Association between prediabetes and dietsry patterns for the elderly in rural China
-
摘要:
目的 分析我国农村老年人不同膳食模式及其他影响因素与糖尿病前期的关系。 方法 选取2010―2012年中国居民营养与健康状况监测中的4 577名60~79岁的农村人群作为研究对象。采用食物频率调查问卷进行膳食调查，采用因子分析法提取膳食模式，并运用Logistic回归分析模型分析膳食模式与糖尿病前期之间的关系；应用分类树模型分析糖尿病前期的影响因素。 结果 因子分析得到3种膳食模式，分别为面食杂粮模式、水果蔬菜模式和水产豆蛋模式；Logistic回归分析模型分析发现面食杂粮模式与水产豆蛋模式存在交互作用，其中面食杂粮模式T3水平且水产豆蛋模式T1水平与糖尿病前期低风险相关（OR=0.518，95% CI：0.328~0.819），而面食杂粮模式T1水平且水产豆蛋模式T4水平与糖尿病前期高风险相关（OR=2.060，95% CI：1.347~3.151）；分类树模型筛选出8个影响因素：中心性肥胖、体重指数（body mass index，BMI）、收入水平、年龄、水产豆蛋模式、面食杂粮模式、水果蔬菜模式和饮酒，以及3个农村老年人糖尿病前期高危人群。 结论 膳食因素与糖尿病前期密切相关，采用平衡膳食，保持健康体重和良好的生活习惯有助于降低农村老年人糖尿病前期的发病风险。 Abstract:Objective We aimed to find out the dietary patterns and analyze the relationship between influencing factors (especially dietary patterns) and prediabetes in Chinese rural elderly. Methods Data were obtained from the Chinese nutrition and health surveillance during 2010 to 2012, and a total of 4 577 participants aged 60~79 years old were included. A food frequency questionnaire was performed to conduct a dietary survey. Factor analysis was used to extract dietary patterns, while Logistic regression models were used to analyze relationship between dietary patterns and prediabetes. In addition, classification tree model was applied to analyze the influencing factors of prediabetes. Results Three dietary patterns were found among the elderly participants by factor analysis, including the wheat-coarse cereals, fruit-vegetables, and aquatic-beans-eggs patterns. The result of Logistic analysis showed that there was an interaction between wheat-coarse cereals pattern and aquatic-beans-eggs pattern.In detail, the T3 level of wheat-coarse cereals pattern and the T1 level of aquatic-beans-eggs pattern was associated with low risk of prediabetes(OR=0.518, 95% CI:0.328-0.819), while the T1 level of wheat-coarse cereals pattern and the T4 level of aquatic-beans-eggs pattern was associated with high risk of prediabetes(OR=2.060, 95% CI:1.347-3.151). In addition, eight influencing factors and three group at high-risk for prediabetes were identified by classification tree model. The influencing factors were central obesity, BMI, income, age, aquatic-beans-eggs pattern, wheat-coarse cereals pattern, fruit-vegetables pattern, and drinking. Conclusions Our results suggest that dietary factors are closed related to prediabetes, and adopting a balanced diet, maintaining a healthy weight and good living habits may help to reduce the risk of prediabetes in chinese rural eldery. -
Key words:
- Prediabetes /
- Elderly /
- Dietary pattern /
- Classification tree
-
图 1 糖尿病前期影响因素分类树模型中心性肥胖分支情况
注：否为血糖正常；是为糖尿病前期。
Figure 1. Central obesity branches in the classification tree model of prediabetes influencing factors
图 2 糖尿病前期影响因素分类树模型非中心性肥胖分支情况
注：否为血糖正常；是为糖尿病前期。
Figure 2. Non-central obesity branches of the prediabetes influencing factors in the classification tree model
表 1 糖尿病前期影响因素赋值表
Table 1. Assignment table of influencing factors of prediabetes
变量 赋值 性别 1=男；2=女 年龄组(岁) 1=60~；2=65~；3=70~；4=75~79 文化程度 1=小学及以下；2=初中；3=高中；4=大专及以上 收入水平(元) 1=低收入；2=中等收入；3=高收入；4=不回答 BMI(kg/m2) 1=正常体重；2=低体重；3=超重；4=肥胖 吸烟 0=否；1=是 饮酒 0=否；1=是 身体活动 0=不充足；1=充足 中心性肥胖 0=否；1=是 静坐时间(h/d) 1=＜3；2=3~；3=＞5 膳食模式因子得分 1=T1；2=T2；3=T3；4=T4 
