Smoking habits and related attitudes were assessed in a sample of 4059 11- to 16-year-olds who also identified their best friends from among their fellow respondents. Subjects' responses were directly collated with those of their friends and indicated a clear covariation of smoking status (controlling for sex and age) as anticipated from previous research in which adolescents have been asked to report on the smoking habits of their friends. Such covariation, however, was not specific to smoking habits, but generalized to related measures of attitude and normative beliefs, alcohol use, health locus of control, school performance, spending habits and socio-economic status. Similarities on these other attributes were much the same, whether or not friends shared each others' smoking habits. It is concluded that these data argue against a simplistic view of unidirectional 'peer group influence' and invite an interpretation of friendship choice based on multiple dimensions of similarity.

Risk factors for the uptake of cigarette smoking were examined prospectively in 2159 non-smoking secondary schoolchildren aged 11-13 who participated in a survey in 1983 and were followed up 30 months later, by which time 35 per cent had taken up smoking. In a multivariate logistic model, the strongest predictors to emerge were prior experimentation with cigarettes and sex, with more girls (41%) than boys (30%) starting to smoke. Other predictors of taking up smoking were being uncertain about smoking in the future, reporting having been drunk, having a boy or girl friend, believing teachers and friends would not mind if they took up smoking, and giving lower estimates of prevalence of smoking among teachers. Parental smoking behaviour and attitudes, beliefs about the effects of smoking on health, opinions about smoking and perceived strictness of parents did not predict take up of smoking when other variables were controlled for. The odds of taking up smoking varied from 0-24 (risk = 0 19) for a child with the most favourable combination of risk factors to 3-49 (risk = 0178) for a child with the worst prognosis. These results differ from those of many cross sectional studies and hence indicate the importance ofa prospective approach to this type of research.Most cigarette smokers take up the habit during adolescence. A recent UK national survey indicated that the proportion of regular smokers increases sharply between the first and fifth years of secondary school and that 40% of sixteen year olds smoke.' It is important to be able to identify those children most likely to take up smoking during these school years.The following factors have emerged from British studies as being associated with smoking: parental smoking and precept; sibling and peer smoking; social class; pocket money; alcohol consumption; truancy; lower academic achievement; participation in social activities associated with older and mixed sex peer groups; rejection ofthe health hazards associated with smoking; and personality characteristics such as rebelliousness, neuroticism and extraversion.2-9 However, most of these studies were cross sectional in design and it is not clear whether the factors identified actually predated the uptake of smoking or were merely concomitants. Also these cross sectional studies have not been able to separate factors associated with smokers who have recently acquired the habit from those affecting long standing smokers.'0 Prospective studies should be more informative, although when smokers are included in the baseline sample,7 they suffer from similar interpretative problems.Prospective studies of adolescent smoking carried out in the United States have found future intention to smoke to be a predictor of the onset of adolescent smoking in addition to various combinations of the above factors.'"-"' Results obtained from univariate analyses have conflicted with those derived from a multivariate approach.'8 This paper presents the results of a prospective

Saliva cotinine concentrations in 569 non-smoking schoolchildren were strongly related to the smoking habits of their parents. When neither parent smoked the mean concentration was 0 44 ng/ml, rising to 3-38 ng/ml when both parents were cigarette smokers. Mothers' smoking had a stronger influence than did fathers' (p <0 01). In addition, there was a small independent effect of number of siblings who smoked (p <0 01). The dose of nicotine received from fathers' smoking was estimated as equivalent to the active smoking of 30 cigarettes a year, that from mothers' smoking as equivalent to smoking 50 cigarettes a year, and that from both parents smoking as equivalent to smoking 80 cigarettes a year. This unsolicited burden may be prolonged throughout childhood and poses a definite risk to health.

The self-reported smoking habits and intentions of over 10,000 adolescents, aged 11 to 16 years, were related to a variety of attitudinal and normative variables. Current or previous smoking experience was a more important predictor of future intention to smoke than attitude or subjective norm (others' perceived tolerance of subjects' smoking). Parental opposition to smoking was a more important direct predictor of subjects' smoking intention than parental smoking behaviour.

Results from a survey of 10,579 schoolchildren aged 11-16 years indicated various relationships between age, smoking behaviour, health locus of control beliefs, expressed concern with different consequences of illness and beliefs in the efficacy of different strategies for staying healthy. In particular, smokers, compared with non-smokers, showed less belief in the importance of 'powerful others' or 'personal control' but more belief in the importance of 'chance' as an influence on health outcomes. With regard to consequences of illness, smokers cared more than non-smokers about not being allowed to see friends, but less about getting behind in school or being a nuisance to others; they were also less convinced of the benefits of preventive health behaviours such as eating good food and taking exercise. Implications for health education approaches are discussed.