AI Support During Emergency Calls

What is it about?

The study demonstrates how artificial intelligence can support telephone nurses in handling emergency calls. These calls are often complex, and the clinical assessment and healthcare response rely on telenurses' competence. So-called symptom checkers—patient-facing apps—are increasingly used to bypass telenurses and automate the assessment process. However, symptom checkers cannot serve all patient groups, like people with low health literacy or digital literacy or with complex health conditions affecting their urgent health problems. Instead of trying to automate the clinical assessment, the researchers in this study have collaborated with telenurses to learn how they work and how artificial intelligence can aid them in serving all patient groups. The endeavour has resulted in a working prototype for an AI tool that suggests tailored questions based on how a call has progressed so far. A telenurse can select questions and confirm or deny symptoms based on a caller's account, leading to structured documentation that aids judgment during the call. The tool marks symptoms that indicate high urgency, alone or in combination. Telenurses and artificial intelligence collaborate during the assessment. However, telenurses remain in control and can leverage their expertise to respond to patients’ needs.

Featured Image

Photo by ThisisEngineering RAEng on Unsplash

Photo by ThisisEngineering RAEng on Unsplash

Why is it important?

Artificial intelligence has opened up new areas for automation. However, automating assessments of medical urgencies by symptom checkers negatively affects several patient groups. Alternative initiatives that do not aim for full automation but for supporting telenurses by artificial intelligence tend to focus on identifying single conditions, like stroke or blood poisoning. Such single-purpose initiatives have limited ramifications for telenurses' work. This study takes a different approach by trying to enhance the decision-making process through continuous nurse-AI collaboration. The intention is to leverage the combined strengths of humans and artificial intelligence, preparing the stage for humans and machines to learn from each other in the long run.

Perspectives