基于分类树模型对高尿酸血症的影响因素分析

王雪莹; 李楠; 郭佳桐; 冷松

doi:10.16462/j.cnki.zhjbkz.2019.11.013

基于分类树模型对高尿酸血症的影响因素分析

doi: 10.16462/j.cnki.zhjbkz.2019.11.013

116023 大连, 大连医科大学附属第二医院健康管理中心

基金项目:

辽宁省自然科学基金项目 20170540239

详细信息

通讯作者:
冷松, E-mail: dllengsong@126.com

中图分类号: R181.2
计量
- 文章访问数: 229
- HTML全文浏览量: 197
- PDF下载量: 21
- 被引次数: 0
出版历程
- 收稿日期: 2019-04-29
- 修回日期: 2019-08-02
- 刊出日期: 2019-11-10

Analysis on the associated factors of hyperuricemia: based on classification tree model

Health Management Center, Second Affiliated Hospital of Dalian Medical University, Dalian 116023, China

Funds:

Liaoning Natural Science Foundation Project 20170540239

More Information

Corresponding author: LENG Song, E-mail: dllengsong@126.com

摘要

摘要: 目的了解影响高尿酸血症（hyperuricemia，HUA）的生理、社会心理及膳食等生活习惯因素，为HUA的干预提供参考依据。方法对2016-2017年在大连医科大学附属第二医院健康管理中心体检的成年人进行问卷调查和体格测量，比较基本资料、行为习惯、心理因素等对HUA患病风险的影响。采取分类树模型分析各研究因素与HUA的关系。结果研究共纳入研究对象4 118人，平均年龄为（52.8±7.9）岁。统计发现，30岁以下及40~50岁、吸烟、饮酒、静坐为主工作、对工作不满意、精神压力极大的人群HUA检出率更高。因子分析共提取出3种膳食模式：传统模式，油炸腌熏甜点模式，优质蛋白模式，累计方差贡献率53.886%。多因素模型显示，优质蛋白模式、体育锻炼和空腹血糖（fasting plasma glucose，FPG）与HUA呈负相关，男性、吸烟、饮酒、血脂异常及体重指数（body mass index，BMI）为HUA的危险因素。性别分析显示，男、女优质蛋白模式均与HUA呈负相关，油炸腌熏甜点模式为女性HUA的独立危险因素。分类树模型显示男性、血脂异常、高BMI等级、油炸腌熏甜点模式是HUA的危险因素。结论不同性别人群HUA的影响因素不同，应采取个性化的健康管理。超重或肥胖男性应控制吸烟、饮酒等行为，女性更应注意油炸腌熏甜点食物的摄入，提倡优质蛋白饮食模式及适当体育锻炼。
- 血尿酸 /
- 高尿酸血症 /
- 危险因素
Abstract: Objective To explore the associations between physiological factors, psychosocial factors, dietary habits, lifestyles and hyperuricemia (HUA) and provide the evidence for hyperuricemia intervention. Methods From 2016 to 2017, adults during their checkup in the health management center of the Second Affiliated Hospital of Dalian Medical University were interviewed with self-designed questionnaire and received the health examinations. To determine the influence of socio-economic data, life style habits and psychological factors on the risk of HUA, the classification tree model was adopted. Results A total of 4 118 subjects were enrolled in the study, with an average age of (52.8±7.9) years. The results showed that the prevalence of HUA was higher among people under 30 years old and 40-50 years old who are smoking, drinking and doing the sedentary work, who unsatisfied with their work and under a great mental stress. Factor analysis model identified three dietary patterns, traditional model, fried and smoked food and dessert model and high quality protein model, the cumulative variance contribution rate was 53.886%. In multivariate model, it was found that high-quality protein pattern, physical exercise and fasting plasma glucose (FPG) were negatively associated with HUA. Male, smoking, drinking, having dyslipidemia, higher body mass index (BMI) were risk factors for HUA. Subgroup analysis showed that in both male and female, the high-quality protein patterns were negatively associated with HUA. The pattern of fried and smoked food and dessert patterns was the independent risk factors for HUA in female. Classification tree model showed that male, dyslipidemia, higher BMI grade, and the fried and smoked food and dessert pattern were the risk factors for HUA. Conclusions The influencing factors of HUA in different gender were different, which indicate the individualized health management should be adopted. Male who is overweight or obese should quit smoking and drinking. Women should pay more attention to avoid of the excess intake of fried and smoked food and dessert. Meanwhile, a high-quality protein diet and more physical exercise should be encouraged.
- Serum uric acid /
- Hyperuricemia /
- Risk factors

HTML全文

图 1 HUA危险因素分类树模型

Figure 1. HUA risk factor classification tree model

下载: 全尺寸图片幻灯片

表 1 高尿酸血症检出情况[n(%)]

