Purpose: There is lack of evidence on the effects of health-promoting programmes among adolescents. Health behaviour models and studies seldom compare the underlying factors of unhealthy behaviours between different adolescent age groups. The main objective of this study was to investigate factors including sociodemographic parameters that were associated with vulnerability to health-damaging behaviours and non-adoption of health-enhancing behaviours in different adolescent age groups. Methods: A survey was conducted among 10,590 pupils in the age groups of 13–14, 15–16 and 17–18 years. Structural equation modelling was performed to determine whether health-damaging behaviours (smoking and alcohol consumption) and non-adoption of health-enhancing behaviours (regular meal habits and physical activity) shared an underlying vulnerability. This method was also used to determine whether gender and socio-economic status were associated with an underlying vulnerability to unhealthy behaviours. Results: The findings gave rise to three models, which may reflect the underlying vulnerability to health-damaging behaviours and non-adoption of health-enhancing behaviours at different ages during adolescence. The four behaviours shared what was interpreted as an underlying vulnerability in the 15–16-year-old age group. In the youngest group, all behaviours except for non-participation in physical activity shared an underlying vulnerability. Similarly, alcohol consumption did not form part of the underlying vulnerability in the oldest group. Lower socio-economic status was associated with an underlying vulnerability in all the age groups; female gender was associated with vulnerability in the youngest adolescents and male gender among the oldest adolescents. Conclusions: These results suggest that intervention studies should investigate the benefits of health-promoting programmes designed to prevent health-damaging behaviours and promote health-enhancing behaviours in adolescents of different ages. Future studies should examine other factors that may contribute to the underlying vulnerability in different age groups.