Sorbonne Paris Nord University, Inserm, INRAE, Cnam, Nutritional Epidemiology Research Team (EREN), Epidemiology and Statistics Research Center, University of Paris (CRESS), 93017, Bobigny, France.
Nutritional Epidemiology Surveillance Team (ESEN), Santé Publique France, The French Public Health Agency, Bobigny, France.
Eur J Nutr. 2023 Feb;62(1):261-274. doi: 10.1007/s00394-022-02978-w. Epub 2022 Aug 12.
To assess the cross-sectional association between dietary indexes (DI) that underlie, respectively, the Nutri-score (NS), the proposed Canadian 'High In' Symbol (CHIL) and the Diabetes Canada Clinical Practice Guidelines (DCCP) with food consumption, nutrient intakes and metabolic markers.
1836 adults (18-74 years) participating in the representative ESTEBAN study, conducted in mainland France in 2014-2016, were included in the analysis. Food consumption was assessed with three repeated 24 h dietary recalls. Anthropometric measurements and biomarkers of metabolic risk (cholesterol-total, LDL (Low Density Lipoprotein), HDL (High Density Lipoprotein)-triglycerides, glucose) were obtained through a clinical examination and fasting blood draw. The DI were assessed for their association with food consumption, dietary intakes and metabolic biomarkers as quintiles and continuous variables using multi-adjusted linear regression. Heathier diets were assigned to lower scores.
Correlations between scores ranged from + 0.62 between CHIL-DI and NS-DI to + 0.75 between NS-DI and DCCP-DI. All DIs discriminated individuals according to the nutritional quality of their diets through food consumption and nutrient intakes (healthier diets were associated with lower intakes of energy, added sugars and saturated fat; and with higher intakes of fiber, vitamins and minerals). NS-DI was associated with blood glucose (adjusted mean in Q1 = 5 vs. Q5 = 5.46 mmol/dl, p = 0.001) and DCCP-DI was associated with BMI (Q1 = 24.8 kg/m vs. Q5 = 25.8 kg/m, p = 0.025), while CHIL showed no significant association with any anthropometric measures or biomarkers.
This study provides elements supporting the validity of the nutrient profiling systems underlying front-of-package nutrition labellings (FOPLs) to characterize the healthiness of diets.
评估 Nutri-score（NS）、加拿大“高”符号（CHIL）和加拿大糖尿病临床实践指南（DCCP）所依据的膳食指数（DI）与食物消费、营养素摄入和代谢标志物之间的横断面关联。
纳入了 2014 年至 2016 年在法国大陆进行的代表性 ESTEBAN 研究中的 1836 名成年人（18-74 岁）。采用三次 24 小时膳食回忆评估食物消费。通过临床检查和空腹采血获得代谢风险的生物标志物（总胆固醇、低密度脂蛋白（LDL）、高密度脂蛋白（HDL）-甘油三酯、血糖）。使用多因素线性回归，将 DI 评估为五分位数和连续变量，以评估其与食物消费、膳食摄入量和代谢生物标志物的关联。分数越低表示饮食越健康。
评分之间的相关性从 CHIL-DI 和 NS-DI 之间的+0.62 到 NS-DI 和 DCCP-DI 之间的+0.75。所有 DI 均通过食物消费和营养素摄入来区分个体的饮食营养质量（更健康的饮食与较低的能量、添加糖和饱和脂肪摄入有关，而与较高的纤维、维生素和矿物质摄入有关）。NS-DI 与血糖有关（Q1=5 时的调整均值与 Q5=5.46 mmol/dl 相比，p=0.001），DCCP-DI 与 BMI 有关（Q1=24.8 kg/m 与 Q5=25.8 kg/m 相比，p=0.025），而 CHIL 与任何人体测量指标或生物标志物均无显著关联。
本研究提供了支持 FOPLs 中营养素成分系统有效性的元素，以表征饮食的健康程度。
