Deciding which patients in the emergency room are the most sick or injured is tricky business. The fear of passing over a critically ill patient is one of the things that keeps emergency room doctors up at night. To make triage decisions more objective and accurate, Scott Levin developed an electronic triage system.

Levin is an associate professor in the schools of medicine and engineering, and he is the director of the Systems Institute, an interdivisional center that employs a multidisciplinary approach to understand and re-engineer systems.

Using data about the patient, Levin’s system analyzes a patient’s risk of a critical event, like needing surgery or intensive care, and the likelihood of the patient being admitted to the hospital. Based on these results, it assigns a triage level, nicknamed a “Hopscore.”

Hospitals currently use the Emergency Severity Index to prioritize patients in the emergency room on a scale of 1 to 5, with 1 being the most sick and 5 being the least. The problem is, approximately half of all patients are assigned a level 3, which mixes a large group of sick and healthy people.

Hopscore uses the same scale from 1 to 5, but a computer algorithm assigns the triage level. “The hope is that we can identify patients who are at risk more quickly,” says Levin. “We also want people to move through the Emergency Department faster when they don’t need as much care.”

Last year, Levin conducted a retrospective study of the systems at The Johns Hopkins Hospital and Howard Country General Hospital. Now it’s being tested at these facilities and Levin is gathering  feedback from nurses to make the system better.

In just six months’ time, Hopscore could be live at Johns Hopkins facilities. After that, it could move to global partner hospitals and then to commercialization.