Wen Jun, Shi Xiaowen, Liu Yan, Zhuang Rongjuan, Guo Shuliang, Chi Jing
Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Chongqing Medical University, Chongqing Medical University, Chongqing, China.
Front Nutr. 2025 Jul 7;12:1555888. doi: 10.3389/fnut.2025.1555888. eCollection 2025.
Epidemiological investigations on the association of handgrip status and asthma risk still remain understudied. This research aims to investigate the associations of handgrip strength (HGS), relative handgrip strength (RHGS), low HGS, and asthma risk, as well as the mediating role of nutritional status, using data from the Survey of Health, Ageing and Retirement in Europe (SHARE).
This investigation included 27,185 participants for a cross-sectional study and 18,047 participants for a prospective cohort study from SHARE. Four machine learning models, the Shapley Additive Explanations (SHAP) model, restricted cubic spline (RCS), cumulative occurrence curve, logistic regression, and Cox regression were used to comprehensively evaluate the performance of handgrip status in predicting asthma risk. Finally, the mediation effect model was employed to evaluate the role of nutritional status in the relationship between grip strength and asthma risk.
The cross-sectional investigation suggested that both HGS (OR: 0.98, 95% CI: 0.98-0.99) and RHGS (OR: 0.61, 95% CI: 0.51-0.73) were negatively linked to the risk of asthma, and low HGS was a risk factor for asthma (OR: 1.52, 95% CI: 1.24-1.87). And the prospective cohort investigation with a median follow-up time of 30 months further confirmed that both HGS (HR: 0.98, 95% CI: 0.97-1.00) and RHGS (HR: 0.52, 95% CI: 0.37-0.73) were negatively linked to the risk of asthma. Among the four machine learning models used to evaluate handgrip status and the risk of asthma, eXtreme Gradient Boosting (XGBoost) showed better predictive performance. The SHAP model based on XGBoost suggested that the top five crucial indicators for predicting asthma risk were RHGS, HGS, country, age, and chronic lung disease. Finally, the mediation effect model suggested that malnutrition partially mediated the relationship between low HGS and increased risk of asthma, with a mediation proportion of 2.71%.
This investigation suggested that lower HGS and RHGS were linked to a higher risk of asthma, and handgrip status could be used as an independent marker of asthma risk in European populations. And malnutrition partially mediated the relationship between low HGS and asthma risk. Improving muscle strength could be a potential preventive strategy against asthma, with implications for public health and clinical practice.
关于握力状况与哮喘风险之间关联的流行病学调查仍研究不足。本研究旨在利用欧洲健康、老龄化与退休调查(SHARE)的数据,调查握力(HGS)、相对握力(RHGS)、低握力与哮喘风险之间的关联,以及营养状况的中介作用。
本调查纳入了SHARE中的27185名参与者进行横断面研究,以及18047名参与者进行前瞻性队列研究。使用四种机器学习模型,即夏普利值加法解释(SHAP)模型、受限立方样条(RCS)、累积发生率曲线、逻辑回归和Cox回归,全面评估握力状况对哮喘风险的预测性能。最后,采用中介效应模型评估营养状况在握力与哮喘风险关系中的作用。
横断面调查表明,HGS(比值比:0.98,95%置信区间:0.98 - 0.99)和RHGS(比值比:0.61,95%置信区间:0.51 - 0.73)均与哮喘风险呈负相关,低HGS是哮喘的一个风险因素(比值比:1.52,95%置信区间:1.24 - 1.87)。中位随访时间为30个月的前瞻性队列研究进一步证实,HGS(风险比:0.98,95%置信区间:0.97 - 1.00)和RHGS(风险比:0.52,95%置信区间:0.37 - 0.73)均与哮喘风险呈负相关。在用于评估握力状况与哮喘风险的四种机器学习模型中,极端梯度提升(XGBoost)显示出更好的预测性能。基于XGBoost的SHAP模型表明,预测哮喘风险的前五个关键指标是RHGS、HGS、国家、年龄和慢性肺病。最后,中介效应模型表明,营养不良部分介导了低HGS与哮喘风险增加之间的关系,中介比例为2.71%。
本调查表明,较低的HGS和RHGS与较高的哮喘风险相关,握力状况可作为欧洲人群哮喘风险的独立标志物。并且营养不良部分介导了低HGS与哮喘风险之间的关系。改善肌肉力量可能是预防哮喘的一种潜在策略,对公共卫生和临床实践具有重要意义。