基于随机森林算法和Logistic回归分析的老年肌肉衰减症影响因素

陈新宇; 周军良; 李婷婷; 田甜; 汪敏; 贾红

doi:10.16462/j.cnki.zhjbkz.2022.03.020

基于随机森林算法和Logistic回归分析的老年肌肉衰减症影响因素

doi: 10.16462/j.cnki.zhjbkz.2022.03.020

陈新宇¹,
周军良¹,
李婷婷¹,
田甜¹,
汪敏²,
贾红^1, ,

1.
646000 泸州，西南医科大学公共卫生学院
2.
646000 泸州，西南医科大学附属医院营养科

基金项目:

达能营养中心膳食营养研究与宣教基金 DIC2019-10

详细信息

通讯作者:
贾红，E-mail: jhong_lz@163.com

中图分类号: R443+8, R592
计量
- 文章访问数: 1021
- HTML全文浏览量: 539
- PDF下载量: 114
- 被引次数: 0
出版历程
- 收稿日期: 2021-05-12
- 修回日期: 2021-08-13
- 网络出版日期: 2022-03-17
- 刊出日期: 2022-03-10

Influential factors of sarcopenia in older adults based on random forest and Logistic regression

1.
School of Public Health, Southwest Medical University, Luzhou 646000, China
2.
Department of Nutrition, The Affiliated Hospital of Southwest Medical University, Luzhou 646000, China

Funds:

Danone Nutrition Center Dietary Nutrition Research and Education Fund DIC2019-10

More Information

Corresponding author: JIA Hong, E-mail: jhong_lz@163.com

摘要

摘要: 目的分析老年肌肉衰减症的影响因素，为老年肌肉衰减症干预提供参考依据。方法通过对泸州市某社区及医院439例老年人进行问卷调查，应用随机森林算法对影响因素进行重要性排序并降维，将降维后的变量纳入Logistic回归分析模型，分析影响因素的作用方向和相对危险度。结果随机森林算法显示：变量数为7时袋外估算误差率最低，依次是BMI、糖尿病、年龄、高血压、中心性肥胖、饮酒、锻炼时间。Logistic回归分析显示：年龄、BMI、适量饮酒、中心性肥胖、糖尿病是肌肉衰减症的影响因素(均有P＜0.05)。结论 BMI、糖尿病、年龄、高血压、中心性肥胖、饮酒、锻炼时间是对肌肉衰减症影响较大的7个因素。建议加大健康生活方式的宣传力度，加强对高血压、糖尿病的检测和控制。
- 肌肉衰减症 /
- 影响因素 /
- 随机森林 /
- Logistic回归
Abstract: Objective To analyze the influencing factors of sarcopenia in older adults, and providing a reference for sarcopenia intervention. Methods The structured questionnaires were used to investigate 439 elder persons admitted to communities and hospitals in Luzhou City. The random forest method was used to sort the importance and reduce the dimension of the independent variables. The Logistic regression model was used to analyze the action direction and relative risk of independent variables. Results The results of the random forest analysis showed that the estimated error rate was the lowest when the number of variables was seven, followed by the body mass index, diabetes, age, hypertension, abdominal obesity, alcohol drinking and exercise duration. Logistic regression analysis showed that age, BMI, alcohol drinking, abdominal obesity, diabetes were influencing factors of sarcopenia in the elderly (all P < 0.05). Conclusion The top seven influencing factors of sarcopenia were BMI, diabetes, age, hypertension, abdominal obesity, alcohol drinking, and exercise duration. It is recommended to increase the education of healthy lifestyles, and strengthen the detection and control of diabetes and hypertension.
- Sarcopenia /
- Influencing factors /
- Random forest /
- Logistic regression

HTML全文

图 1 变量重要性排序

Figure 1. Importance ranking of variables

下载: 全尺寸图片幻灯片

图 2 逐步随机森林袋外数据分类错误率图

Figure 2. Classification error rate chart of stepwise random forest out-of-bag data