下载: 导出CSV
表 2 3种膳食模式及其因子载荷
Table 2. Three dietary patterns and factor loadings
面食杂粮模式 水果蔬菜模式 水产豆蛋模式 食物类 因子载荷 食物类 因子载荷 食物类 因子载荷 面及面制品 0.804 水果 0.615 水产 0.530 其他谷类 0.664 蔬菜 0.589 大豆及制品 0.491 蛋类 0.363 坚果 0.517 蛋类 0.469 糕点 0.481 奶及奶制品 0.404 薯类 0.339 注：表中仅列出因子载荷＞0.3的食物类。
下载: 导出CSV
表 3 3种膳食模式与糖尿病前期关系的Logistic回归分析模型分析结果
Table 3. Analysis results of Logistic regression models of the relationship between three dietary patterns and prediabetes
膳食模式 OR(95% CI)值 χ2值 P值 OR(95% CI)值a χ2值 P值 水果蔬菜模式 10.069 0.018 9.909 0.028 T1 1.000 1.000 T2 1.110(0.914~1.349) 1.103 0.294 1.130(0.913~1.399) 1.267 0.260 T3 0.862(0.704~1.057) 2.039 0.153 0.874(0.699~1.091) 1.418 0.234 T4 0.835(0.679~1.027) 2.905 0.088 0.834(0.664~1.046) 2.448 0.118 面食杂粮模式水产豆蛋模式 58.123 ＜0.001 55.482 ＜0.001 T1*T1 1.000 1.000 T1*T4 2.001(1.363~2.937) 12.536 ＜0.001 2.060(1.347~3.151) 11.103 0.001 T3*T1 0.566(0.373~0.859) 7.143 0.008 0.518(0.328~0.819) 7.923 0.005 注：表中仅列出有统计学意义的交互作用结果；aOR值为调整OR值，调整因素为性别、年龄、收入水平、吸烟、饮酒、BMI、中心性肥胖、身体活动、静坐时间。
下载: 导出CSV
表 4 3个糖尿病前期高危人群
Table 4. Three high-risk groups of prediabetes
群体 糖尿病前期高危人群 1 中心性肥胖(男性≥90 cm/女性≥85 cm)且BMI≥28 kg/m2、中等收入 2 正常腰围(男性＜90 cm/女性＜85 cm)、65~79岁、水产豆蛋模式因子得分T4水平、不饮酒但水果蔬菜模式T2及以下水平 3 正常腰围(男性＜90 cm/女性＜85 cm)、65~79岁、水产豆蛋模式因子得分T3及以下水平、面食杂粮模式T2及以下水平且饮酒
下载: 导出CSV
表 5 糖尿病前期影响因素的分类树模型终端节点情况
Table 5. Terminal nodes of the classification tree model for prediabetes influencing factors
终端节点 节点 收益 响应(%)c 指数(%)d 例数a 比例(%)b 例数 比例(%) 10 114 2.49 52 6.06 45.61 243.33 17 87 1.90 35 4.08 40.23 214.61 16 211 4.61 54 6.29 25.59 136.52 9 281 6.14 65 7.58 23.13 123.40 18 170 3.71 37 4.31 21.76 116.10 5 792 17.30 172 20.05 21.72 115.85 14 187 4.09 34 3.96 18.18 96.99 15 617 13.48 110 12.82 17.83 95.10 3 1 432 31.29 205 23.89 14.32 76.37 12 686 14.99 94 10.96 13.70 73.10 注：a表示在终端节点中样本类别为“1”即糖尿病前期的例数；b为收益n与根节点中糖尿病前期病例数(858)的比值；c表示终端节点样本中样本类别为“1”即糖尿病前期的比例；d为[n%(终端节点中样本类别为“1”)/n%(根节点中样本类别为“1”)]×100%。
下载: 导出CSV
-
[1] 宁光, 赵文华.中国糖尿病回顾与展望[M].北京:人民卫生出版社, 2018.Ning G, Zhao WH. Review and prospect of diabetes in China[M]. Beijing:People's Medical Publishing House, 2018. [2] Wang L, Gao P, Zhang M, et al. Prevalence and ethnic pattern of diabetes and prediabetes in China in 2013[J]. JAMA, 2017, 317(24):2515-2523. DOI: 10.1001/jama.2017.7596. [3] Tabák AG, Herder C, Rathmann W, et al. Prediabetes:a high-risk state for diabetes development[J]. Lancet, 2012, 379(9833):2279-2290. DOI: 10.1016/S0140-6736(12)60283-9. [4] 中华人民共和国国家卫生和计划生育委员会疾病预防控制局.中国居民营养与慢性病状况报告[R].北京: 人民卫生出版社, 2015.Disease Prevention and Control Bureau of National Health and Family Planning Commission of the People's Republic of China. Report on the status of nutrition and chronic diseases of Chinese residents[R]. Beijing: People's Medical Publishing House, 2015. [5] Zhang ML, Zhu YF, Li P, et al. Associations between dietary patterns and impaired fastingglucose in Chinese men:a cross-sectional study[J]. Nutrients, 2015, 7(9):8072-8089. DOI: 10.3390/nu7095382. [6] 赵丽云, 马冠生, 朴建华, 等. 2010-2012年中国居民营养与健康状况监测总体方案[J].