SBIR Phase I: System for Patient Risk Stratification through Electronic Health Record Analytics

此小型企业创新研究（SBIR）第一阶段项目的广泛影响/商业潜力侧重于使用分析和技术来提高医疗保健领域的效率和有效性。最近的立法变化正在推动医疗保健生态系统中的所有参与者承担更大的责任。这个第一阶段项目特别包括自动评估患者风险的技术，从而降低出院后再入院率。这个第一阶段的项目有潜力通过提供具有成本效益的再入院控制解决方案来应对新的立法压力，为供应商和付款人领域的广泛客户提供支持。就商业潜力而言，美国国家科学院医学研究所估计，可预防的医院再入院每年浪费了200亿美元的医疗保健支出。拟议的I期概念验证的患者风险分层以支持再入院控制的潜在市场约为1亿美元。未来，该研究项目将作为支持更广泛的人口健康分析的基础，其潜在市场每年超过5亿美元，并以每年24%的速度增长。拟议的项目旨在开发一个数据挖掘系统，从电子病历中获取和分析信息，以便在病人出院后对他们进行风险分层。利用联邦法规要求的互操作性标准，该系统将以与供应商无关的方式无缝地聚合来自多个电子病历系统的数据。定制的分析引擎将检测紧急模式并对每个患者进行推断。再入院的风险。如果成功，这项研究将验证端到端概念，并表明这种方法更广泛地适用于人口健康方面的一些最大挑战。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project focuses on using analytics and technology to drive greater efficiency and effectiveness in healthcare. Recent legislative changes are driving all players within the healthcare ecosystem toward greater accountability. This Phase I project specifically includes technologies to automatically assess patient risk and thereby reduce post-discharge readmissions rates. This Phase I project has the potential to support a broad range of customers across both the provider and the payer landscape, by providing cost-effective readmissions control solutions that respond to new legislative pressures. In terms of commercial potential, the Institute of Medicine of the National Academies has estimated that preventable hospital readmissions account for $20 billion/year in wasteful healthcare spending. The addressable market for the proposed Phase I proof-of-concept for patient risk stratification to support readmission control is approximately $100MM. In the future, this research project will serve as a foundation to support broader population health analytics, the addressable market for which exceeds $500MM/year and is growing at a rate of 24% annually.The proposed project aims to develop a data mining system to capture and analyze information from electronic medical records in order to risk-stratify patients after they have been discharged from hospital. Leveraging interoperability standards that are required by federal regulation, the system will seamlessly aggregate data from multiple electronic medical record systems in a vendor-agnostic manner. A custom analytics engine will detect emergent patterns and draw inferences about each patient?s risk of readmission. If successful, this research will validate the end-to-end concept and suggest the broader applicability of this approach to some of the greatest challenges in population health.