非综合征性唇腭裂高危因素与发病预测模型研究

Study on risk factors and morbidity predictive model of nonsyndromic cleft lip and palate

目的探讨非综合征性唇腭裂(nonsyndromic cleft lip and palate,NSCL/P)发病的主要危险因素;评估这些主要危险因素在NSCL/P发病中的相对重要性,最终确立NSCL/P发病概率的预测模型,为优生网络的构建奠定基础。方法采用1∶1配对病例对照研究,病例组来源于2006年9月至2007年9月在潍坊医学院附属医院、潍坊市人民医院、菏泽市立医院、烟台毓璜顶医院口腔科住院,年龄在12岁以下患有NSCL/P的儿童76例;对照组为来源于同一机构门诊或病房或同一居住区符合配对条件的非唇腭裂儿童76名。根据拟定的42项危险因素编制调查表,对病例组患儿与对照组儿童的父母进行调查,数据经审核后录入Excel 2003建立数据库。首先使用条件Logistic回归对资料进行单因素分析,再对单因素筛选的变量结合专业知识进行多因素分析,筛选主要危险因素并建立回归模型,根据危险因素分别建立分类树与LogitBoost算法的发病概率预测模型,采用受试者工作特征曲线(ROC曲线)对两模型进行评价,从而确立本研究中NSCL/P发病概率的预测模型。结果病例组与对照组作对比分析,进入条件Logistic回归模型的变量有:母亲孕期感染史(P=0.010)、家族遗传史(P=0.009)、母孕期饮食是否规律(P=0.007)、胎次(P=0.004)、母亲孕期异常情绪史(P<0.001)、父亲学历(P<0.001)。经ROC曲线评价,确立分类树模型可用来预测NSCL/P的发病概率。结论母亲孕期感染、家族遗传、母亲孕期饮食不规律、胎次、母亲孕期异常情绪是NSCL/P发病的促进因素,且其对NSCL/P发病的影响作用依次增强;父亲学历是该病的保护因素。经ROC曲线评价,最终确立分类树模型为NSCL/P发病概率的预测模型。

Objective To discuss the main risk factors of nonsyndromic cleft lip and palate (NSCL/P), to evaluate the relative importance among these risk factors in occurrence of NSCL/P, and to propose the reasonable predictive model for aristogenesis net. Methods A hospital-based 1:1 matched case-control study design was applied to our epidemiological study.A total of 76 nonsyndromic cleft lip and palate children aged 0~12 years were selected from Department of Stomatology of Affiliated Hospital of Weifang Medical College, People's Hospital of Weifang,Heze Municipal Hospital,Yuhuangding Hospital of Yantai from Sep.2006 to Sep. 2007 as collecting cases, and control cases were chosen from the same section or in the same living area with the matched patients. A self-administered questionnaire prepared according to risk factors of NSCL/P was used to collect information, that is, the parents of patients and controls of NSCL/P were asked some questions about these risk factors and the answers were filled in the questionnaire by investigators. After the investigation, these data were cleaned up by pair and input into computer to establish data base in Excel 2003. First carry on the multivariate stepwise regression analysis in univariate condition Logistic regression analysis foundation, and then discover the risk factors from multitudinous possible influencing factor and establish the regression model. Establish the classification tree and the LogitBoost predictive models separately according to the risk factors,and appraise the two models using the ROC curve and establish the reasonable predictive model of NSCL/P. Results There were 6 risk factors in the model which were related to NSCL/P,namely,maternal infectious history (P=0.010), genetic family history (P=0.009), maternal disorder diet (P=0.007), birth order (P=0.004),maternal stress during the first trimester of pregnancy(P<0.001)and fathers' education level(P<0.001). After the evaluation with ROC curve,the classification tree model was established,which was used to predict the morbidity of NSCL/P. Conclusion Birth order, maternal stress during the first trimester of pregnancy, genetic family history, maternal disorder diet and maternal infectious history are promoting factors of NSCL/P,and the effects on occurrence of NSCL/P are enhanced. Fathers' education level is a protecting factor. According to the investigation,the classification tree model is established as the predictive model of the morbidity of NSCL/P.