Understanding the effect of rurality and social risk factors on barriers to care and surgical outcomes.

背景：《使命法案》改善了退伍军人在退伍军人内部获得护理的机会 行政管理(VA)和社区系统。一个基本的假设是，更快的护理和更多的选择 会带来更好的护理。然而，护理支离破碎与更长的住院时间、再住院时间、 和死亡率。少数族裔和低社会经济水平的人术后并发症和再入院比例较高 状态(SES)患者。低SES也与脆弱有关，这是30天最好的预测因素之一 术后并发症和再次入院治疗。尽管对健康有深远的影响 结果，在退伍军人管理局质量措施的风险调整中没有社会风险因素，进一步加剧了 少数族裔和低社会经济地位人群的差异。这一战略可能会进一步限制用于护理的资源 弱势群体，因为许多退伍军人在经济上处于不利地位，可能会增加可避免的 护理服务的交付成本。另一个主要问题是护理分散。尽管如此，非退伍军人护理和 绩效指标中不存在护理碎片化。我们的目标是确定社会风险因素和 影响手术结果的护理碎片化，以告知退伍军人管理局质量指标政策和机构 资源配置。我们通过将手术结果数据与1)退伍军人管理局/中心联系起来，改进了目前的做法 对于Medicare和Medicaid Services(CMS)索赔数据，2)退伍军人管理局收费文件，以识别在 3)使用更细粒度的社会风险因素和邻里劣势。 重要性/影响：我们的重要性是使用社会风险因素、乡村、居住环境等来模拟手术结果 弱势社区和护理支离破碎，以确定造成医疗保健差距的因素 并告知退伍军人事务部的政策。其影响是使用社会风险因素和护理碎片化来制定质量指标。 高铁与发展优先领域：农村健康、健康公平、医疗保健价值和医疗保健信息学。 创新：结合不同的数据源，使用两种传统的参数方法开发预测模型 和探索性机器学习技术，为临床医生和管理员提供结果和 进行必要的经济分析，以改变体制做法，使我们最脆弱的退伍军人受益。 具体目标： 目标1：通过评估种族、种族、社会经济地位、地点的贡献来确定影响手术结果的因素 住院和护理分散对手术并发症、再入院和死亡率的影响 假设：使用少数族裔身份、社会经济地位、居住地和护理分散程度将确定 术后并发症、再入院和死亡率的重要危险因素 目标2：评估社会风险因素和护理分散对医院绩效指标的影响 再入院和死亡率 假设：在风险调整模型中显著包括社会风险因素和护理碎片化 更改退伍军人医院再入院和死亡率方面的绩效排名 目的3：确定居住地、护理碎片化、社会经济地位和少数民族地位与急性发作的关系 和长期退伍军人外科医疗服务利用情况，为退伍军人管理局资源分配提供信息 假设：低SES、乡村、护理碎片化和少数族裔地位与较高的VA资源相关 利用 方法：使用传统的参数和探索性机器学习技术进行定量分析 在不同的数据集上执行，以使用护理碎片化来开发手术结果的预测模型， 农村和社会风险因素风险调整为医疗合并症，并应用于退伍军人管理局质量指标。 实施/下一步：使用社会风险因素和护理部署质量指标模型 退伍军人制度内部的碎片化。调整资源配置，以应对社会风险因素。

Background: The Mission Act provides improved Veteran access to care both within the Veterans Administration (VA) and community systems. An underlying assumption is that faster care with more choices results in better care. However, care fragmentation is associated with increased length of stay, readmissions, and mortality. Postoperative complications and readmissions are higher in minority and low socioeconomic status (SES) patients. Low SES is also associated with frailty, one of the best predictors of 30-day postoperative complications and hospital readmissions. Despite having a profound influence on health outcomes, social risk factors are absent from risk adjustment for VA quality measures, further exacerbating disparities in minority and low SES populations. This strategy may further constrain resources to care for vulnerable populations, as many Veterans are economically disadvantaged and potentially adding avoidable costs to care delivery. Another major issue is care fragmentation. Nevertheless, the impact of non-VA care and care fragmentation is absent in performance metrics. Our goal is to identify social risk factors and levels of care fragmentation that affect surgical outcomes to inform VA quality metric policy and institutional resource allocation. We improve upon current practice by joining surgical outcomes data with 1) VA/Centers for Medicare & Medicaid Services (CMS) claims data, 2) VA fee-basis files to identify encounters outside of the VA health system and 3) using more granular proxy social risk factors and neighborhood disadvantage. Significance/Impact: Our significance is modeling surgical outcomes using social risk factors, rurality, living in a disadvantaged neighborhood and care fragmentation to identify factors contributing to health care disparities and to inform VA policy. The impact is to develop quality metrics using social risk factors and care fragmentation. HSR&D priority areas: Rural Health, Health Equity, Health Care Value and Health Care Informatics. Innovation: Joining diverse data sources to develop predictive models using both traditional parametric methods and exploratory machine learning techniques to provide clinicians and administrators with outcomes and economic analyses necessary to change institutional practices to benefit our most vulnerable Veterans. Specific Aims: Aim 1: Identify factors affecting surgical outcomes by assessing the contributions of ethnicity, race, SES, place of residence and care fragmentation to surgical complications, readmissions and mortality Hypothesis: Using ethnic/racial minority status, SES, place of residence and care fragmentation will identify important risk factors for postoperative complications, readmissions, and mortality Aim 2: Assess the impact of social risk factors and care fragmentation on hospital performance metrics for readmissions and mortality Hypothesis: Including social risk factors and care fragmentation in risk adjustment models significantly changes VA hospital performance rankings with respect to readmissions and mortality Aim 3: Determine the relationship of place of residence, care fragmentation, SES and minority status to acute and long-term VA surgical health care utilization to inform VA resource allocation Hypothesis: Low SES, rurality, care fragmentation and minority status are associated with higher VA resource utilization Methodology: Quantitative analyses using traditional parametric and exploratory machine learning techniques performed on diverse datasets to develop predictive models of surgical outcomes using care fragmentation, rurality and social risk factors risk adjusted for medical comorbidities and applied to VA quality metrics. Implementation/Next Steps: Deployment of quality metric models using social risk factors and care fragmentation within the VA system. Adjusting resource allocation to account for social risk factors.