About the speaker: Agni Orfanoudaki is an Associate Professor of Operations Management at the Saïd
Business School of Oxford University, where she leads the Data-Driven Decisions Lab.
Alongside her role, Agni is a Management Studies Fellow at Exeter College and a visiting
scholar at the Harvard Kennedy School as a Harvard Data Science Initiative Fellow. Prior to
joining Oxford, Agni received a PhD in Operations Research from the Massachusetts
Institute of Technology. Her primary research interests lie at the intersection of optimization
and machine learning with applications to healthcare and insurance. She has collaborated
with numerous institutions, including a major medical society, two international reinsurance
companies, and more than eight hospitals in the US and Europe.


Abstract: Tackling hospital emergency department (ED) overcrowding is a paramount
challenge for healthcare systems. To combat this issue, an innovative approach is to identify
patients who can be served vertically (i.e., in a seated position) and route them to a dedicated
area termed the Vertical Processing Pathway (VPP). Successfully implementing this design
requires a clear understanding of which patients should be routed to the VPP and when.
Currently, the decision on how to leverage the VPP is conducted in an ad-hoc fashion. To
assist our partner hospital and other EDs in capturing the value of the VPP, we develop a
machine learning model that provides personalized risk scores predicting whether each
arriving patient will need an ED bed. We use the derived scores as input to a stochastic
patient flow model and analytically characterize the optimal VPP policy that minimizes
the length of stay. Employing simulation analyses, we identify the settings in which our proposed
VPP design is preferable in terms of operational performance to traditional ED flow
approaches, such as “fast track” or “physician in triage.” Finally, we derive an interpretable
VPP patient streaming protocol and conduct a before-and-after experiment where we
leverage empirical analyses to evaluate the impact of implementing it in practice. Our
findings demonstrate that our protocol is highly effective, leading to significant reductions in
patient length of stay without any adverse effect on quality of care measures. Our work
results in a VPP protocol that is generalizable to other EDs, offering operational
improvements without requiring additional resources.