Risk Factors Analysis For Early Childhood Caries And Construction Of Caries Risk Assessment Model

Objective:1.To understand the prevalence of caries in preschool children aged 3-5 years old in Shanxi Province and to explore the related factors of early childhood caries.2.To construct a caries risk assessment model using survey data for preschool children in Shanxi Provience.3.To evaluate the differentiation and calibration of the model using ROC curve and Hosmer-Lemeshow googness-of-fit test respectively,so as to better judge the practical application value of the model.Methods:1.Deta collection:Using the method of multi-stage stratified random sampling,12investigation sites at county(district)level were selected from 6 prefectures(cities)in the whole province.Then collect the list of kindergartens in the investigated counties(districts),sort them according to the scale of students aged 3-5 years old,select 3kindergartens from each district and county by proportional probability sampling method,and finally investigate a total of 5616 preschool children aged 3-5 years old.The data were collected by clinical examination combined with questionnaire survey.2.Data analysis:SPSS22.0 software was used for statistical analysis,descriptive statistical indicators(frequency,percentage,etc.)were used to describe the distribution and prevalence of the subjects,and chi-square test was used to determine the risk factors of early childhood caries.The independent factors affecting caries in children were screened by logistic regression,and the difference was statistically significant(P<0.05),the caries risk assessment model was established.3.Model evaluation:The ROC curve was used to evaluate the discrimination of the model,and the Hosmer-Lemeshow goodness-of-fit test was used to evaluate the calibration of the model.Results:1.In this survey,a total of 5,616 children aged 3-5 were investigated.After deleting the missing and abnormal data,the total number of remaining samples was 5,204,and the data efficiency rate was 92.7%.Among the subjects,the prevalence rate of early children caries was 58.9%,of which the caries prevalence rates of male and female were 58.5%and 59.2%respectively,and there was no significant statistical difference between them(P=0.588).The prevalence rates of caries in urban and rural children were 56.0%and61.7%respectively,and there was significant difference between them(P<0.01).The caries prevalence rates of 3-year-old,4-year-old and 5-year-old children were 47.5%,62.9%and 66.4%,with statistical differences(P<0.01).The univariate analysis showed that between the caries-free group and the caries-affected group,age,urban and rural areas,saliva ph,mother’s physical condition during pregnancy,feeding method,weaning time,children’s sleeping habits with bottles,whether parents chew food to feed their children,daily brushing times,whether parents help children brush their teeth,toothache in the past 12 months,whether they have visited their teeth,the reasons for their dental visits,whether they have regular oral examinations,frequency of eating sweets,frequency of drink sweetened beverages/sweetened milk,frequency of eating fruit,whether they have a partial eclipse,whether their diet is regular,frequency of colds was statistically significant(P<0.05).2.2.Logistic regression analysis showed that age,urban and rural areas,saliva ph,feeding method,weaning time,sleeping habits with bottles,feeding children with chewing food,toothache in the past 12 months,regular oral examination,frequency of eating sweets,frequency of sweetened milk and fruit intake and cold frequency were independent risk factors for early childhood caries.According to the screening variables,the prediction model Logit(p)is established:Logit(p)=-1.618+0.344*age-0.314*urban and rural-0.198*saliva ph-0.171*feeding mode+0.109*weaning time-0.245*bottle sleeping habits-0.197*chewing food feeding children+0.651*toothache in the past 12months+0.374*regularoralexamination+0.133*frequencyofeating sweets+0.066*frequency of sweetened milk+0.06*frequency of fruits+0.186*cold frequency.3.The ROC curve was used to evaluate the binary Logistic regression model.The area under the curve of the established model was 0.708,and the 95%confidence interval was 0.694-0.722,P=0.000,indicating that the predictive model had a good discriminative ability.The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the calibration of the model.The results showed that Hosmer-Lemeshowχ~2=9.362,P=0.316>0.05,indicating that the difference between the predicted value of the model and the actual observed value is not statistically significant,and the prediction model has a good calibration ability.Conclusion:1.The prevalence of ECC in preschool children aged 3-5 in Shanxi Province was58.9%,which increased with age.The prevalence of ECC in rural children was higher than that in urban areas.The prevalence of ECC in Shanxi Province is lower than that of developing countries,but much higher than that of developed countries,and there are children with high caries risk.This suggests that oral prevention work has a long way to go.It is necessary to actively screen people with high caries,carry out targeted prevention,and promote the process of pediatric oral diagnosis and treatment.2.The early childhood caries in Shanxi Province is related to a variety of factors.The related factors mainly include age,urban and rural areas,saliva ph,etc.Among them,toothache experience in the past 12 months is the strongest indicator of caries risk.3.The caries risk assessment model was established by survey data of preschool children aged 3-5 in Shanxi Province.The model can better predict the caries situation of preschool children in Shanxi Province after testing,and is suitable for large-scale screening and identification of high-risk children.