Journalists are increasingly aware of the negative impact that can result as a consequence of the way they report cases of suicide, writes Dr Raj Persaud

The who calculates that around the world approximately one million people take their own lives annually. The recent attempt to calculate this disturbing figure coincides with the first real exertion by public health agencies globally to actively commit themselves to suicide reduction.

(Incidentally the phrase "commit suicide" is sometimes regarded by suicidologists as unhelpful, because it is linked to the pre-1960s stigmatised view of suicide as a crime. One commits a crime so should the word commit be used with suicide?)

However, after many years of worrying increases in rates of youth suicide in most Western countries, a remarkable decline is now occurring. But the motives behind such trends remain deeply puzzling to suicide experts.

For example reductions in overall suicide rates have occurred not only in Finland, Sweden, Norway and Denmark which have a structured suicide reduction strategy, but also in Hungary and The Netherlands, which don't possess anysuch national prevention programme.

Amidst this puzzle of how to reduce suicide rates comes some startling research which suggests that one way forward is to alter the way the media reports suicide itself.

The possible effect of media accounts of self-harm on actual suicide is termed the "Werther Effect" after Goethe's novel which was blamed as having lead several young men to take their lives, in the same way as young Werther, after the publication of the book.

Psychiatric research from all round the world has since demonstrated a temporary rise in suicide rates following newspaper reports on actual suicides, as well as when suicide occurs in soap operas or TV serials.

Some studies confirm that suicides, motor vehicle deaths (which are possibly disguised suicides) and non-fatal accidents all rise in the week immediately following soap opera suicide stories and, furthermore, these rises are statistically very significant.

Other research has found a "dose-response" relationship - in other words the more prominent the media report on suicide the more the impact on real suicides. For example investigators have found imitation is more likely when the suicide is not only front page, but covered by large headlines (particularly with the word suicide prominent), heavily publicised, with a lengthy report containing pictures. Another factor is if the suicide concerns celebrities.

Some of the most recent research into this effect was conducted by Dr Elmar Etzersdorfer and Dr Gernot Sonneck of the University of Vienna who measured the number of suicides occurring in Vienna on the subway system through the 1980s and 1990s.

Shortly after the subway system in Vienna was built in 1978 it became an increasingly popular means of suicide in the 1980s. The mass media reported these events in very dramatic and extensive ways with headlines and pictures of the deceased. The Austrian Association for Suicide Prevention began to study this mass media reporting and concluded that many suicides could be ascribed to imitations of the behaviour found in these reports.

The theory is that such accounts form a kind of "natural advertisement" for the idea of suicide. A person in a suicidal crisis is ambivalent and therefore possibly prone to suggestions in both directions. A media report that allows one to identify with the person described may work as the last trigger for the decision to commit suicide. Also a person who is constricted in a severe crisis and cannot think of a way out could find the solution for his unbearable solution formulated in a media report.

As a result of these theories a press campaign was launched in mid-1987 by the Austrian Association for Suicide Prevention informing journalists about possible negative consequences of their reporting and offering alternative ways of dealing with those issues.

The effect of the campaign was that media reports changed markedly and immediately. Reports on suicidal behaviour in general became much more moderate than before and for the first time several subway suicides were even left unreported.

After the media campaign in mid-1987 a sharp drop in suicides and attempts was recorded. The decrease from the first half of 1987 to the second half was 84.2 per cent for suicides and attempts taken together. In the subsequent immediate years the number of suicides and attempts stayed low.

Other media organisations around the world are now beginning to take the issue seriously - for example BBC guidelines say "Reported suicides may encourage others. We should"not add to this risk"Reports should usually avoid details of method"unless there is a good reason to go into detail." The Independent Television Commission's code makes similar points, urging producers where appropriate to seek advice from such organisations as Samaritans.

However, there are occasional lapses, such as the recent publication on the front of the G2 section of The Guardian newspaper of a photograph of a young woman who had killed herself a short while previously. What was particularly tragic about this story was that the family had not given permission for the use of the picture on this occasion, and had also been quoted in the subsequent article without authorisation. The Guardian's readers' editor later formally accepted his newspaper should change the way it reports suicides.

Of course we still don't know if recent reduction in national suicide rates have something to do with more responsible media handling of the issue, but there is also a sense in which the solution is not that media should shun the question entirely. As a society we need to discuss suicide in an honest manner in order to educate the public about it.

For example one theory about why suicide rates among young people went down in some countries following the death of Kurt Cobain, lead singer of the grunge metal band Nirvana, was that his lover, Courtney Love, publicly issued a response rejecting the sentiments in his suicide note.

After all, if we were to ban all media reports of suicide, where would that leave us over public performances of Romeo and Juliet, with its famous suicide scene?