中国6城市中老年人群颈椎骨关节炎患病危险因素的非条件Logistic回归分析

背景中老年人的颈椎骨关节炎发病率较高,国外对引起颈椎骨关节炎的危险因素已进行了一些研究,但大多数因素尚在讨论之中.目的调查国内不同地区间中老年人群颈椎骨关节炎的患病危险因素,为颈椎骨关节炎的社区预防干预提供依据.设计横断面调查.单位卫生部北京医院联合西安交通大学第二医院、河北医科大学中西医结合研究所、上海华东医院、南方医院、哈尔滨医科大学第二附属医院、四川大学华西医院共同完成.对象调查于2005-07/08完成.采用分层多阶段整群抽样方法,选择西安、石家庄、上海、广州、哈尔滨市及成都6城市40岁及以上具有正式户口的常住居民6 218人,均自愿参加调查.其中男2 916人,女3 302人,年龄男性40～94岁,女性40～86岁.方法对调查人群进行颈椎骨关节炎的流行病学问卷调查,对有临床症状者行X射线摄片.本次调查的基本抽样单位是居委会(城市)和村委会(农村).抽样方法以各城市为总体,分为城市和农村两层,第一阶段抽取区(县),第二阶段抽取街道(乡),第三阶段抽取居委会(村委会).颈椎骨关节炎诊断标准为颈椎骨关节炎临床症状阳性加X射线Kellgren&Lawrence分级2级及以上.问卷调查内容包括6个方面包括一般情况、现病史、既往史、体格检查、X射线片检查情况及疾病诊断情况,共计94个问题141个变量指标.分析颈椎骨关节炎患病率的影响因素应用多因素非条件Logistic回归法,表示疾病与暴露因素之间联系强度的指标用比值比(OR),若OR＞1,说明疾病发生与暴露因素呈正关联;若OR＜1,说明疾病发生与暴露因素呈负关联.主要观察指标各城市颈椎骨关节炎的患病率和OR.结果纳入调查对象6 218人,全部进入结果分析,无脱落.①国内6城市40岁及以上人群颈椎骨关节炎总患病率为23.6%,各城市患病率相比差异有显著性意义(P＜0.01).②Logistic回归分析结果年龄(OR=1.010～1.058)、使用蹲坑排便(OR=1.024～1.997)和高血压病史(OR=1.815～3.078)为大部分地区共同危险因素;北方地区的共同危险因素有日常爬楼梯或爬坡(OR=1.018～1.020);而喝色酒(OR=3.451,西安)、父亲骨关节炎史(OR=2.491,西安)、糖尿病史(OR=5.013,石家庄)、母亲骨关节炎史(OR=2.045,上海)、吸烟(OR=6.857,广州)、开始饮酒年龄(OR=3.044,广州)及专职体育运动(OR=9.020,哈尔滨)等危险因素分别在不同地区出现.结论国内6地区颈椎骨关节炎发病具有共同的危险因素,不同地区主要危险因素又有一定差异,可为今后颈椎骨关节炎的防治提供参考数据,减少医疗资源浪费.

Non-conditional Logistic regression analysis on risk factors of cervical osteoarthritis in the middle-aged and elderly from 6 cities of China

BACKGROUND: Incidence rate of cervical osteoarthritis in the middleaged and elderly is high. Some researches on risk factor causing cervical osteoarthritis have been performed abroad, but most of the factors are being discussed.OBJECTIVE: To investigate the occurring cervical osteoarthritis risk factors in the middle-aged and elderly from different regions of China and provide evidences for prevention and intervention of cervical osteoarthritis in community.DESIGN: Cross-sectional study.SETTING: Beijing Hospital, Ministry of Health, together with Second Hospital of Xi'an Jiantong University, Institute of Integrated Traditional and Western Medicine of Hebei Medical University, Shanghai Huadong Hospital, Nanfang Hospital, Second Affiliated Hospital of Harbin Medical University, West China Hospital of Sichuan University.PARTICIPANTS: The investigation was conducted from July to August2005. On the basis of stratified multi-stage cluster sampling method, 6 218formal registered permanent residents of over 40 years old from Xi'an,Shijiazhuang, Shanghai, Guangzhou, Harbin and Chengdu were enrolled.They all agreed to join the investigation voluntarily. There were 2 916males of 40-94 years and 3 302 females of 40-86 years.METHODS: Questionnaire investigation of epidemiology of cervical osteoarthritis was performed in the testees, and radiograph was used in the persons with clinical symptom. The basic sample unit was neighborhood committee (city) and village committee (countryside). Sampling method:Taking each city as a whole, composed of two levels, namely city and countryside, in the first phase the persons were extracted from district (county),in the second phase from sub-district (countryside), in the third phase from neighborhood committee (village eommittee). Diagnosis standard of cervical osteoarthritis was positive clinical symptom and 2 grade or above of radiograph Kellgren & Lawrence grading. The content of questionnaire contained 6 aspects: general condition, history of present illness, history of past illness, physical check-up, radiographs and disease diagnosis, totally94 questions and 141 variation indexes. Influential factors of prevalence rate of cervical osteoarthritis were analyzed using multifactor non-conditional Logistic regression analysis. Odds ratio (OR) was used for expressing index of strength of relationship between disease and exposures. If OR ＞ 1,it was indicated that there was positive correlation between disease occurrence and exposures. If OR ＜ 1, it was suggested that there was negative correlation between disease occurrence and exposures.MAIN OUTCOME MEASURES: Prevalence rate of cervical osteoarthritis in each city and OR.RESULTS: Totally 6 218 investigational subjects were included in the result analysis, without drop out. ①Total prevalence rate of cervical osteoarthritis in population of 40 years or above from 6 domestic cities was23.6%. There was significnat difference of prevalence rate in each city (P＜0.01). ②Result of Logistic regression analysis: Age (OR=1.010-1.058),defecation with squat ting pot (OR =1.024-1.997) and history of hypertension (OR =1.815-3.078) were common risk factor in most areas. In northern area the common risk factor compos ed of daily stair climbing or grade climbing (OR =1.018-1.020), while drinking colored wine (OR=3.451, Xi'an), history of osteoarthri tis of father (OR =2.491, Xi'an), history of diabetes (OR =5.013, Shijiazhuang), history of osteoarthritis of mother (OR =2.045, Shanghai), smoking (OR =6.857, Guangzhou), age of starting drinking (OR =3.044, Guangzhou) and full-time athletic sports (OR=9.020, Harbin), etc. emerged in different areas.CONCLUSION: The onset of cervical osteoarthritis has the same risk factor in 6 domestic areas, and main risk factor in different areas has certain differences, which can provide reference data for the prevention and cure of cervical osteoarthritis for the future and reduce waster of medical resources.