Smoking, however, remains the 'big killer'. Scientists who yesterday presented the good news on alcohol emphasised that while moderate drinking may reduce mortality, 'moderate or immoderate smoking kills'.
The 13-year study of 12,000 doctors found that one or two drinks a day appeared to cut mortality by at least one-sixth when compared with abstainers. Alcohol also protected the men against some common diseases, including heart and lung disorders.
But the scientists warned that a fine line existed between the benefits of alcohol and the risks. Men who had several drinks a day experienced more deaths than moderate drinkers from heart and blood vessel disease, respiratory disorders, and alcohol-related illness such as cirrhosis and some cancers.
The type of alcohol drunk by the men was not included in the study but the interaction between pure ethanol (alcohol) and blood chemistry appears to be responsible for the beneficial effects.
Professor Richard Peto, of the Imperial Cancer Research Fund's Cancer Studies Unit at Oxford, dismissed suggestions that the results, published in the British Medical Journal, would be seen as a 'licence to booze' by some men, and he warned that the study did not include young people or women. 'One or two glasses of wine a day is good for you, but one or two bottles of wine isn't . . . If men consistently have more than about two or three drinks a day then this increases their overall death rates in middle or old age. But, men who don't drink at all have higher death rates in middle or old age than men who average one or two drinks a day.'
Eric Appleby, director of Alcohol Concern, welcomed the findings which concur with current advice. Sensible drinking limits are currently a maximum of 21 units a week for men (one unit is a glass of wine, a measure of spirits or half a pint of beer or lager) and 14 units for women. The limits are under review and may be raised.
Professor Peto, together with Professor Sir Richard Doll, monitored the drinking habits and health of the doctors, who were among 34,000 initially recruited to a study of smoking and death which began in 1951. It is the world's longest-running such project.
Presenting the 40-year follow-up (to 1991) results yesterday which are also published in tomorrow's issue of the BMJ, Sir Richard, the epidemiologist who with Austin Bradford Hill first showed smoking as a cause of lung cancer, said that it was clear that half of all regular smokers will eventually be killed by their habit.
Smoking affected 25 different causes of death and increased 24 of them. It appeared to inhibit Parkinson's disease, Sir Richard said, but 'overall, smoking probably caused about 100 times as many deaths as it prevented'. The findings emphasised the benefits of giving up smoking. Doctors who stopped in middle-age substantially increased their life-expectancy.
Virginia Bottomley, the Secretary of State for Health, yesterday pledged her support for an increase in tobacco taxation, claiming that it was one of the most effective ways to reduce smoking.