Table 1. Detection of hyperuricemia [n(%)]

特征	人数	HUA(%)	95% CI值	χ²值	P值
性别				245.524	< 0.001
男	2 867	754(26.3)	(24.7，27.9)
女	1 251	64(5.12)	(3.89, 6.34)
年龄(岁)				11.251	0.024
< 30	21	7(33.34)	(11.35，55.32)
30~	190	30(15.79)	(10.56，21.02)
40~	1 159	242(20.88)	(18.54，23.22)
50~	1 994	414(20.76)	(18.98，22.54)
60~90	754	125(16.57)	(13.92，19.24)
吸烟				73.739	< 0.001
不吸烟	2 433	378(15.54)	(14.10，16.98)
已戒烟	411	121(29.44)	(25.02，33.87)
吸烟	1 274	319(25.04)	(22.66，27.42)
饮酒				120.080	< 0.001
不饮酒	1 360	146(10.74)	(9.09，12.38)
已戒酒	455	82(18.02)	(14.48，21.57)
饮酒	2 303	590(25.62)	(23.83，27.40)
体育锻炼				10.323	0.001
是	2 270	410(18.06)	(16.48，19.65)
否	1 848	408(22.08)	(20.19，23.97)
工作方式				14.996	0.001
静坐为主	2 349	512(21.80)	(20.13，23.47)
轻度活动	1 739	304(17.48)	(15.69，19.27)
体力劳动	30	2(6.67)	(-2.80，16.14)
生活满意度				1.894	0.755
很不满意	19	5(26.32)	(4.51，48.12)
不满意	48	9(18.75)	(7.30，30.20)
一般	423	93(21.98)	(18.02，25.95)
满意	2 738	537(19.61)	(18.12，21.10)
很满意	890	174(19.55)	(16.94，22.16)
工作满意度				18.023	0.001
很不满意	14	3(21.43)	(-3.16，46.01)
不满意	75	17(22.67)	(12.97，32.36)
一般	965	147(15.23)	(12.96，17.50)
满意	2 483	536(21.58)	(19.97，23.21)
很满意	581	115(19.79)	(16.54，23.04)
精神压力				23.095	< 0.001
没有压力	765	110(14.38)	(11.89，16.87)
压力较少	684	135(19.74)	(16.75，22.73)
一般	1 078	214(19.85)	(17.47，22.24)
压力较大	1 392	308(22.13)	(19.94，24.31)
压力极大	199	51(25.63)	(19.51，31.75)
沮丧情绪				4.211	0.122
没有	964	193(20.02)	(17.49，22.55)
偶尔	2 765	563(20.36)	(18.86，21.86)
经常	389	62(15.94)	(12.28，19.59)
焦虑情绪				1.185	0.553
没有	933	185(19.83)	(17.27，22.39)
偶尔	2 724	550(20.19)	(18.68，21.70)
经常	461	83(18.00)	(14.48，21.52)

下载: 导出CSV

表 2 膳食因子载荷量

Table 2. Dietary factor load

食物	传统模式	油炸腌熏甜点模式	优质蛋白模式
谷类	0.773	-0.209	0.063
肉类	0.776	0.214	-0.050
鱼、水产品	0.221	0.451	0.213
蔬果	0.667	-0.259	0.265
奶类	-0.154	0.101	0.725
蛋类	0.336	-0.015	0.665
豆类	0.120	0.114	0.703
甜点	-0.140	0.608	0.196
油炸	-0.173	0.786	0.024
腌熏	-0.028	0.701	-0.090
特征根	2.252	1.987	1.150
方差贡献率	18.961	18.621	16.304
累计贡献率	18.961	37.582	53.886

下载: 导出CSV

表 3 总人群高尿酸血症回归模型

Table 3. Regression model of total population hyperuricemia

因素	分类	单因素	调整模型
因素	分类	OR(95% CI)值	OR(95% CI)值
传统模式	T1	1.00	1.00
	T2	0.97(0.77~1.22)	0.94(0.74~1.21)
	T3	1.31(1.05~1.64)	1.11(0.88~1.40)
	T4	1.24(1.00~1.55)	1.05(0.83~1.33)
油炸腌熏甜点模式	T1	1.00	1.00
	T2	1.21(0.96~1.52)	1.01(0.79~1.28)
	T3	1.28(1.02~1.61)^a	1.02(0.79~1.29)
	T4	1.39(1.10~1.74)^b	1.09(0.85~1.39)
优质蛋白模式	T1	1.00	1.00
	T2	0.78(0.64~0.97)^a	0.83(0.66~1.03)
	T3	0.66(0.53~0.82)^b	0.74(0.59~0.94)^a
	T4	0.63(0.50~0.78)^b	0.77(0.61~0.97)^a
性别	男=1，女=0		3.58(2.59~4.96)^b
年龄			1.00(0.99~1.01)
吸烟	不吸烟=1		1.00
	已戒烟=2		1.14(0.95~1.38)
	吸烟=3		1.38(1.07~1.80)^a
饮酒	不喝酒=1		1.00
	已戒酒=2		1.02(0.74~1.41)
	喝酒=3		1.31(1.04~1.65)^a
体育锻炼	是=1，否=0		0.79(0.67~0.94)^b
血脂异常	是=1，否=0		1.86(1.53~2.25)^b
BMI(kg/m²)			1.12(1.09~1.15)^b
FPG(mmol/L)			0.88(0.83~0.94)^b
注：^aP < 0.05，^bP < 0.01。

