《中国康复理论与实践》

《中国康复理论与实践》 ›› 2026, Vol. 32 ›› Issue (1): 1-12.doi: 10.3969/j.issn.1006-9771.2026.01.001

• 专题 平衡功能 • 上一篇    下一篇

体力活动和平衡能力与残疾的关联性研究——基于NHANES的横断面研究

韩雪娇, 史冀鹏(), 卫金鹏, 朱天瑞, 徐红旗, 何慕寒   

  1. 东北师范大学运动能力测评与提升研究中心，吉林长春市 130024
  • 收稿日期:2025-08-07 修回日期:2025-11-05 出版日期:2026-01-25 发布日期:2026-01-28
  • 通讯作者: 史冀鹏(1975-)，女，汉族，辽宁抚顺市人，博士，副教授，主要研究方向：运动生理学。E-mail： shijp2006@163.com
  • 作者简介:韩雪娇(2001-)，女，汉族，河南焦作市人，硕士研究生，主要研究方向：运动与健康促进。
  • 基金资助:
    国家社会科学基金项目(22BTY075)

Association of physical activity and balance ability with disability: a cross-sectional study based on NHANES

HAN Xuejiao, SHI Jipeng(), WEI Jinpeng, ZHU Tianrui, XU Hongqi, HE Muhan   

  1. Research Center of Exercise Capacity Assessment and Promotion, Northeast Normal University, Changchun, Jilin 130024, China
  • Received:2025-08-07 Revised:2025-11-05 Published:2026-01-25 Online:2026-01-28
  • Contact: SHI Jipeng, E-mail: shijp2006@163.com
  • Supported by:
    National Social Science Fund of China(22BTY075)

RichHTML

PDF (PC)

摘要：

目的 探究体力活动和平衡能力与残疾之间的关联性，并分析平衡能力在体力活动与残疾关联中的中介作用。

方法 基于2021年至2023年美国国家健康和营养调查(NHANES)数据，构建多变量Logistic回归模型评估体力活动和平衡能力对残疾的独立关系，分四步调整变量。交互效应分析构建含交互项的Logistic回归模型。采用Baron & Kenny三步法检验平衡能力中介作用，计算效应比例。采用Delta方法等评估中介效应显著性，Bootstrap法验证稳健性，设随机种子。多项敏感性分析验证结果。

结果 共纳入3 902例受试者，残疾检出率为17.7%。多因素回归显示，在调整所有协变量后，低中强度体力活动(LIPA) (OR = 0.489, 95%CI 0.380~0.629, P < 0.001)和高强度体力活动(HIPA) (OR = 0.493, 95%CI 0.371~0.656, P < 0.001)均与残疾风险显著降低相关，而平衡能力异常者残疾风险显著升高(OR = 1.579, 95%CI 1.266~1.970, P < 0.001)。交互效应分析表明，体力活动与平衡能力的交互效应均不显著(P > 0.05)，但LIPA主效应保持稳健(β = -0.597, SE = 0.221, OR = 0.550, P = 0.007)，平衡能力异常仍与残疾风险升高相关(β = 0.577, SE = 0.231, OR = 1.780, P = 0.012)。中介效应分析显示，平衡能力在LIPA与残疾之间发挥稳健的中介作用，中介效应比例为21.1%，间接效应路径a×b显著(P < 0.001)，Bootstrap法验证的95%置信区间均未包含0。

结论 体力活动和平衡能力与残疾之间存在显著关联，平衡能力在二者之间可能具有潜在的中介作用。平衡功能可能在残疾风险评估中及干预具有重要价值。

关键词: 体力活动, 平衡能力, 残疾, 美国健康和营养调查, 中介效应

Abstract:

Objective To investigate the association of physical activity and balance ability with disability, and to further examine the mediating role of balance ability in the relationship between physical activity and disability.

Methods Based on data from 2021 to 2023 National Health and Nutrition Examination Survey (NHANES) cycle, a multivariable Logistic regression model was constructed to evaluate the independent association of physical activity and balance ability with disability, with covariates adjusted in four steps. An interaction model was further developed to assess the multiplicative interaction between physical activity and balance ability. Mediation analysis was performed using the Baron and Kenny three-step approach to examine the mediating role of balance ability, and the proportion of the mediation effect was calculated. The significance of the mediation effect was assessed using the Delta method, and robustness was verified through the Bootstrap method with a fixed random seed. Multiple sensitivity analyses were conducted to ensure the stability of the results.

Results A total of 3 902 participants were included, with a prevalence of disability of 17.7%. Multivariable regression analysis showed that, after adjusting for all covariates, both light-intensity physical activity (LIPA) (OR = 0.489, 95%CI 0.380 to 0.629, P < 0.001) and high-intensity physical activity (HIPA) (OR = 0.493, 95%CI 0.371 to 0.656, P < 0.001) were significantly associated with a reduced risk of disability, whereas impaired balance ability was significantly associated with an increased risk (OR = 1.579, 95%CI 1.266 to 1.970, P < 0.001). The interaction effect analysis showed that the interaction between physical activity and balance ability were not significant (P > 0.05), however, the main effect of LIPA remained robust (β = -0.597, SE = 0.221, OR = 0.550, P = 0.007), while impaired balance ability was significantly associated with an increased risk of disability (β = 0.577, SE = 0.231, OR = 1.780, P = 0.012). The mediation analysis further indicated that balance ability played a robust mediating role in the association between LIPA and disability, with a mediation proportion of 21.1%. The indirect effect (a × b) was statistically significant (P < 0.001), and the 95% confidence intervals derived from the bootstrap method did not include zero.

Conclusion Physical activity and balance ability are significantly associated with disability, and balance ability may mediate the relationship between physical activity and disability, highlighting its potential value in disability risk assessment and intervention strategies.

Key words: physical activity, balance ability, disability, National Health and Nutrition Examination Survey, mediation effect

中图分类号: