Prevalence of obesity among youth with intellectual disabilities: Comparison of cut-points

OBJECTIVE: The purpose of this study was to compare the prevalence of obesity across two international growth references [World Health Organization (WHO); International Obesity Task Force (IOTF)] among a large sample of youth with intellectual disabilities (ID). Discrepancies between published obesity prevalence rates among youth with ID are likely due to the use of different growth references. However, no studies have directly compared the performance of multiple growth references for youth with ID. METHODS: The records of 3,736 youth (59% male; 8-18 years) were available from the Special Olympics International Healthy Athletes Health Promotion database (2007-2011) after data cleaning. BMI (kg/m²) was calculated from anthropometric measurements and prevalence of obesity was determined using WHO and IOTF criteria. Differences between growth references were examined through descriptive statistics (proportions, 95% CI), proportions tests and kappa statistics. Analyses were conducted for the entire sample, by gender, and by age with bootstrapping procedures. RESULTS: Estimates of obesity prevalence from WHO (30.91% ± 1.49%) criteria were significantly higher than IOTF (25.96% ± 1.41%; p<.001) cut-points. Differences are substantially larger when stratified by age and gender. The largest difference observed was 15%. Despite large differences between prevalence estimates, agreement of obesity classification was high. Kappa statistics ranged from .69 to .99, with most above .80 indicating good agreement. CONCLUSIONS: Prevalence estimates of obesity for youth with ID varied significantly between growth references. Differences are largest when prevalence rates are calculated by specific age groups and genders and should be considered when interpreting results across studies.