Objective: To describe the longitudinal development of smoking behaviour from adolescence (age 13 years) into young adulthood (age 27 years) and to analyse the longitudinal relationship between smoking behaviour and biological risk factors for cardiovascular disease.
Design: Longitudinal.
Setting: Amsterdam school children.
Methods: This study was carried out as part of the Amsterdam Growth and Health Study, an observational longitudinal study. Six measurements were carried out over a period of 15 years in 84 males and 98 females, and included a questionnaire regarding smoking behaviour and blood analysis for cholesterol parameters. Tracking analysis and generalized estimating equations analysis were used.
Results: Both the percentage of smokers and the amount of tobacco smoked by smokers increased until the age of 21 years. After that smoking behaviour remained more or less stable in males and decreased slightly in females (period effect). Between 16 and 21 years half the number of persons tried to stop smoking but failed to succeed. The development of smoking behaviour appeared to be related to low systolic blood pressure and to development of a negative risk profile regarding hypercholesterolaemia (i.e. low high-density lipoprotein (HDL) cholesterol and high ratio of total cholesterol and HDL cholesterol). There was no relationship between smoking and the amount of subcutaneous fat.