The authors of a new comprehensive analysis of antidepressants for children and teenagers say the benefits of treatment trump the small risk of increasing some patients' chances of having suicidal thoughts and behaviours.

The risk they found is lower than the one the U.S. Food and Drug Administration identified in 2004, the year the agency warned the public about the drugs' risks for children.

After the warning, U.S. youth suicides increased and some mental health experts said reluctance to try antidepressants might be to blame.

The new analysis includes data from seven studies that were not part of the FDA analysis, including two large pediatric depression trials that were unavailable three years ago.

Researchers analyzed data on 5,310 children and teenagers from 27 studies. They found that for every 100 kids treated with antidepressants, about one additional child experienced worsening suicidal feelings above what would have happened without drug treatment.

In contrast, the FDA analysis found an added risk affecting about two in 100 patients.

There were no suicides in any of the studies.

The antidepressants included Prozac, Paxil, Zoloft, Celexa, Lexapro, Effexor, Serzone and Remeron.

"The medications are safe and effective and should be considered as an important part of treatment," said study co-author Dr. David Brent of the University of Pittsburgh School of Medicine. "The benefits seem favourable compared to the small risk of suicidal thoughts and behaviour."

Antidepressants worked best when used to treat anxiety, the analysis found, with 69 per cent improving on antidepressants, compared to 39 per cent on dummy pills.

Antidepressants worked moderately well for treating obsessive-compulsive disorders in young people, 61 per cent compared with 32 per cent for placebos. They worked less well but were still effective in treating depression, 61 per cent versus 50 per cent.

Adolescents responded better than children to treatment for depression and anxiety, the researchers found. They also found that only Prozac worked better than dummy pills in depressed children younger than 12.

Careful monitoring needed

Effectiveness of the drugs was measured in the studies using widely accepted rating scales.

The analysis appears in Wednesday's Journal of the American Medical Association.

Dr. John March, chief of child and adolescent psychiatry at Duke University Medical Center, welcomed the study as "the most comprehensive analysis of the data yet put together."

He said the suicidal behaviour risk, although lower than that found by the FDA, demands that doctors and families watch for warning signs.

"You can't say, 'Take these and call me in six weeks,'" March said. "You have to monitor carefully the benefits and adverse events."

The study didn't measure the effects of talk therapy, March pointed out. He said cognitive behavioural therapy used with antidepressants can lower suicide risk and speed up recovery for depressed youth.

The study was supported by grants from the U.S. National Institute of Mental Health and the Robert Wood Johnson Foundation.