WEDNESDAY, Aug. 5, 2026 (HealthDay News) -- A data-driven design can optimize a vertical processing pathway (VPP), identifying patients who can be served safely while seated in the emergency department (ED), according to a study published online July 9 in Management Science.
Arshya Feizi, Ph.D., from Harvard University in Boston, and colleagues developed a machine learning model that provides personalized risk scores predicting whether each arriving patient needs an ED bed. This scoring was used to develop a stochastic patient flow model and analytically characterize the optimal VPP policy. Pre- and post-VPP protocol implementation were compared.
The researchers found that the implemented protocol led to an 11-minute (4.2 percent) reduction in ED length of stay in a partner hospital. There were no adverse effects on quality-of-care outcomes. This statistically significant effect persisted after controlling for confounding factors and sources of endogeneity.
"The 11-minute per-visit reduction translates to substantial benefits for an individual ED: a medium-sized ED with 40,000 annual visits could recapture approximately 6,800 bed-hours annually, enabling treatment of nearly 2,000 additional patients and potentially generating $3 million in additional health care system reimbursements from payers for these visits," the authors write.
