Abstract:Objective To analyze the correlation and intensity between several common chronic non-communicable diseases by the Apriori algorithm,and to provide data support for the screening and prevention of chronic non-communicable diseases.Methods 3 519 subjects aged 18 and above were selected from ten community in Shangcheng by multi-stage sampling method. The circumstances suffering from chronic non-communicable diseases were questioned and the physical examinations were performed. Apriori modeling algorithm in SPSS Clementine 12.0 data mining software was used to analyze the correlation and correlation strength among chronic non-communicable diseases.Results A total of 3 519 questionnaires were distributed and 3 345 valid questionnaires were recovered. The effective response rate was 95.06%. The prevalence of hypertension,dyslipidemia,diabetes,coronary heart disease,chronic respiratory disease,cerebral apoplexy and malignant tumor was 43.47%,The foreword incidence rate was hypertension(31.53%),dyslipidemia(16.65%)and diabetes(10.43%). There were 7 strong association rules for these 7 chronic non-communicable diseases,and strength of {dyslipidemia,coronary heart disease}→{hypertension},{diabetes,dyslipidemia}→{hypertension} and {coronary heart disease}→{hypertension} ranked first,second and third according to confidence. Among people aged 60 above,14 strong association rules were found,the strength of {diabetes,dyslipidemia,coronary heart disease}→{hypertension},{diabetes,coronary heart disease}→{hypertension} and {dyslipidemia,coronary heart disease}→{hypertension} ranked first,second and third according to confidence. The association rules of men and women were similar; seven strong association rules were found,with strength of {diabetes,dyslipidemia, coronary heart disease}→{hypertension},{diabetes,coronary heart disease}→{hypertension} and {dyslipidemia,coronary heart disease}→{hypertension} ranked top three according to confidence.Conclusion These seven common chronic non-communicable diseases were related to each other,especially hypertension and among people over the age of 60.
施明明,李娜,胡锦峰. 关联规则在社区居民常见慢性病关联性分析中的应用[J]. 预防医学, 2018, 30(8): 766-770.
SHI Ming-ming,LI Na,HU Jin-feng. Application of association rules in analyzing the relationship between common chronic diseases. Preventive Medicine, 2018, 30(8): 766-770.
[1] WHO. Global burden of disease Estimates for 2000-2012[R/OL] . [2017-12-13] . http://www.portal.pmnch.org/healthinfo/ global_burden_disease/estimates/en/. [2] WANG S,MARQUEZ P,LANGENBRUNNER J,et al. Toward a healthy and harmonious life in China:stemming the rising tide of non-communicable diseases[R] . The World Bank,2011:1-48. [3] WU Y,DANG J.Blue book of aging:China report of the development on aging cause[M] . Beijing:Social Sciences Academic Press,2013:1-274. [4] Wang X Q,Chen P J. Population ageing challenges health care in China[J] . Lancet,2014,383(9920):870. [5] 翟铁民,柴培培,魏强,等. 我国慢性非传染性疾病卫生费用与筹资分析[J] . 中国卫生经济,2014,33(2):14-17. [6] FILLENBAUM G G,PIEPER C F,COHEN H J,et al. Comorbidity of five chronic health conditions in elderly community residents:determinants and impact on mortality[J] . J Gerontol A Biol Sci Med Sci,2000,55(2):M84. [7] WOLFF J L,STARFIELD B,ANDERSON G. Prevalence,expenditures,and complications of multiple chronic conditions in the elderly[J] . Archives of Internal Medicine,2002,162(20):2269-2276. [8] SHIN A M,LEE I H,LEE G H,et al. Diagnostic analysis of patients with essential hypertension using association rule mining[J] . Healthcare Informatics Research,2010,16(2):77-81. [9] KIM H S,SHIN A M,MI K K,et al. Comorbidity study on type 2 diabetes mellitus using data mining[J] . Korean Journal of Internal Medicine,2012,27(2):197-202. [10] 罗旋,闫晓芳,罗明明,等. 海淀区居民慢性病相关危险因素的关联规则[J] . 中南大学学报(医学版),2017,42(5):570- 574. [11] 李澄,谢雁鸣,张寅,等. 基于关联规则Apriori算法的真实世界复方苦参注射液治疗乳腺恶性肿瘤西药联合应用特征研究[J] .中南药学,2017,15(11):1513-1519. [12] 王乾,吕亚男,李东红,等. 基于关联规则的乳腺肿块多模检索[J] . 哈尔滨理工大学学报,2017,22(2):124-128,134. [13] Y Zhao,Y Yao,Z Liu.An efficient distributed algorithm for mining association rules[M] // Ivan Stojmenovic. Parallel and Distributed Processing and Applications. Berlin:Springer,2007:41-44. [14] AGRAWAL R,LMEILINSKI T,SWAMI A. Mining association rules between sets of items inlargedatabases[EB/OL] .[2017- 12-13] . http://www.almaden.ibm.com/cs/quest/papers/sigmod93.pdf. [15] 郁旷明,寇俊,叶俊杰,等. 上城区居民主要慢性病患病情况调查[J] . 浙江预防医学,2013,25(4):37-39. [16] 周晓红,席胜军,赵琪,等. 杭州市下城区流动人口慢性病患病情况及危险因素分析[J] . 浙江预防医学,2015,27(12):1202-1205,1209. [17] 沈丽丽. 拱墅区慢性病患病情况及相关危险因素分析[D] . 杭州:浙江大学,2015. [18] 任仙龙,胡冬梅,王文娟,等. 关联规则在社区居民慢性病患病率分析中的应用[J] . 中国卫生统计,2013,30(6):818-820. [19] 李慧,叶浩森,徐郁,等. 珠海市居民慢性病患病情况及其影响因素分析[J] . 中华疾病控制杂志,2018,22(1):14-18. [20] 游弋,李宁,吴明,等. 辽宁省城区居民主要慢性病流行现况及影响因素分析[J] . 预防医学,2018,30(1):35-40,45. [21] 仇冰玉,许文青,曹姗,等. 老年疾病负担与公共卫生服务[J] . 社区医学杂志,2014,12(1):15-18.