The drugs in question -- Zyprexa from Eli Lilly, Seroquel from AstraZeneca, and Risperdal from Janssen Pharmaceutical -- belong to a class of medications known as atypical antipsychotics. They are used to treat schizophrenia and other psychoses, and are commonly prescribed for elderly patients in long-term care facilities.

The conclusion of researchers cast future prescribing habits of physicians treating Alzheimer’s disease into question. According to the researchers, about a third of the estimated 2.5 million Medicare beneficiaries in nursing homes have taken at least one of these medications.

advertisement

The three drug manufacturers noted that the Food ands Drug Administration had not approved these drugs for treatment of Alzheimer’s disease and said they did not market them for this purpose. In fact, prescribing information for the drugs warns that patients with Alzheimer’s-related psychosis "are at increased risk of death compared to placebo."

Still, it is not uncommon for physicians to use promising drugs "off label" -- or not for the sanctioned purpose -- to treat patients when they believe the drug holds some benefit.

"The question is whether these drugs have a place in the treatment of Alzheimer’s patients at all," said Dr. Jason Karlawish, an associate professor of medicine at the University of Pennsylvania who wrote an editorial accompanying the study. "I think the answer is yes, but only for a subgroup of patients who can tolerate them, and in facilities that have the expertise to manage the side effects."

Results of the study were released after researchers followed 421 Alzheimer’s patients for 12 weeks. In each case, the patients suffered from disabling agitation, delusions or hallucinations. Each patient was randomly assigned to receive either dummy pills or one of the three antipsychotic drugs. Doctors adjusted the doses as needed, tracked how long they stayed on the drugs, and noted their improvement, if any.

Experts say that the amount of time a patient spends on a medication is an important measure of its usefulness, because patients often stop taking a drug if it is not doing any good or if the side effects are intolerable.

At the end of 12 weeks, "there were no significant differences between the groups with regard to improvement" on a scale that measured symptom relief, said Dr. Lon S. Schneider, professor of psychiatry, neurology and gerontology at the University of Southern California School of Medicine, and the lead author of the study.

Side effects were closely monitored by the researchers. They included sedation in 15%-24% of the patients and confusion in 6%-18%, both conditions that can increase the risk of falls. Twelve percent of the patients on either Zyprexa or Risperdal experienced Parkinson’s-like symptoms, including tremors.