ASSESSING RISK FOR EARLY CHILDHOOD CARIES

尽管许多研究已经检查了早期儿童龋齿（ECC），但其背后的病因仍然未知。近二十年来，研究人员一直试图开发敏感和特定的龋齿风险评估工具来识别有风险的个体，但收效甚微。本研究将利用全州和地方的横截面数据源来开发、测试和完善ECC风险关联模型，该模型将采用多因素和多层次的方法来研究各种个人家庭和社区特征，包括人口统计学、社会经济学、文化适应、健康信念和行为、获得护理和生理学，使用计算密集型的统计数据挖掘方法，如人工神经网络。目的是：利用1993-4年加州全州口腔健康需求评估的个人和家庭数据，并辅以社区数据，建立ECC风险关联模型；测试和完善上述ECC风险关联模型，并使用旧金山纵向氟化物清漆试验的对照组，使用个人、家庭和社区水平的数据，测试和完善最终的ECC风险关联模型作为风险预测模型。将开发用于非牙科环境的风险评估模型和针对种族/族裔的模型。这项研究与三个已经获得资助的研究，以及其他项目和拟议中心的核心相关联。该项目将为圣伊西德罗干预研究、旧金山焦点小组和凯撒永久医疗服务研究提供投入，研究与ECC相关的因素，并提供风险评估模型，可根据其研究数据进行测试和改进。这是第一个同时用数据挖掘技术检查与ECC相关的个人、家庭和社区水平特征的研究。

Although many studies have examined Early Childhood Caries (ECC), the etiology behind demonstrated racial/ethnic differences remain unknown. For almost two decades, researchers have tried to develop sensitive and specific caries risk assessment tools to identify at-risk individuals, but with limited success. This study will utilize statewide and local cross-sectional data sources to develop, test and refine an ECC risk association model, which will multifactorial and multilevel approach looking at various individual family and community characteristics, including demographics, socioeconomics, acculturation, health beliefs and behaviors, access to care, and physiology, using computationally intensive statistical data mining methods such as artificial neural networks. Aims are: to develop an ECC risk association model with individual- and family-level data from the 1993-4 statewide California Oral Health Needs Assessment supplemented with neighborhood-level data; to test and refine the above ECC risk association model and to test and refine th final ECC risk association model as a risk prediction model with the control group from a longitudinal fluoride varnish trial in San Francisco using individual, family, and neighborhood-level data. Risk assessment models to be used in non-dental settings and race/ethnicity specific models will be developed. This study interfaces with three already funded studies, as well as the other projects and the cores of the proposed center. The project will provide input to the San Ysidro intervention study, the San Francisco focus group, and the Kaiser Permanente health services research study about factors associated with ECC and provide risk assessment models that can be tested and refined with their study data. This is the first study to simultaneously examine individual-, family- and community-level characteristics related to ECC with data mining techniques.