Is smoking a communicable disease? Effect of exposure to ever smokers in school tutor groups on the risk of incident smoking in the first year of secondary school

Objective: To estimate the effect of joining a first year secondary school tutor group with a high prevalence of ever smoking on the risk of incident smoking in schoolchildren. Design: Cross-sectional questionnaire survey. Setting: 10 randomly selected secondary schools in Nottinghamshire, UK. Participants: Pupils in years (grades) 7-11 (aged 11-16 years). Main outcome measure: Incident smoking in the first year of secondary education, defined as pupils who reported smoking their first cigarette during year 7. Results: Of 6522 pupils (75% of those eligible) who completed the questionnaire, 17% were current smokers and 49% had ever smoked, of whom 23% had started smoking in year 7. Incident smoking in year 7 was more common in girls, in children with parents or siblings who smoke, and in more deprived children, and was independently increased in relation to the proportion of ever smokers in the year 7 tutor group joined by the child (adjusted odds ratio of incident smoking for a child joining a year 7 tutor group in the highest relative to the lowest quartile of ever smoking prevalence 1.45, 95% confidence interval [Cl] 1.11 to 1.89). Exposure to ever smokers in year 7 tutor groups also accounted for most of the increased risk of incident smoking associated with socioeconomic deprivation. Conclusions: The risk of incident smoking in children entering secondary education is independently increased by exposure to other ever smokers in school tutor groups. Incident smoking in adolescents is thus to some extent a communicable disorder, and may be partly preventable by policies that reduce exposure to smoking at school.