中华预防医学杂志, 2016, 50(3):204-207. DOI: 10.3760/cma.j.issn.0253-9624.2016.03.002.Zhao LY, Ma GS, Piao JH, et al. Scheme of the 2010-2012 Chinese nutrition and health surveillance[J]. Chin J Prevent Med, 2016, 50(3):204-207. DOI: 10.3760/cma.j.issn.0253-9624.2016.03.002. [7] World Health Organization. Definition, diagnosis and classification of diabetes mellitus and its complications: report of a WHO consultation.Part1.Diagnosis and classification of diabetes mellitus[R]. Geneva: World Health Organization, 1999. [8] 中华人民共和国国家卫生和计划生育委员会. WS/T 428-2013中华人民共和国卫生行业标准《成人体重判定》[S].北京: 中国标准出版社, 2013.Health and Family Planning Commission of the People's Republic of China. WS/T 428-2013 people's republic of China health industry standard "adult weight determination"[S]. Beijing: China Standard Press, 2013. [9] 中国营养学会.中国居民膳食指南(2016)[M].北京:人民卫生出版社, 2016.Chinese Nutrition Society. Dietary guidelines for Chinese residents (2016)[M]. Beijing:People's Medical Publishing House, 2016 [10] 宓伟, 练武, 杨曼丽, 等.烟台市老年人膳食模式及其影响因素[J].卫生研究, 2017, 46(1):27-31. DOI: 10.19813/j.cnki.weishengyanjiu.201.01.005.Mi W, Lian W, Yang ML, et al. Analysis on dietary patterns and influencing factors among the elderly in Yantai City[J]. J Hygiene Res, 2017, 46(1):27-31. DOI: 10.19813/j.cnki.weishengyanjiu.201.01.005. [11] 朱徐慧, 章荣华, 周标, 等.浙江省老年人膳食结构与代谢综合征的关系[J].中华老年医学杂志, 2015, 34(10):1131-1136. DOI: 10.3760/cma.j.issn.0254-9026.2015.10.023.Zhu XH, Zhang RH, Zhou B, et al. Association between dietary patterns and metabolic syndrome among elderly people in Zhejiang Province[J]. Chin J Geriatr, 2015, 34(10):1131-1136. DOI: 10.3760/cma.j.issn.0254-9026.2015.10.023. [12] Ke Q, Chen C, He F, et al. Association between dietary protein intake and type 2 diabetes varies by dietary pattern[J]. Diabetol Metab Syndr, 2018, 10(48). DOI: 10.1186/s13098-018-0350-5. [13] Zhu Z, He H, Zhou J, et al. Soy product consumption and type 2 diabetes among adults in Beijing, China[J]. Asia Pac J Clin Nutr, 2019, 28(1):150-156. DOI: 10.6133/apjcn.201903_28(1).0020. [14] Vitale M, Masulli M, Calabrese I, et al. Impact of a Mediterranean dietary pattern and its components on cardiovascular risk factors, glucose control, and body weight in people with type 2 Diabetes:a real-life study[J]. Nutrients, 2018, 10(8):pii:E1067. DOI: 10.3390/nu10081067. [15] 武艳华, 史宝林, 葛丽平.分类与回归树分析方法及其在医学研究中的应用[J].河北北方学院学报(医学版), 2008, 25(6):72-73. DOI: 10.3969/j.issn.2095-1396.2008.06.039.Wu YH, Shi BL, Ge LP. Classification and regression tree analysis method and its application in medical research[J]. J Heibei North Univ(Medical Edition), 2008, 25(6):72-73. DOI: 10.3969/j.issn.2095-1396.2008.06.039. [16] Liao QJ, Zheng Z, Xiu S, et al. Waist circumference is a better predictor of risk for frailty than BMI in the community-dwelling elderly in Beijing[J]. Aging Clin Exp Res, 2018, 30(11):1319-1325. DOI: 10.1007/s40520-018-0933-x. [17] 关红军, 鲁俊华, 李丽娟, 等.应用分类树模型研究乳腺癌危险因素[J].中华乳腺病(电子版), 2013, 7(3):179-183. DOI: 10.3877/cmmj.issn.1674-0807.2013.03.006.Guan HJ, Lu JH, Li LJ, et al. Taxonomic tree model to study risk factors of breast cancer[J]. Chin J Breast Dis(Electronic Edition), 2013, 7(3):179-183. DOI: 10.3877/cmmj.issn.1674-0807.2013.03.006. -
点击查看大图
图(2) / 表(5)
计量
- 文章访问数: 411
- HTML全文浏览量: 228
- PDF下载量: 42
- 被引次数: 0