Operationalizing Machine Learning and Discrete Event Simulation Models to Improve Clinic Efficiency

运用机器学习和离散事件模拟模型来提高诊所效率

• 批准号: 10664923
• 负责人: Nicole Gray Weiskopf
• 金额: $ 32.73万
• 依托单位: OREGON HEALTH & SCIENCE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-08-01 至 2024-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10664923
• 关键词: Affect; Ambulatory Care Facilities; Appointment; Books; Childhood; Client satisfaction; Clinic; Clinical; Data; Data Set; Effectiveness; Electronic Health Record; Event; Face; Fast Healthcare Interoperability Resources; Feeling; Glaucoma; Length; Machine Learning; Maps; Methodology; Methods; Modeling; Ophthalmology; Outpatients; Patient Care; Patient Schedules; Patients; Physicians; Production; Provider; Quality of Care; Regulation; Reporting; Research; Resources; Schedule; Testing; Time; Visit; Wait Time; Work; application programming interface; barrier to care; burnout; data access; effectiveness evaluation; electronic health data; follow-up; improved; individual patient; machine learning model; models and simulation; novel; post implementation; predictive modeling; pressure; prospective; tool

PROJECT SUMMARY Physicians often report feeling pressured to see more patients to maintain revenue, while having less available time for patient care. Systematic data-driven methods for efficiently scheduling patients are important as physicians are pressured to see more and more patients. We propose that real time prediction models of patient visit lengths, the likelihood of missing appointments, and of patient wait times will help schedule patients more efficiently. Clinics will be able to safely overbook to avoid empty slots from missed appointments, have guidance for scheduling urgent add-on patients, and provide wait time estimates for patients when there are delays. We will develop methodologies for accessing data needed for these predictions in real time and propose that the integration of these models into workflows will improve scheduling accuracy, patient wait time, and patient satisfaction, while also increasing clinic volumes.

项目摘要 医生经常报告感到压力很大，以期看到更多的患者维持收入，而可用的可用性较少 患者护理的时间。有效安排患者的系统数据驱动方法很重要，因为 医生被迫看越来越多的患者。我们提出了实时预测模型的 患者访问长度，失踪约会的可能性以及患者等待时间将有助于安排 患者更有效。诊所将能够安全地预订，以避免因错过的约会而空空位， 有指导安排紧急附加患者的指导，并在那里为患者提供等待时间的估计 是延迟。我们将开发用于实时访问这些预测所需数据的方法论， 建议将这些模型集成到工作流程中将提高调度准确性，患者的等待时间， 和患者满意度，同时增加了临床体积。