies that market “me-too” drugs—medications that have virtually the same mode of action as other drugs in the same class and generally differ only in side-effect profile, but are often significantly more expensive than older drugs that are as effective.
“It’s very good to have this type of head-on comparison to provide perspective and guidance for decision makers on mental health services,” he said. Carpenter suggested that clinicians have an ethical duty to pay attention to cost-effectiveness.
Others who reviewed the report for Psychiatric News agreed it offers valuable information for payers and policymakers, but questioned the degree to which individual clinicians should apply cost-effectiveness in the treatment of individual patients.
Eric Slade, Ph.D., says he believes cost-effectiveness data are primarily useful for mental health system administrators responsible for population-level policy.
“This is likely to be useful information for payers and policymakers since it provides a rational basis for determining that populations of patients should first be treated with HD before they receive treatment with PP,” said Stephen Marder, M.D., a professor of psychiatry and behavioral health and the director of the section on psychosis in the Department of Psychiatry at the University of California, Los Angeles.
“My concern is that this information may not translate well for a prescriber and a patient who need to deal with other considerations,” Marder said. “If the patient is concerned about weight gain, HD was shown to have an advantage. If an individual is vulnerable to developing discomforting akathisia or if there is a worry about tardive dyskinesia, the calculation of a QALY loses its meaning and interferes with a decision that is in the best interests of an individual.”
Eric Slade, Ph.D., an associate professor of psychiatry at the University of Maryland, agreed. “I’m not sure that clinicians should consider cost,” he said. “Their job is to identify the best course of treatment for each patient. The cost of antipsychotic treatment, particularly injectables, is generally borne by Medicaid and other public payers. Cost-effectiveness findings are aimed at mental health system planners, typically state mental health administrations or their equivalent at the local, state, or federal levels.”
Slade added, “The cost-effectiveness findings from the Rosenheck study imply that from a cost-effectiveness perspective, it may be worth having Medicaid provide some regulatory guidance regarding patient access to PP. Although PP results in slightly greater quality of life, the cost of achieving that improvement is greater than the conventional threshold of acceptable cost.”
Jim Sabin, M.D., says decisions about resource allocation are best made within systems of care that accommodate the imperatives of population health and the needs and preferences of individual patients.
Jim Sabin, M.D., is the director of the ethics program at Harvard Pilgrim Health Care and has written extensively about the ethics of resource allocation within systems of care. He hailed the Rosenheck study as “remarkably and admirably explicit about cost issues and unembarrassed about it.”
He continued, “We have been so reluctant to entertain these questions in the United States, as if health and money can’t be thought of together. But a symptom of this reluctance is what has happened to our health care costs.”
Sabin, who is also a professor in the departments of Population Medicine and Psychiatry at Harvard, acknowledged a “serious conundrum” translating cost-effectiveness—which serves as a useful guide to policymakers concerned with population health—into clinical decision-making for the individual patient. But he emphasized that such decisions can be made ethically and rationally within a system of care (as opposed to ad hoc decision-making by individual prescribers) that accommodates the imperatives of population health and the individual.
“A well-functioning system serves both perspectives,” he told Psychiatric News. “Balancing population health and individual considerations can open up very thorny and challenging questions that are ideally addressed within systems in a thoughtful way, with no pre-existing rules about how powerfully an individual preference should drive the allocation of resources.”
An organized, systemic approach to resource allocation can also avoid singling out particular groups for cost-cutting, he said. “My concern is that a stigmatized group, such as people with severe schizophrenia, could become the focus for resource allocation applying cost-effectiveness approaches in ways that aren’t applied to less stigmatized but equally or more costly areas. That’s a real social risk.” ■
Read a related blog post by Sabin titled “When Is Rationing Ethically Acceptable?”
