July 3 (HealthDay News) -- Taking prescription medications can help reduce violent behavior in some schizophrenia patients, Duke researchers report.

It included 1,445 schizophrenia patients randomly selected to receive one of five antipsychotic medications: olanzapine, perphenanize, quetiapine, risperidone or ziprasidone.

Patient violence declined significantly when patients took their medications as prescribed, but only among patients whose prior risk for violence could be linked to psychotic problems. There was little or no improvement in a subgroup of patients with a history of childhood conduct problems who were more likely to be violent at the start of the study.

The study also found that older antipsychotics are as effective as newer drugs in reducing violent behavior.

"Contrary to the expectations of many clinicians and some research, this study found no benefit for newer medications over an older medication in reducing the risk for violence over the six-month study period," study author Jeffrey Swanson, a professor in psychiatry and behavioral sciences, said in a prepared statement. "In fact, one of the newer medications, quetiapine, performed worse than the first-generation drug perphenanize."

The findings were published in the July issue of the British Journal of Psychiatry.

"This is the first large randomized controlled study to compare the effectiveness of several commonly prescribed medications for schizophrenia on reducing community violence," Swanson said. "Serious violent behavior is not frequent among people with schizophrenia, but when it does occur, the results can be costly and tragic."

"In the past, we've not understood very well why a small proportion of patients with schizophrenia become seriously violent, while most do not -- and why medication seems to prevent violent behavior in some and not others," study co-author Marvin Swartz, a professor of psychiatry, said in a prepared statement.

"These findings tell us that people with schizophrenia may behave violently for reasons not directly related to their mental illness. If that's the case, then treatment for psychotic symptoms alone may not eliminate the risk of violence," Swartz said.

SOURCE: Duke University Medical Center, news release, July 1, 2008