LAUSR

Abstract The present study set out to examine a number of psychological and socio-demographic factors that influence adult tobacco use in a longitudinal birth cohort. The study was based on a nationally representative sample of 5223 people born in Great Britain in 1970 with nine follow-ups. Data used in this study were collected when cohort members were born, and at age 10, 16, 30, 34 and 38 years. Current tobacco use status (yes/no) and lifelong tobacco use status (ever/never) at age 38 years were the two outcome measures respectively. Logistic regression analyses showed that among the 5223 participants with complete data, childhood self-esteem, teenage locus of control, adult psychological distress, educational achievement and occupational prestige were significant, and independent, predictors of both current and lifelong tobacco usage. Locus of Control measured at 16 years predicted smoking over 22 years later and suggests early intervention in changing fatalistic beliefs may have long term consequences. Psychological distress (malaise), shown to be very stable over time, was in the odds ratio, the most important factor. Implications and limitations are discussed.