Reassessing BMI thresholds for obesity: a population-based study in Turkish adults

BACKGROUND: The global prevalence of obesity has been steadily increasing, posing major public health challenges. Current universal BMI cut-off points may not accurately reflect obesity risk across different ethnicities, including the Turkish population. OBJECTIVE: This study aimed to evaluate the relationship between body mass index (BMI) and body fat percentage (BF%) and determine appropriate BMI cut-off points for diagnosing obesity in Turkish adults. METHODS: A cross-sectional study was conducted among Turkish individuals aged 18-65 in four major districts of Kayseri. Anthropometric measurements were obtained, including height, weight, waist circumference, and hip circumference. Body composition was assessed using bioelectrical impedance analysis (BIA). ROC (Receiver Operating Characteristic) curve analysis was performed to identify optimal BMI cut-off points for predicting obesity based on BF%. Pearson correlation analysis evaluated the relationship between BMI and BF%. RESULTS: The participants had a mean age of 39.9 ± 13.3 years, and 51.6% were female. A strong positive correlation between BMI and BF% was found for both women (r = 0.86, p < 0.001) and men (r = 0.68, p < 0.001). The optimal BMI cut-off point for detecting obesity was 28.76 kg/m² for the total sample, 28.80 kg/m² for women, and 28.76 kg/m² for men. Using these new thresholds, the obesity prevalence increased from 27.7% to 33.6%. The AUC values indicated excellent discriminative power for women (AUC = 0.98) and good discriminative power for men (AUC = 0.91). CONCLUSION: The current universal BMI cut-off points (≥30 kg/m²) may underestimate obesity prevalence in the Turkish population. Establishing national and gender-specific BMI thresholds could improve the accuracy of obesity diagnosis and support more effective public health interventions targeting chronic non-communicable diseases.