Abstract

In 7735 men aged 40-59, selected at random from general practices in 24 towns throughout Britain, pronounced differences were noted in the prevalences of smoking and drinking between the social classes. Social class differences also existed for frequency and quantity of drinking, type of beverage, and several aspects of smoking behaviour. Increasing amounts of smoking were associated with higher prevalences of moderate to heavy drinking, particularly in daily rather than weekend drinkers. Between drinking groups, however, the relation with smoking was more U-shaped, with light and heavy drinkers smoking more than moderate drinkers. The lowest rates of moderate to heavy smoking were observed in frequent light drinkers, particularly in the nonmanual workers. The proportion of moderate to heavy drinkers was no higher among ex-cigarette smokers than among current smokers. When the data were examined by town of residence social class differences persisted. Controlling for social class still showed pronounced differences between towns in both smoking and drinking behaviour. These data confirm that town of residence and social class have independent effects on smoking and drinking. The established regional and social class differences in cardiovascular disease may be due in part to the independent influences of town and social class on smoking and drinking behaviour.