This study was performed to assess the influence of smoking on periodontal disease severity. Data concerning periodontal status and smoking habits were collected from 889 periodontal patients: 340 male and 549 female, 21 to 76 years of age, 47.4% being non smokers and 52.6% smokers. Periodontal parameters, recorded by the same examiner (PMC), were: gingival recession (GR), Pocket depth (PD), Probing attachment level (PAL), and mobility (M). The influence of age, sex and tobacco consumption on these periodontal parameters was statistically evaluated using an analysis of variance (ANOVA) with covariates. A non-linear effect model was also fitted by taking the natural logarithms of the response variables (GR, PD, PAL) closer to biomedical phenomena. Mobility was analyzed by a chi2-test. The effect of smoking on periodontitis showed no association with age or with sex. Smoking, age and sex were shown to be statistically significant for periodontitis, by performing both univariate (t-test for equal means) and multivariate tests. p-values for smoking and periodontitis were: GR (p=0.000), PD (p=0.000), PAL (p=0.000) and M (P=0.015). Smoking one cigarette per day, up to 10, and up to 20, increased PAL by 0.5%, 5% and 10%, respectively. The impact of tobacco is comparable to the impact resulting from the factor of age in this sample, increasing PAL by 0.7% for each year of life. Comparison between smokers of less than 10 cigarettes per day (PAL mean 3.72 mm +/-0.86) and non-smokers (PAL mean 3.84 +/- 0.89) showed no differences in PAL (p=0.216), while comparison for smokers from 11 to 20 cigarettes (PAL mean 4.36 +/- 1.23) and for more than 20 cigarettes (PAL mean 4.50 +/- 1.04) demonstrated significant differences (p=0.000). These findings suggest that: (1) tobacco increases periodontal disease severity; (2) this effect is clinically evident above consumption of a certain quantity of tobacco.