Huang Cunrong, Lopes Andre, Britton Annie
Research Department of Epidemiology and Public Health, University College London, United Kingdom.
Cancer Research UK & Cancer Trials Centre, University College London, United Kingdom.
Diabetes Res Clin Pract. 2025 Jul;225:112268. doi: 10.1016/j.diabres.2025.112268. Epub 2025 May 20.
We compared ability of five adiposity indicators [body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR), waist-by-height (WHT.5R), and a body shape index (ABSI)] to identify current diabetes and their prospective associations with diabetes.
Baseline data were from 7,979 participants of UK Whitehall II study, of whom 7,488 diabetes-free participants were followed-up (median = 16.0 years) for incident diabetes (n = 940). According to five indices' cut-points, participants were separately classified into low-value groups and high-value groups. We cross-sectionally investigated ability of the indicators to identify existing diabetes by receiver operating characteristic curve analysis, and explored prospective associations between elevated indices and diabetes using Cox regression analysis.
Waist-based indicators were superior to BMI in identifying diabetes. High WHtR (≥0.5) demonstrated the highest multivariable-adjusted HR [2.64 (95 % CI 2.29, 3.03)]. Across all indicators, associations between elevated indicators and diabetes were stronger in younger participants. In combined analyses, "low BMI but high WHtR" had higher risk for diabetes [2.20 (95 % CI 1.65, 2.95)] than "high BMI but low WHtR" [1.34 (95 % CI 1.05, 1.70)].
Waist-based indicators are more strongly associated with diabetes than BMI. WHtR, an easy-to-calculate, waist-based index with a sex- and race-independent cut-point, may be useful for diabetes prevention.
我们比较了五个肥胖指标[体重指数（BMI）、腰围（WC）、腰高比（WHtR）、腰乘高（WHT.5R）和体型指数（ABSI）]识别当前糖尿病的能力及其与糖尿病的前瞻性关联。
基线数据来自英国白厅II研究的7979名参与者，其中7488名无糖尿病参与者接受了糖尿病发病随访（中位数=16.0年）（n=940）。根据五个指标的切点，参与者被分别分为低值组和高值组。我们通过受试者工作特征曲线分析横断面研究了这些指标识别现有糖尿病的能力，并使用Cox回归分析探讨了指标升高与糖尿病之间的前瞻性关联。
基于腰围的指标在识别糖尿病方面优于BMI。高WHtR（≥0.5）显示出最高的多变量调整后HR[2.64（95%CI 2.29，3.03）]。在所有指标中，指标升高与糖尿病之间的关联在年轻参与者中更强。在综合分析中，“低BMI但高WHtR”患糖尿病的风险[2.20（95%CI 1.65，2.95）]高于“高BMI但低WHtR”[1.34（95%CI 1.05，1.70）]。
基于腰围的指标与糖尿病的关联比BMI更强。WHtR是一个易于计算的、基于腰围的指数，其切点不依赖性别和种族，可能对糖尿病预防有用。
