The Government - or at any rate that part of it represented by Virginia Bottomley - has decreed in the White Paper Health of the Nation that, come the millennium, the suicide rate ought to be reduced by 15 per cent, and it evidently expects the healing professions to do the reducing. Possibly it has been stung by accusations of callous disregard for the underclasses: many people tend to assume that suicide has increased among the poor.

The figure of 15 per cent has presumably been plucked out of the air. Suicide rates may be higher or lower by 2000, but any change will probably have little to do with the Government's campaign. The figures for this century suggest that changes in the suicide rate are often the unexpected consequences of technological developments.

Compared with other European countries, British suicide rates - at around 12 per 100,000 of the adult male population - are low. In 1987, we were less suicidal than Hungarians (66 per 100,000), Austrians (42) and West Germans (25); but more suicidal than the Irish (9) and Spanish (7).

Until the beginning of this century, would-be suicides had a limited choice of methods, most of them unpleasant: stabbing, poisoning and drowning were among the most common. Handguns were rare in Britain and hanging was unpopular, possibly because the public associated it with capital punishment. In countries that used other methods of execution, suicide by hanging was (and is) much more common.

The introduction of railways did not provide much of an aesthetic advance for would-be suicides. In any case, many people are as unenterprising in dying as in living, and intending suicides are much more likely to use facilities that are on their doorstep. That is why the number of British suicides increased rapidly - from about 2,200 in 1890 to 3,500 in 1910 - after the introduction of domestic gas. By the Twenties, the gas oven had become the most common method, and a testimony to Britain's lead in idustrialisation. No other country had such a well-developed domestic gas network. The use of traditional methods remained at much the same level as before.

Overdoses of sedative drugs were not very numerous before the Fifties, possibly because until the NHS, fewer people consulted doctors. Before the First World War, laudanum and other opiates - including heroin - could be bought freely at any pharmacy, but opiate poisoning never really caught on.

Between 1963 and 1975, the replacement of coal gas by natural gas made gas ovens harmless. Suicides fell from a peak of about 5,500 a year in 1963 to about 3,500 in 1975, despite an increasing number of attempted and successful suicides with overdoses of drugs, many of them prescribed to relieve unhappiness. No other country has experienced such a sharp decline in peacetime suicides - because no other country suddenly lost its most popular method. The Samaritans initially claimed the credit, until it was pointed out that during this period the number of suicides by other methods had either increased or remained stable.

After 1975, as people turned to other methods such as car exhausts and hanging - which had lost much of its stigma since the abolition of the death penalty - the suicide rate started to climb steadily by an average of 3.2 per cent a year. But after Margaret Thatcher's arrival in Downing Street in 1979, the annual rate of increase fell sharply to an average of less than 1 per cent for the next five years. There may have been a slight beneficial Falklands effect in 1982-83. Wars generally lower the suicide rate, although military suicides (some of them redefined as acts of heroism) increase.

This levelling-off of the suicide rate has continued. There were 4,321 suicides in 1980, a figure that has fluctuated only slightly since (by 1991 it had fallen to 3,893) - although the suicide figure for women has declined dramatically from 1,692 in 1980 to 886 in 1991.

So if the Government is looking for a success on the 'caring' front to compensate for its Gradgrind image in welfare matters, it could do worse than point to the suicide figures. After all, suicide generally increases at times of high unemployment, as it certainly did during the Thirties, but this time round the rates for men have not increased much. A decline in suicides among the over-forties has been more than balanced by an increase among younger men, but this trend has appeared in most of Western Europe.

Women may have benefited disproportionately from all those part-time jobs that Labour complains about, but there are other possible explanations for the halving of female suicides. Taking impulsive overdoses has long been a predominantly female habit. It has declined somewhat since a peak in 1977, partly because doctors are more cautious in prescribing. Consequently, although overdoses were the commonest method of successful suicide in the Seventies, since the mid-Eighties they have been edged into third place by hanging and car exhaust poisoning.

It rather looks as if women, with their traditional distaste for violent methods of suicide and their equally traditional (if declining) distaste for mechanical activities, are committing suicide less often because they are less likely than men to have the sort of DIY skills that are now necessary. Perhaps the decline in non-fatal self-poisoning since 1977 also indicates a more constructive attitude to the misfortunes of life and an increased willingness to tough things out. If so, this is just as well. A report in the mid-Seventies calculated that if overdoses continued to increase at the existing rate, by 1984 every female medical bed would be filled with a recovering self-poisoner.

Although the increase in car exhaust suicides may be evidence of growing prosperity, the rise is much greater than the growth in car ownership during the same period. Interestingly, catalytic converters, now mandatory on new cars produced in the European Community, greatly reduce emissions of carbon monoxide, the lethal component of exhaust fumes and pre-North Sea gas. This device, introduced for purely environmental reasons, may eventually make a significant impact on suicide rates in the way that North Sea gas once did; if so, europhobes will presumably complain that Brussels, having made life difficult for them, is now trying to do the same for death.

Paracetamol overdoses currently account for about 300 suicides a year. They could, in theory, be virtually abolished if the manufacturers incorporated the antidote acetylcysteine into all tablets. Unfortunately, that would greatly increase the price, so paracetamol makers are unlikely to be enthusiastic.

Suicide and alcohol abuse are related. Increasing the price of alcohol would probably reduce both conditions somewhat, but that is unlikely to find much favour either. Every year two or three hundred people use antidepressants to commit suicide. The higher cost and lower toxicity of newer drugs should act as a deterrent to their use.

Even if measures such as increasing the price of alcohol or adding the antidote to paracetamol reduced the suicide rate by 2000, more people might resort to methods with higher success rates, such as hanging or using uncatalysed exhausts. If that happened, some currently unsuccessful suicide attempts might end in death. Mrs Bottomley is a trained and experienced social worker. She would not be the first member of that well-meaning profession to achieve the opposite of her intentions.

The author is a consultant psychiatrist.