下载: 全尺寸图片幻灯片

图 3 肌肉衰减症影响因素的多因素Logistic回归分析

Figure 3. Multivariate Logistic regression analysis of influencing factors for Sarcopenia

下载: 全尺寸图片幻灯片

表 1 两组研究对象基本情况比较[n (%)]

Table 1. Baseline characteristics of study participants [n (%)]

变量	非肌肉衰减症组	肌肉衰减症组	χ²/Z值	P值
年龄(岁)			-3.702	0.001
60~＜70	159(53.9)	37(34.9)
70~＜80	117(39.7)	53(50.0)
≥80	19(6.4)	16(15.1)
学历			-0.102	0.919
初中及以下	173(59.0)	62(59.6)
高中及以上	120(41.0)	42(40.4)
性别			0.008	0.928
男性	146(49.5)	53(50.0)
女性	149(50.5)	53(50.0)
BMI(kg/m²)			-6.895	＜0.001
＜24	122(41.4)	85(80.2)
24~＜28	135(45.8)	19(17.9)
≥28	38(12.9)	2(1.9)
高血压			0.015	0.903
是	126(42.7)	46(43.4)
否	169(57.3)	60(56.6)
糖尿病			21.845	＜0.001
是	45(15.3)	39(36.8)
否	250(84.7)	67(63.2)
高脂血症			1.293	0.256
是	50(16.9)	13(12.3)
否	245(83.1)	93(87.7)

下载: 导出CSV

参考文献(17)

