Research of a model development for tooth loss prediction

本研究的目的是评估社区居住成年人牙齿脱落的三年发生率和危险因素。研究对象为新泻县某镇有牙成人269人（男性128人，女性141人），分别于1997年（基线检查）和2000年（随访检查）进行口腔检查。随访率为37.2%。基线时平均年龄为60.6岁（SD=12.8）。我们采用多元逻辑回归模型，通过口腔检查变量和基线调查问卷，量化牙齿特异性和个人水平的牙齿脱落危险因素。三年内牙齿脱落的发生率在牙齿水平为4.9%，在个人水平为48%。平均每年掉牙0.33颗。个体水平多元logistic回归分析结果显示，牙齿脱落在10 ~ 27颗现牙、蛀牙、口腔症状、1年内就诊史、不使用牙间刷和牙线的人群中较为常见。采用多牙水平logistic回归分析，发现智齿、蛀牙、冠牙、桥基牙和活动牙的牙齿脱落率较高。

The purpose of this study was to evaluate three-year incidence of and risk factors for tooth loss in community-dwelling adults. Subjects were 269 (male 128,female 141) dentate adults in a town of Niigata Prefecture who visited both oral examination that were carried out in 1997 (baseline examination) and 2000 (follow-up examination). The follow-up rate was 37.2%. The average age at baseline was 60.6 (SD=12.8). We used multiple-logistic regression model to quantify tooth-specific and person-level risk factors for tooth loss by using variables of oral examination and questionnaire at baseline.Three-year incidence of tooth loss was 4.9% in tooth-level, 48% in person-level. The average number of missing teeth per year was 0.33.The results of person-level multiple logistic regression analysis showed that tooth loss was more common in persons with 10-27 present teeth, decayed teeth, oral symptom, history of dental visit within a year and habit not using interdental brush or dental floss. By using tooth-level multiple logistic regression analysis showed that incidence of tooth loss is higher in wisdom tooth, decayed tooth, crowned tooth, abutment tooth of bridge and tooth with mobility.