Integration of an NLP-based application to support medication management

描述（由申请人提供）：患者电子健康记录（EHR）上准确完整的药物列表对于防止处方和管理错误至关重要。有意义使用的第1阶段要求认证的EHR能够为用户提供执行药物核对的能力。然而，大多数以前的研究都发生在住院设置，而在门诊设置的药物协调是重要的和具有挑战性的。此外，临床记录包含也需要核对的关键药物信息。我们这项研究的目标是开发新的方法和系统，使用自然语言处理（NLP）和其他技术，以促进在门诊环境中的药物和解过程。我们的具体目标是：1）确定在药物协调过程中使用临床记录和基于NLP的系统的需求、用例、工作流程问题、障碍和促进因素; 2）设计通用系统架构和应用程序，该应用程序将NLP系统和基于Web的用户界面集成到现有药物协调系统中; 3）在两个初级保健诊所试点这项研究，并衡量所提出的方法和工具的利用率、可用性、性能和可行性;以及4）分发我们的方法和工具，并使其广泛提供给其他研究人员和医疗机构用于非商业用途。 公共卫生相关性：患者电子健康记录（EHR）上准确完整的药物列表对于防止处方和管理错误至关重要。在这项研究中，我们将开发新的方法和工具，使用自然语言处理和其他技术，以促进药物和解过程。我们将实施该系统，并在门诊环境中评估我们的方法。

DESCRIPTION (provided by applicant): An accurate and complete medication list on a patient's electronic health record (EHR) is critical to prevent prescribing and administration errors. Stage 1 of Meaningful Use requires certified EHRs to be capable of providing a user with the ability to perform medication reconciliation. However, most previous studies have taken place in the inpatient setting, while medication reconciliation in the outpatient setting is importnt and challenging. In addition, clinical notes contain critical medication information that also need to be reconciled. Our goal of this study is to develop novel methods and a system using natural language processing (NLP) and other technologies to facilitate the medication reconciliation process in the ambulatory setting. Our specific aims are to : 1) identify the requirements, use cases, work flow issues, barriers to and facilitators of using clinical notes and a NLP-based system in the medication reconciliation process; 2) design a generic system architecture and an application that integrates an NLP system and a web-based user interface within an existing medication reconciliation system; 3) pilot this study in two primary care clinics and measure the utilization, usability, performance and feasibility of the proposed methods and the tool; and 4) distribute our methods and the tool and to make them widely available to other researchers and healthcare institutions for non-commercial use. PUBLIC HEALTH RELEVANCE: An accurate and complete medication list on a patient's electronic health record (EHR) is critical to prevent prescribing and administration errors. In thi study, we will develop novel methods and a tool using natural language processing and other technologies to facilitate the medication reconciliation process. We will implement the system and evaluate our approach in the outpatient setting.