Bidwell Jonathan, Ji Lixuan, Ammar Nariman, Fort Daniel, Fort Lisa
Ochsner Health, New Orleans, LA.
Tulane University, New Orleans, LA.
AMIA Jt Summits Transl Sci Proc. 2023 Jun 16;2023:42-51. eCollection 2023.
We developed the Ochsner Emergency Department Overcrowding Scale (OEDOCS) to help us measure and respond to crowding among diverse-sized Emergency Departments (ED) within our network. Not satisfied with our current Emergency Department (ED) crowding score, we first surveyed our ED staff to report perceived crowding and then developed models to predict perceived crowding from our Electronic Health Record (EHR) data. Staff at two ED locations, one large and one small, were asked to report a perceived crowding level between 0-200 every four hours for over 3 months. In addition, we collected Electronic Health Record (EHR) data during the same period. Next, we investigated models for predicting perceived crowding. Linear regression performed the best with an RMSE of 41.77 and 41.98% RMSE improvement over our previous crowding score. We have made OEDOCS publicly available.
我们开发了奥施纳急诊科拥挤度量表（OEDOCS），以帮助我们衡量和应对网络内不同规模急诊科（ED）的拥挤情况。由于对我们当前的急诊科（ED）拥挤度评分不满意，我们首先对急诊科工作人员进行了调查，以报告他们感知到的拥挤情况，然后开发模型，根据我们的电子健康记录（EHR）数据预测感知到的拥挤情况。要求两个急诊科地点（一个大的和一个小的）的工作人员每四小时报告一次0至200之间的感知拥挤程度，持续3个多月。此外，我们在同一时期收集了电子健康记录（EHR）数据。接下来，我们研究了预测感知拥挤情况的模型。线性回归表现最佳，均方根误差（RMSE）为41.77，与我们之前的拥挤度评分相比，RMSE改善了41.98%。我们已将OEDOCS公开。
