He Ting, Yan Yifei, Wang Dongdong, Peng Tingting, Jin Liuyin
Shaoxing Seventh People's Hospital (Affiliated Mental Health Center, Medical College of Shaoxing University）, Shaoxing, China.
The First Clinical Medical College, Guangdong Medical University, Zhanjiang, China.
BMC Public Health. 2025 Mar 19;25(1):1068. doi: 10.1186/s12889-025-22245-x.
The Healthy Eating Index (HEI-2015) assesses dietary quality, and lower scores may be associated with an increased risk of frailty. However, few epidemiological studies have examined the relationship between HEI-2015 and the Frailty Index (FI). This study explores the association between HEI-2015 and FI using data from the U.S. National Health and Nutrition Examination Survey (NHANES), analyzing how factors such as gender and age influence this relationship.
TNHANES data (2007-2018) were analyzed using logistic regression models to assess the HEI-2015-frailty association. Frailty diagnosis was based on physical activity, strength, fatigue, weight change, and gait speed, with FI scores ranging from 0 to 1. A threshold of 0.21 classified frailty. The R package "DALEX" was used for feature importance analysis to enhance the prediction of frailty. From this analysis, we selected 10 key factors to further improve the accuracy of frailty prediction.
Of 14,300 participants, 16.2% (2,322) were classified as frail. Frail participants had lower income, higher BMI, lower physical activity, and lower HEI-2015 scores. Higher HEI-2015 scores were associated with reduced frailty risk (adjusted OR: 0.69, 95% CI: 0.56-0.87, P < 0.01). The negative association was stronger in women, higher-income groups, and those with higher education (P < 0.01). Feature importance analysis showed HEI-2015 was the top predictor of frailty.
Higher HEI-2015 scores are linked to lower frailty risk. Promoting healthy eating may prevent frailty, especially in high-risk groups, with education and demographic factors influencing this relationship.
健康饮食指数（HEI-2015）用于评估饮食质量，得分较低可能与身体虚弱风险增加有关。然而，很少有流行病学研究探讨HEI-2015与虚弱指数（FI）之间的关系。本研究利用美国国家健康与营养检查调查（NHANES）的数据，探讨HEI-2015与FI之间的关联，并分析性别和年龄等因素如何影响这种关系。
使用逻辑回归模型分析NHANES数据（2007-2018），以评估HEI-2015与虚弱之间的关联。虚弱诊断基于身体活动、力量、疲劳、体重变化和步速，FI得分范围为0至1。以0.21为阈值划分虚弱。使用R包“DALEX”进行特征重要性分析，以增强对虚弱的预测。通过该分析，我们选择了10个关键因素，以进一步提高虚弱预测的准确性。
在14300名参与者中，16.2%（2322名）被归类为虚弱。虚弱参与者的收入较低、体重指数较高、身体活动较少且HEI-2015得分较低。较高的HEI-2015得分与较低的虚弱风险相关（调整后的比值比：0.69，95%置信区间：0.56-0.87，P<0.01）。这种负相关在女性、高收入群体和受过高等教育的人群中更强（P<0.01）。特征重要性分析表明，HEI-2015是虚弱的首要预测因素。
较高的HEI-2015得分与较低的虚弱风险相关。促进健康饮食可能预防虚弱，尤其是在高危人群中，教育和人口因素会影响这种关系。
