Prevalence of overweight and obesity in Sardinian adolescents

To estimate the prevalence of overweight and obesity among adolescents in Sardinia and to examine the association with several biological and geographic factors. A cross-sectional study was performed in 3,946 unselected adolescents (2,011 boys, 1,935 girls; aged 11-15 years) attending the public secondary schools in 33 Sardinian municipalities: 28 semi-rural, 5 urban, sub-grouped according to their geographic location (mountain, hillside and plain). Oversized children were measured and their BMI defined as being above normal values according to parameters provided by the International Obesity Task Force (IOFT) by Cole et al. (BMI for age > or = 95th percentile). Relative risk for overweight and obesity was calculated using Poisson regression analysis: risks associated to each covariate were reciprocally adjusted. The 95% confidence interval (CI) of the estimated risk was calculated using Wald's formula (RR, RR = log(n) beta +/- 1.96 se(beta)). The overall prevalence rate found for overweight and obesity was 14.9% (95% C.I.: 13.7-16.1%) and 3.7% (95% C.I. 3.1-4.3%), respectively. Overweight rate showed no association with gender, whereas belonging to the female sex constituted a significant protection against obesity. Increasing age in the range 12-14 years was protective against both overweight and obesity in the whole sample. A similar finding however was not observed for obesity in girls or overweight in boys, when considered separately. Boys, but not girls, living in urban areas displayed a modest though significant 20% increase in overweight and a 40% decrease in obesity risk. Living in a mountainous area conveyed a 30% decrease in risk of overweight and a 50% decrease in risk of obesity, when compared to living on the plains and hillside combined. However, the small sample size of study subjects living in mountainous areas generated extremely wide 95% confidence intervals, thereby preventing the drawing of any significant conclusions. In comparison with other surveys performed by the IOFT, Sardinian adolescents show a low prevalence rate for oversize, emphasizing a marked discrepancy with the general north-south rising trend of oversize observed throughout Europe. Geographic location, aesthetic or other age related factors seem to exert a different gender-specific influence on overweight and obesity. The present report is cross sectional and the consequences of overweight and obesity on individuals over time are not traceable. However, the outcome of