Prevalence of obesity in Italian adolescents: does the use of different growth charts make the difference?

Background Since populations are becoming increasingly multi-ethnic, the use of local or international charts is a matter of debate. This study aimed to evaluate how the choice of cut-off thresholds affected prevalence of underweight (UW), overweight (OW), obesity (OB) in 1200 11–12-year Italian adolescents, and how their somatic growth depended on parental origin. Methods The height, weight and body mass index were expressed as standard deviation score (SDS) using Italian (ISPED-2006) and UK (UK-1990) charts. The classification of UW/OW/OB was computed with the IOTF international cut-offs, and thresholds were identified as centiles corresponding to BMI values of 18.5/25.0/30.0 kg/m 2 at 18-year in ISPED-2006 or UK-1990 references. Results About 30% participants had non-Italian parents, above all from North-Africa and Romania. Referring to the UK-1990 charts, all groups showed negative mean SDS for height, and positive SDS for weight and BMI. Referring to the ISPED-2006 charts, all mean SDS were negative. Percentage of UW individuals was higher in accordance with ISPED-2006 than with UK-1990 charts, whereas percentages of OW/OB were higher with UK-1990 than ISPED-2006 charts. The results obtained using IOFT cut-offs were similar to UK-1990 cut-offs. These results were due to the different shape of age-dependent cut-off centiles. Independently by the parental origin, the percentages of adolescents classified as OW/OB were closer to the expected values using the ISPED-2006 then the UK-1990 cut-offs. The results suggested the use of the Italian references for adolescents with immigrant parents. Conclusion The use of local charts seems more appropriate at least in Italian adolescents in the age range studied.