下载: 导出CSV

表 4 不同性别高尿酸血症回归模型

Table 4. Regression Model of Hyperuricemia in Different Sexes

因素	分类	男单因素	男调整模型	女单因素	女调整模型
因素	分类	OR(95% CI)值	OR(95% CI)值	OR(95% CI)值	OR(95% CI)值
传统模式	T1	1.00	1.00	1.00	1.00
	T2	0.89(0.69~1.14)	0.94(0.73~1.23)	1.05(0.50~2.20)	0.87(0.39~1.91)
	T3	1.05(0.83~1.34)	1.08(0.84~1.38)	1.56(0.74~3.27)	1.31(0.59~2.87)
	T4	0.97(0.76~1.23)	1.00(0.78~1.29)	2.21(1.08~4.52)^a	1.58(0.72~3.46)
油炸腌熏甜点模式	T1	1.00	1.00	1.00	1.00
	T2	1.06(0.83~1.37)	1.00(0.77~1.30)	0.96(0.44~2.07)	0.90(0.40~2.05)
	T3	1.08(0.84~1.39)	0.98(0.75~1.27)	1.46(0.70~3.03)	1.37(0.62~3.01)
	T4	1.12(0.87~1.44)	1.01(0.78~1.32)	2.15(1.08~4.51)^a	1.97(1.00~4.30)^a
优质蛋白模式	T1	1.00	1.00	1.00	1.00
	T2	0.82(0.65~1.02)	0.84(0.67~1.06)	0.61(0.28~1.31)	0.69(0.30~1.58)
	T3	0.69(0.54~0.87)^b	0.73(0.57~0.93)^a	0.93(0.46~1.87)	1.06(0.49~2.29)
	T4	0.70(0.55~0.88)^b	0.77(0.60~0.99)^a	0.80(0.40~1.62)	0.84(0.19~4.18)
注：^a：P < 0.05，^b：P < 0.01。男、女调整模型调整变量为年龄、吸烟、饮酒、体育锻炼、血脂异常、BMI、FPG。

下载: 导出CSV

表 5 变量赋值说明

Table 5. Variable assignment description

指标名称	定义及赋值
解释变量
性别	女=0，男=1
年龄	连续性变量
吸烟	不吸烟=1，已戒烟=2，吸烟=3
饮酒	不饮酒=1，已戒酒=2，饮酒=3
传统模式	T1=1，T2=2，T3=3，T4=4
油炸腌熏甜点模式	T1=1，T2=2，T3=3，T4=4
优质蛋白模式	T1=1，T2=2，T3=3，T4=4
体育锻炼	不足20分钟体育锻炼=0，每天20分钟以上体育锻炼=1
工作方式	静坐为主=1，轻度活动=2，体力劳动=3
工作满意度	很不满意=1，不满意=2，一般=3，满意=4，很满意=5
精神压力	没有压力=1，压力较少=2，一般=3，压力较少=4，压力极大=5
血脂异常	否=0，是=1
BMI	连续性变量
FPG	连续性变量
BMI等级	消瘦=1，正常=2，超重=3，肥胖=4
因变量	未患HUA=0，患HUA=1

下载: 导出CSV

参考文献(18)

