Comparison of Waist Circumference and Waist-to-Hip Ratio in Assessing Abdominal Obesity and Cardiometabolic Risk Among Women Aged 15-49 Years in India: Evidence From a National Cross-Sectional Survey.

Introduction Abdominal obesity (AO) has become a major public health challenge, particularly in South Asian populations where risk occurs at lower body mass index (BMI) levels. Waist circumference (WC) and waist-to-hip ratio (WHR) are widely used measures of AO, yet their comparative performance in identifying individuals at risk remains unclear. This study aims to compare WC and WHR in estimating AO, examine their agreement and discordance, and assess their association with diabetes and hypertension among women in India. Methods We utilised the data from the fifth round of the National Family Health Survey, 2019-21, a nationally representative cross-sectional survey conducted across all 36 states and union territories of India. The survey collected information from 724,115 women aged 15-49 years. Women who were pregnant at the time of the survey, observations with extreme anthropometric values, and missing values were excluded from the study. The final analytical sample comprised 664,646 women. AO was defined using WC >80 cm and WHR ≥0.85. Agreement between measures was assessed using percentage agreement and Cohen's kappa statistic. Discordance analysis was conducted to compare socioeconomic and behavioural profiles across classification groups. Multivariate logistic regression models were used to examine associations of WC- and WHR-defined AO with diabetes and hypertension after adjusting for socioeconomic and behavioural covariates. Results The prevalence of AO among women was 39.9% using WC and 56.6% using WHR. Agreement between the two measures was fair (66.8%; kappa=0.36), with lower agreement among younger women. Nearly one-fourth of women (24.3%) were classified as abdominally obese by WHR alone compared to 7.6% by WC alone, indicating high discordance between the two measures. Women identified through WC had a higher prevalence of overweight/obesity, urban residence, higher wealth status, and parity, whereas the WHR-only group was more similar to the normal population. WC-defined AO also showed stronger alignment with BMI, while WHR classified a substantial proportion of underweight and normal BMI women as abdominally obese. WC showed a stronger association with cardiometabolic outcomes, with higher odds of diabetes (adjusted odds ratio (AOR)=1.73) and hypertension (AOR=1.51) compared to WHR (AOR=1.33 for both outcomes). Conclusions WC and WHR identify different population groups and are not interchangeable measures of AO. While WHR yields higher prevalence estimates, WC demonstrates stronger alignment with BMI and cardiometabolic risk. Given its simplicity, interpretability, and stronger predictive value, WC may be a more appropriate tool for population-level screening and risk identification among women in India.