Background: A number of epidemiological studies have shown that smoking is a risk factor for periodontal disease. Little is known about the relationship between smoking duration and alveolar bone loss. The purpose of this research was to describe the prevalence of alveolar bone loss according to smoking status in Norway. A dose-response model for duration of tobacco smoking on alveolar bone loss was then developed and discussed. Methods: The study population consisted of 812 individuals living in Norway aged 45 to 64 years old (248 current smokers, 245 former smokers and 319 non-smokers). Alveolar bone loss was measured on bite-wing radiographs. Simple descriptive statistics were used to describe the central tendency and variation in alveolar bone loss. Regression analyses were performed to study the relationship between smoking duration and alveolar bone loss. Results: Mean alveolar bone loss varied between 1.51 mm and 2.64 mm depending on smoking status and age. Mean alveolar bone loss was lowest in non-smokers and highest in current smokers. Given identical smoking status, the mean alveolar bone loss increased with increasing age except for the 2 oldest age groups of current smokers. Conclusions: Our results generate the hypothesis that the relationship between smoking duration and alveolar bone loss was "S-shaped." Assuming that alveolar bone loss is irreversible after smoking cessation, it could be hypothesized that there is a threshold period for tobacco smoking after which the accumulated effect of smoking becomes clinically observable. After a certain number of years of smoking, the effect on alveolar bone loss seems to level out. To test this hypothesis, the relationship between smoking duration and alveolar bone loss should be studied in a prospective study design.