[1]	Yu TY, Jee JH, Bae JC, et al. Serum uric acid: A strong and independent predictor of metabolic syndrome after adjusting for body composition(Article)[J]. Metabolism, 2016, 65(4): 432-440. DOI: 10.1016/j.metabol.2015.11.003.
[2]	Hayashino Y, Okamura S, Tsujii S, et al. Association of serum uric acid levels with the risk of development or progression of albuminuria among Japanese patients with type 2 diabetes: a prospective cohort study[diabetes distress and care registry at Tenri (DDCRT 10)] [J]. Acta Diabetol, 2016, 53(4): 599-607. DOI: 10.1007/s00592-015-0825-x.
[3]	Kanbay M, Solak Y, Afsar B, et al. Serum uric acid and risk for acute kidney injury following contrast[J]. Angiology, 2017, 68(2): 132-144. DOI: 10.1177/0003319716644395.
[4]	Ickmans K, Clarys P, Nijs J, et al. Association between cognitive performance, physical fitness, and physical activity level in women with chronic fatigue syndrome[J]. J Rehabil Res Dev, 2013, 50(6):795-810. DOI: 10.1682/JRRD.2012.08.0156.DOI:10.1007/s12603-012-0076-2.
[5]	Xiong Z, Zhu C, Qian X, et al. Serum uric acid is associated with dietary and lifestyle factors in elderly women in suburban guangzhou in guangdong province of south China[J]. J Nutr Health Aging, 2013, 17(1): 30-34. DOI: 10.1007/s12603-012-0076-2.
[6]	张苗. 中老年人不同性别血尿酸与非酒精性脂肪肝的相关性研究——基于横断面研究和纵向队列研究[D]. 广州: 暨南大学, 2017. Zhang M. Study on the correlation between serum uric acid and non-alcoholic fatty liver in middle-aged and elderly people of different sexes: based on cross-sectional study and longitudinal cohort study[D]. Guangzhou: Jinan University, 2017.
[7]	中国成人血脂异常防治指南修订联合委员会. 中国成人血脂异常防治指南(2016年修订版)[J]. 中华健康管理学杂志, 2017, 11(1): 7-28. DOI: 10.3760/cma.j.issn.1674-0815.2017.01.003. Joint committee issued Chinese guideline for the prevention and treatment of dyslipidemia. 2016 Chinese guideline for the management of dyslipidemia in adults[J]. Chinese Journal of Health Management, 2017, 11(1): 7-28. DOI: 10.3760/cma.j.issn.1674-0815.2017.01.003.
[8]	Li X, Song P, Li J, et al. Relationship between hyperuricemia and dietary risk factors in Chinese adults: a cross-sectional study[J]. Rheumatol Int, 2015, 35(12): 2079-2089. DOI: 10.1007/s00296-015-3315-0.
[9]	Hak AE, Choi HK. Menopause, postmenopausal hormone use and serum uric acid levels in US women——the third national health and nutrition examination survey[J]. Arthritis Res Ther, 2008, 10(5): R116. DOI: 10.1186/ar2519.
[10]	Guan S, Tang Z, Fang X, et al. Prevalence of hyperuricemia among Beijing post-menopausal women in 10 years[J]. Arch Gerontol Geriatr, 2016, 64: 162-166. DOI: 10.1016/j.archger.2016.02.002.
[11]	Zhang M, Chang H, Gao Y, et al. Major dietary patterns and risk of asymptomatic hyperuricemia in Chinese adults[J]. J Nutr Sci Vitaminol (Tokyo), 2012, 58(5): 339-345. DOI: 10.3177/jnsv.58.339.
[12]	Chuang SY, Lee SC, Hsieh YT, et al. Trends in hyperuricemia and gout prevalence: nutrition and health survey in Taiwan from 1993-1996 to 2005-2008[J]. Asia Pac J Clin Nutr, 2011, 20(2): 301-308.
[13]	王子楠, 庄媛媛. 膳食结构对高尿酸血症的影响[J]. 中国老年学杂志, 2015, 35(23): 6937-6938. DOI: 10.3969/j.issn.1005-9202.2015.23.134. Wang ZN, Zhuang YY. Effect of dietary structure on hyperuricemia[J]. Chin J Gerontol, 2015, 35(23): 6937-6938. DOI: 10.3969/j.issn.1005-9202.2015.23.134.
[14]	Liu X, Huang S, Xu W, et al. Association of dietary patterns and hyperuricemia: a cross-sectional study of the Yi ethnic group in China[J]. Food Nutr Res, 2018, 62. DOI: 10.29219/fnr.v62.1380.
[15]	Ekpenyong CE, Daniel N. Roles of diets and dietary factors in the pathogenesis, management and prevention of abnormal serum uric acid levels[J]. PharmaNutrition, 2015, 3(2): 29-45. DOI: 10.1016/j.phanu.2014.12.001.
[16]	Cui L, Meng L, Wang G, et al. Prevalence and risk factors of hyperuricemia: results of the Kailuan cohort study[J]. Mod Rheumatol, 2017, 27(6): 1066-1071. DOI: 10.1080/14397595.2017.1300117.
[17]	Shen Y, Wang Y, Chang C, et al. Prevalence and risk factors associated with hyperuricemia among working population at high altitudes: a cross-sectional study in Western China[J]. Clin Rheumatol, 201938(5): 1375-1384. . DOI: 10.1007/s10067-018-4391-9.
[18]	Li H, Zha X, Zhu Y, et al. An invert U-shaped curve: relationship between fasting plasma glucose and serum uric acid concentration in a large health check-up population in China[J]. Medicine(Baltimore), 2016, 95(16): e3456. DOI: 10.1097/MD.0000000000003456.