Multilevel analyses of oral health conditions among older adults in the All of Us Research Program

摘要 美国人口正在迅速老龄化。美国人的寿命更长，患有更多的慢性病， 更多牙齿由于口腔和全身条件之间的关系，有必要对牙齿护理 更好地与初级保健相结合。在获得牙科保健方面的口腔健康不平等现象普遍存在 在老年人中。常规牙科护理不包括在医疗保险和大的州一级的差异存在， 牙科医疗补助美国老年人的口腔健康研究不足，因为很少有全国性的大型研究， 规模研究与纵向组成部分，包括医疗牙科，个人和社区层面的数据。 有了这个补充，来自美国国立卫生研究院全美国（AoU）研究项目的二级数据分析将 扩展有关51岁以上成年人的信息，可从纵向，全国代表性的美国卫生 和目前正在分析的退休研究（HRS）（NIDCR R 03 DE 030161 -02）。我们的R 03目标是 研究老年人口腔健康的纵向变化与认知，退休， 医疗、情境、社会心理、行为和牙科使用数据。HRS和AoU数据集都提供了 将口腔健康状况与医疗条件、人口统计学、生活方式、心理社会因素联系起来的信息，以及 整体健康。HRS和AoU都包括自我报告调查数据。此外，AoU包括临床- 从参与者的电子健康记录（EHR）在医疗机构中获得的确定信息。AOU 旨在招收多样化的，传统上代表性不足的群体。AoU受控层级数据包括3- 数字邮政编码，提供地理背景信息，以探索社会决定因素， 健康使用多层次分析。老年人口腔健康状况差，部分牙齿脱落和缺牙 对他们的生活质量和功能状态产生负面影响。AoU分析受控层数据集将 基于EHR数据构建了3组具有不同口腔条件的组用于比较：1）完全 无齿，2）部分无齿，和3）有齿的证据（例如，龋齿、牙周炎）。的 具体目标是：目标1a）将社会人口因素的平均值（SD）和频率分布制成表格 在受控层AoU数据以及来自调查和EHR的医疗、认知和社会/生活方式风险指标中 与这3个牙齿状态组相关的数据（上文）。目标1b）使用这些工具开发多变量模型 3牙科状况组和牙科利用，以确定其对自测健康，认知和 其它医学状况（例如，糖尿病、心血管疾病、中风）。目标1c）建立多层次模型 包括使用邮政编码数据来表示参与者居住的地理区域的上下文变量， 检查结果中的个人和地区差异。目的2：将AoU结果与2018 HRS数据进行比较， 评估自评健康，认知和医疗结果的人群差异。HRS和AoU都是 具有不同关键特征的大规模、里程碑式的研究。通过分析这两个数据集，我们将能够 利用每一个最好的功能，以进一步了解老年人的口腔健康。

Abstract The U.S. population is rapidly aging. Americans are living longer, with more chronic conditions, and retaining more teeth. Because of the relationships between oral and systemic conditions, there is a need for dental care to become better integrated with primary care. Oral health inequalities in access to dental care are prevalent among older adults. Routine dental care is not included in Medicare and large state-level variations exist in dental Medicaid benefits. The oral health of older adults in the US is understudied due to few national large- scale studies with longitudinal components that include medical dental, individual and community level data. With this supplement, secondary data analysis from the national NIH All of US (AoU) Research Project will expand information about adults age 51+ available from the longitudinal, nationally representative US Health and Retirement Study (HRS) currently being analyzed (NIDCR R03 DE030161-02). Our R03 goals are to study longitudinal changes in the oral health of older adults in relationship to changes in cognition, retirement, medical, contextual, psychosocial, behavioral and dental utilization data. Both HRS and AoU datasets provide information to link oral health status to medical conditions, demographics, lifestyle, psychosocial factors, and overall health. The HRS and AoU both include self-report survey data. Additionally, the AoU includes clinically- determined information from participants' electronic health records (EHR) obtained in healthcare settings. AoU seeks to enroll diverse, traditionally under-represented groups. The AoU Controlled Tier level data includes 3- digit ZIP code that provides access to geographical contextual information to explore social determinants of health using multilevel analyses. Poor oral health, partial tooth loss, and edentulism among older adults negatively affect their quality of life and functional status. An AoU analytical Controlled Tier dataset will be constructed with 3 groups having different oral conditions based on EHR data for comparison: 1) completely edentulous, 2) partially edentulous, and 3) evidence of being dentate (e.g., dental caries, periodontitis). The specific aims are: AIM 1a) Tabulate the means (SD) and frequency distribution of socio-demographic factors in Controlled Tier AoU data and medical, cognition, and social/lifestyle risk indicators from survey and EHR data associated with these 3 dental status groups (above). Aim 1b) Develop multivariable models using these 3 dental status groups and dental utilization to determine their influence on self-rated health, cognition, and other medical conditions (e.g., diabetes, cardiovascular disease, stroke). Aim 1c) Develop multilevel models including contextual variables using zip code data to represent geographic regions where participants reside to examine individual and area variations in outcomes. AIM 2: Compare AoU findings with 2018 HRS data and assess population differences in self-rated health, cognition, and medical outcomes. HRS and AoU are both large-scale, landmark studies with different key features. By analyzing both datasets, we will be able to take advantage of the best features of each one to further our understanding of the oral health of older adults.