Identifying Patterns of Health Care Utilization among Physical Elder Abuse Victims Using Medicare Data and Legally Adjudicated Cases

项目总结 这项研究的总体目标是：提高对卫生保健利用模式的理解和 身体虐待老年人的相关费用，以改进早期识别和干预并告知 政策。在我们之前的研究中，得到了NIA GEMSSTAR和Beeson对Pi Rosen博士的资助，我们 利用与检察官办公室的独特伙伴关系，全面审查来自更多 超过200名受害者在两个大城市地区，这是最大的法律追溯系列 审理过的身体虐待老人案件。在拟议的研究中，我们计划将这些案例联系起来， 我们已经对其进行了广泛的临床分析，以获得医疗保险索赔数据。这项建议的具体目的 是：(1)使用医疗保险索赔描述受害者之前和之前的医疗保健利用率和模式 在发现后，重点关注卫生专业人员可能错过的识别滥用和 基于受害者特征的差异，(2)比较急诊就诊和住院的比率和模式 从医疗保险索赔数据中通过算法选择的控制组，以及(3) 比较受害者和对照组之间急诊室/医院以外的医疗保健利用模式。这个 所提出的研究将为了解老年人的医疗保健利用模式提供重要的见解 虐待受害者，重点关注是否存在错失的机会和是否出现暗示模式。我们计划 使用复杂的机器学习方法来提高我们识别模式的能力 身体虐待老人的暴露。这将为卫生保健的识别和干预策略提供信息 提供者和收集到的知识将支持卫生信息学工具的未来发展，以确定 潜在的受害者。对相关费用的调查结果将有助于确定身体虐待老年人的范围和影响。 这一创新方法利用了在NIA支持下收集的现有数据，并从成功 对虐待儿童和亲密伴侣暴力的研究方法，同时对其进行扩展。因为不需要额外的 受试者将被前瞻性地纳入这项研究，我们避免了许多典型的 虐待长辈的工作。我们在虐待老人、虐待儿童和疏忽照顾方面的多学科专家团队， 亲密伴侣暴力、急救医学以及统计学、卫生经济学和 计算机科学是唯一能够进行这项研究的学科。此前，我们已经做了开创性的工作 将成人保护服务与老年虐待受害者的保健服务和健康相关结果联系起来 警方数据库的医疗保健数据，这与这项提案高度相关。我们也有很深的经验 使用医疗保险索赔数据和机器学习进行研究。我们研究的长期目标是 利用更好地了解老年虐待受害者使用医疗保健的情况，提高保健能力 提供者识别、干预和防止受害，并告知政策变化以帮助弱势群体 人口。

PROJECT SUMMARY The over-arching aim of this research is to: improve understanding of the patterns of health care utilization and associated costs of physical elder abuse victims to improve early identification and intervention and to inform policy. In our prior research, supported by an NIA GEMSSTAR and Beeson grant to PI Dr. Rosen, we leveraged unique partnerships with prosecutors’ offices to comprehensively examine legal case files from more than 200 victims in two large metropolitan areas, representing the largest retrospective series of legally adjudicated physical elder abuse cases ever examined. For the proposed study, we plan to link these cases, for which we have done extensive clinical analysis, to Medicare claims data. The Specific Aims of this proposal are: (1) To use Medicare claims to describe rates and patterns of health care utilization of victims before and after detection, with a focus on potentially missed opportunities by health professionals to identify abuse and differences based on victim characteristics, (2) To compare rates and patterns of ED visits and hospitalizations of physical elder abuse victims to control groups selected algorithmically from Medicare claims data, and (3) To compare patterns of health care utilization other than EDs/hospitals between victims and controls. The proposed research will provide important insight into the patterns of health care utilization for physical elder abuse victims, focusing on whether missed opportunities exist and suggestive patterns emerge. We plan to employ sophisticated machine learning approaches to increase our ability to identify patterns suggestive of physical elder abuse exposure. This will inform strategies for identification and intervention by health care providers, and knowledge gleaned will support the future development of a health informatics tool to identify potential victims. Findings on associated costs will help define the scope and impact of physical elder abuse. This innovative approach leverages existing data gathered with NIA support and extrapolates from successful research approaches in child abuse and intimate partner violence while expanding on them. As no additional subjects will be prospectively enrolled as part of this research, we avoid many of the ethical concerns typical in elder mistreatment work. Our multi-disciplinary team of experts in elder abuse, child abuse and neglect, intimate partner violence, and emergency medicine as well as specialists in statistics, health economics, and computer science is uniquely able to conduct this research. Previously, we have done seminal work examining health care usage and health-related outcomes of elder abuse victims by linking adult protective services and police databases to health care data, which is highly relevant for this proposal. We also have deep experience in using Medicare claims data and machine learning for research. The long-term goal of our research is to leverage a better understanding of health care use by elder abuse victims to improve the ability of health care providers to identify, intervene, and prevent victimization and to inform policy changes to help this vulnerable population.