[1]	Chen LK, Woo J, Assantachai P, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment[J]. J Am Med Dir Assoc, 2020, 21(3): 300-307. DOI: 10.1016/j.jamda.2019.12.012.
[2]	Cruz-Jentoft A J, Bahat G, Bauer J, et al. Sarcopenia: revised European consensus on definition and diagnosis[J]. Age and ageing, 2019, 48(4): 601. DOI: 10.1093/ageing/afz046.
[3]	孙建琴, 叶梦瑶. 老年肌肉衰减症的筛查、评估与干预[J]. 外科理论与实践, 2020, 25(2): 114-119. DOI: 10.16139/j.1007-9610.2020.02.006. Sun JQ, Ye MY. Screening, evaluation and intervention for sarcopenia in the elderly[J]. J Surg Concepts Pract, 2020, 25(2): 114-119. DOI: 10.16139/j.1007-9610.2020.02.006.
[4]	王慧, 海珊, 刘颖, 等. 成都市社区老人肌少症患病率及相关因素研究[J]. 四川大学学报(医学版), 2019, 50(2): 224-228. DOI: 10.13464/j.scuxbyxb.2019.02.016. Wang H, Hai S, Liu Y, et al. Prevalence of Sarcopenia and Associated Factors in Community-dwelling Elderly Populations in Chengdu China[J]. J Sichuan Univ (Med Sci Edi), 2019, 50(2): 224-228. DOI: 10.13464/j.scuxbyxb.2019.02.016.
[5]	陈姝, 俞静, 狄文娟, 等. 南京市社区老年肌少症流行病学调查及其影响因素分析[J]. 实用老年医学, 2020, 34(8): 793-796. DOI: 10.3969/j.issn.1003-9198.2020.08.010. Chen S, Yu J, Di WJ, et al. Epidemiological survey and influencing factors of sarcopenia in the elderly in a community of Nanjing[J]. Pract Geriatr, 2020, 34(8): 793-796. DOI: 10.3969/j.issn.1003-9198.2020.08.010.
[6]	Galvan-Tejada, CE, JI, et al. An Analysis of Audio Features to Develop a Human Activity Recognition Model Using Genetic Algorithms, Random Forests, and Neural Networks[J]. MOB INF SYST, 2016, 2016(-): DOI: 10.1155/2016/1784101.
[7]	Statistics LB, Breiman L J M L. Random Forests[J]. 2001, 5-32. DOI: 10.1023/A:1010950718922.
[8]	田甜, 李婷婷, 聂志超, 等. 基于随机森林算法的高尿酸血症危险因素分析[J]. 中国卫生统计, 2020, 37(2): 162-165, 169. Tian T, Li TT, Nie ZC, et al. Risk Factors for Hyperuricemia Based on Random Forest Algorithm[J]. Chinese Journal of Health Statistics, 2020, 37(2): 162-165, 169.
[9]	杨月欣, 张环美. 《中国居民膳食指南(2016)》简介[J]. 营养学报, 2016, 38(3): 209-217. DOI: 10.13325/j.cnki.acta.nutr.sin.2016.03.002. Yang YX, Zhang HM. Introduction to "Dietary Guidelines for Chinese Residents (2016)[J]. Acta Nutrimenta Sinica, 2016, 38(3): 209-217. DOI: 10.13325/j.cnki.acta.nutr.sin.2016.03.002.
[10]	Kim M K, Lee W Y, Kang J H, et al. 2014 clinical practice guidelines for overweight and obesity in Korea[J]. Endocrinology and metabolism (Seoul, Korea), 2014, 29(4): 405-409. DOI: 10.3803/EnM.2014.29.4.405.
[11]	陆晔, 刘晓侠, 朱珍妮, 等. 上海市中老年人群优质蛋白质摄入与肌肉衰减症风险的关系研究[J]. 营养学报, 2020, 42(5): 429-434. DOI: 10.13325/j.cnki.acta.nutr.sin.2020.05.003. Lu Y, Liu XY, Zhu ZN, et al. Association Between Daily Intake of High Quality Proteins and Sarcopenia in Aging Adults In Shanghai[J]. Acta Nutrimenta Sinica, 2020, 42(5): 429-434. DOI: 10.13325/j.cnki.acta.nutr.sin.2020.05.003.
[12]	Beaudart C, Rizzoli R, BruyèRe O, et al. Sarcopenia: burden and challenges for public health[J]. Archives of public health, 2014, 72(1): 45. DOI: 10.1186/2049-3258-72-45.
[13]	Keller K, Engelhardt M. Strength and muscle mass loss with aging process. Age and strength loss[J]. Muscles, ligaments and tendons journal, 2013, 3(4): 346-350.
[14]	吴佳佳, 王炜, 祝捷, 等. 2型糖尿病患者肌少症的相关因素[J]. 中华骨质疏松和骨矿盐疾病杂志, 2016, 9(2): 129-135. DOI: 10.3969/j.issn.1674-2591.2016.02.005. Wu JJ, Wang W, Zhu J, et al. Relative factors to sarcopenia in T2DM patients[J]. Chin J Osteoporpsis & Bone Minerres, 2016, 9(2): 129-135. DOI: 10.3969/j.issn.1674-2591.2016.02.005.
[15]	于康. 肌肉衰减症临床营养管理的现状与进展[J]. 中华健康管理学杂志, 2020, 14(6): 505-508. DOI: 10.3760/cma.j.cn115624-20200817-00615. Yu K. The current status and progress of clinical nutritional management for sarcopenia[J]. Chin J Health Manage, 2020, 14(6): 505-508. DOI: 10.3760/cma.j.cn115624-20200817-00615.
[16]	Kuk JL, Ardern CI. Influence of age on the association between various measures of obesity and all-cause mortality[J]. Journal of the American Geriatrics Society, 2009, 57(11): 2077-2084. DOI: 10.1111/j.1532-5415.2009.02486.x.
[17]	孟欣宇, 赵文静, 张金华, 等. 社区老年人身体活动与罹患肌少症的关联[J]. 中华疾病控制杂志, 2020, 24(8): 914-918. DOI: 10.16462/j.cnki.zhjbkz.2020.08.010. Meng XY, Zhao WJ, Zhang JH, et al. Association between physical activity and prevalence of sarcopenia in community－dwelling elderly[J]. Chin J Dis Control Prev, 2020, 24(8): 914-918. DOI: 10.16462/j.cnki.zhjbkz.2020.08.010.