办公室职员肌肉质量减少预测模型的开发与验证  

Development and validation of a prediction model for muscle mass loss in office workers

作　　者：陈瑛翼 游倩 王意 张帆[2] 李凤[2] 季舒铭 徐浩源 饶志勇[1] CHEN Yingyi;YOU Qian;WANG Yi;ZHANG Fan;LI Feng;JI Shuming;XU Haoyuan;RAO Zhiyong(Department of Clinical Nutrition,West China Hospital,Sichuan University;Health Management Center,West China Hospital,Sichuan University;Clinical Research Management Department,West China Hospital,Sichuan University;West China School of Medicine,Sichuan University,Chengdu 610041,Sichuan,China)

机构地区：[1]四川大学华西医院临床营养科 [2]四川大学华西医院健康管理中心 [3]四川大学华西医院临床研究管理部 [4]四川大学华西临床医学院,四川成都610041

出　　处：《山东大学学报(医学版)》2025年第4期26-35,共10页Journal of Shandong University:Health Sciences

基　　金：四川省干部保健科研项目(川干研2022-110)。

摘　　要：目的调查分析办公室职员肌肉质量减少发生率及影响因素,构建并验证风险预测模型,为制定干预方案提供参考。方法选取2023年3月至2023年12月在四川大学华西医院健康管理中心体检的事业单位、机关单位办公室职员和退休者286人,采用食物频率调查问卷、运动情况调查问卷、体格检查、人体成分分析、实验室检查等进行分析。使用R语言(R Studio,版本4.4.1)分析肌肉质量影响因素,建立风险预测模型并进行验证。结果肌肉质量减少发生率26.22%;高血糖、低体质量指数(body mass index,BMI)为肌肉质量减少的危险因素,高血尿素/血肌酐(blood urea nitrogen to serum creatinine,BUN/Cr)、每天实际摄入能量/每日推荐摄入能量比、高新鲜水果、高握力评分、超重是肌肉质量减少的保护因素,并构建模型;验证该模型的ROC曲线下面积为0.83,Youden指数为0.52,最佳风险阈值取14.40%。Hosmer-Lemeshow检验和校准曲线评估模型拟合度较高(χ^(2)=11.98,P=0.152),预测模型的阈值概率值在0.07~0.93。结论本研究预测模型具有较好的预测效果和拟合程度,利于医技护人员评估发生肌肉质量减少的风险,有助于为办公室职员肌肉质量减少高危人群提供参考尽早预防。Objective To investigate and analyze the prevalence of muscle mass loss in office workers and the contributing factors,and to construct and validate a risk prediction model to inform the development of an intervention program.Methods Two hundred and eighty-six office workers from institutions and agencies who underwent medical check-ups at the Health Management Center of West China Hospital of Sichuan University from March 2023 to December 2023 were selected and surveyed using food frequency questionnaires,exercise questionnaires,physical examination,body composition analysis,and laboratory tests.Factors influencing muscle mass were analyzed using R software(RStudio,version number 4.4.1),and a risk prediction model was established and validated.Results The prevalence of muscle mass loss was 26.22%;high glucose and low BMI were risk factors for muscle mass loss,high BUN/Cr,high actual daily energy intake/recommended daily energy intake ratio,high fresh fruits intake,high grip strength score and overweight were protective factors for muscle mass loss,and a model was constructed;the area under the ROC curve of the model was verified to be 0.83,and the Youden index was 0.52.The optimal risk threshold was set as 14.40%.The Hosmer-Lemeshow test and calibration curve assessed that the fit of the model was high(χ^(2)=11.98,P=0.152),and predicted that the threshold probability value of the model was in the range of 0.07-0.93.Conclusion The prediction model in this study has a good predictive effect and good fit,which will facilitate medical and technical nurses to assess the risk of developing muscle mass loss and helps to provide a reference for early prevention for office workers at high risk of muscle mass loss.

关 键 词：肌肉质量减少 发生率 影响因素 预测模型 肌少症 

分 类 号：R685[医药卫生—骨科学] R459.3[医药卫生—外科学]