The finding is important because it appears to shoot down the theory that a new class of antidepressants, which includes big sellers like Prozac and Paxil, are particularly harmful, especially to children.
Regulators in a number of countries have forced manufacturers to include warnings on their labels, a condition not applied to older classes of antidepressants.
There have also been calls for the drugs to be banned for use by children.
But the new study, published in today's edition of the Journal of the American Medical Association, said there is "no substantial difference" in the effect of various drugs on young people aged 10 to 19, though it cautioned that data are limited.
The research also suggests that suicide risk does not increase when patients are weaned off the drugs, another significant finding.
The research team, led by Hershel Jick of the collaborative drug-surveillance program at Boston University, said the seemingly perverse effect -- treatment leading to an increased suicide risk -- is likely explained by the fact that "patients start to take an antidepressant when their depression, which naturally fluctuates over time, is at its worst."
They said while they "cannot exclude" the possibility that the drugs themselves are causing depression to worsen, that is highly unlikely, particularly because two classes of drugs affect the brain differently.
The research is based on data from 160,000 users of four antidepressant drugs in the United Kingdom from 1993 to 1999. The four drugs studied included fluoxetine (sold under the brand name Prozac) and paroxetine (Paxil), which are both selective serotonin reuptake inhibitors (SSRIs); and amitriptyline and dothiepin, both antidepressants known as tricyclics.
Overall, in the first nine days that patients began taking antidepressants, their risk of committing suicide was 38 times that of people who had been taking the drugs for more than 90 days.
(However, there were only 17 suicides, and none in the 10- to 19-year-old group.) Similarly, the new users were four times more likely than the seasoned users to have "non-fatal suicidal behaviours."
Notably, the risks were similar for all four of the drugs.
"Whatever the cause of this striking finding, the current study provides persuasive evidence that it occurs with a similar frequency in the recipients of the four antidepressants studied, and the timing of starting antidepressants therapy is an important factor," the researchers wrote.
In the study, only Paxil users had a slightly higher risk of suicidal behaviour, and the researchers said that may be explained by the fact that it tends to be prescribed to the most severely depressed.
(Tricyclics are generally not used in those cases because patients can overdose, something that cannot occur with SSRIs.)
In an editorial also published in today's edition of JAMA, Simon Wessely of the Department of Psychological Medicine at the Institute of Psychiatry in London, said the study results "confirm that antidepressant prescription is indeed associated with suicidal behaviour and strongly so.
This simply means that antidepressants are being prescribed for the right indication and that they do not immediately eliminate suicide risk. That we knew."
But there was also a growing belief that SSRIs increase the risk still further and the results of the new study, Dr. Wessely said, "do not offer much support for the hypothesis."
He said that, given the anxiety caused by media reports about SSRIs, it is unlikely that the new study alone will restore the confidence of the public and regulatory bodies in these drugs but he nonetheless praised the authors for "providing useful data in what is still a somewhat messy situation."
According to IMS Canada, a company that tracks prescription-drug trends, there were 15.7 million prescriptions last year for SSRIs, and sales topped $990-million last year.
Paxil alone has sales of more than $280-million a year.