NEW YORK (Reuters Health) - Newer antipsychotics are often prescribed for conditions they aren't approved to treat -- with questionable benefit, according to new research.

The medications, known as atypical antipsychotics, include risperidone (sold as Risperdal), olanzapine (Zyprexa), quetiapine (Seroquel) and aripiprazole (Abilify).

While those drugs have been approved in the United States for a few psychiatric conditions, such as schizophrenia and bipolar disorder, doctors may also prescribe them "off-label" to other patients, including people with substance abuse and eating disorders -- typically when they haven't responded to more standard treatments.

"There are several conditions in psychiatry that are pretty difficult to treat," said study author Dr. Alicia Ruelaz Maher.

"Often in psychiatry we think, if something works for one condition, it could possibly be effective for another one," Maher, from RAND Health in Santa Monica, California, told Reuters Health.

People with anxiety or obsessive-compulsive disorder, for example, are often treated with antidepressants known as selective serotonin reuptake inhibitors (SSRIs). But those don't always work.

More and more, doctors are prescribing atypical antipsychotics off-label, experts say.

Yet, Maher noted, "there's a general lack of evidence" on whether atypical antipsychotics can help patients who don't have schizophrenia or bipolar disorder.

She and her colleagues looked back at the scientific literature and combined data from 162 studies that compared an atypical antipsychotic to a drug-free placebo pill for conditions that fell outside the drug's approved uses. They found another 231 studies that kept track of side effects linked to the drugs. Nearly all of the research was funded by drugmakers.

Taken together, the studies showed that atypical antipsychotics had no effect in patients with eating disorders or drug and alcohol abuse, and it wasn't clear if they helped people with personality disorders or post-traumatic stress disorder.

The drugs did have a small but consistent benefit for dementia patients with psychotic symptoms, however.

More people with an anxiety disorder taking Seroquel in particular got better compared with patients who took a placebo. And for those with obsessive-compulsive disorder, Risperdal worked better than a placebo. Still, those specific findings came from just three studies each.

Dr. Randall Stafford, who has studied trends in antipsychotic use at Stanford University and was not involved in the new study, cautioned that most research hasn't evaluated these drugs for off-label conditions for more than a few months at a time.

"This may result in a more favorable assessment than would have been the case if patients had been followed longer," he told Reuters Health in an email. "In fact, the studies included do not tell us about the efficacy and safety of these drugs as they are typically used in the real-world; that is, for long periods of time."

According to the new findings, published in the Journal of the American Medical Association, the drugs also had side effects. For instance, one in 10 elderly people treated with Zyprexa developed tremors and one in 53 on Risperdal suffered a stroke. In agreement with the strong warning on the drugs' labels, one in 87 seniors died from taking them.

Younger people on the drugs experienced weight gain, fatigue and other side effects.

Maher said that the decision to use one of these drugs for a non-approved condition should be based on a discussion between patients and their doctors, especially because some people may be more susceptible to side effects than others.

"Each individual patient has to be considered as opposed to, 'This is good for this condition,'" she said.

If a doctor is prescribing an atypical antipsychotic for anxiety or an eating disorder, for example, "you want to ask him or her on what basis they expect a response, as well as how safe you as an individual may be form these various side effects," Maher advised.

SOURCE: http://bit.ly/rnslfz Journal of the American Medical Association, online September 27, 2011.