Hospitals conduct trials using voice assistants for patient follow-up calls
Policy & SafetyThere's An AI For That · 2h ago

Hospitals conduct trials using voice assistants for patient follow-up calls

A two-year medical trial is evaluating voice assistants designed to handle routine check-in calls for patients suffering from long-term illnesses. Linguists studied real clinical interactions to design conversational pacing and empathy patterns that help stretched health services maintain patient contacts.

University of YorkNHS

The Blend

Health researchers in the United Kingdom are testing an automated voice assistant designed to make phone check-ins with patients managing chronic conditions. According to the University of York, the two-year trial focuses initially on individuals with osteoporosis, a condition causing fragile bones that places heavy operational demands on specialized clinic staff.

To make the system sound less robotic, linguists studied real interactions between medical staff and patients to map out natural speech rhythms and supportive responses. Developed by technology firm Ufonia alongside academic researchers, the voice assistant carries out structured calls while safety experts evaluate whether it accurately captures patient information for clinical review. Participation is voluntary, and the automated calls serve as an extra layer of monitoring alongside standard healthcare appointments.

For everyday patients, automated follow-ups could mean more frequent monitoring without requiring extra clinic visits or putting more burden on overworked staff. If the trial proves successful, project leads hope to adapt the voice tool for other widespread conditions such as diabetes and heart failure, potentially scaling up routine check-ins across public health systems.

However, it remains uncertain how comfortable older or more vulnerable patients will feel discussing their health with a machine over time. While designing convincing conversational patterns is an important technical step, establishing long-term trust and ensuring the system never misses subtle signs of distress will be the true test before widespread adoption.

Written independently by AI News Smoothie from the reporting listed below. Facts belong to the original publishers. Follow the links for their full coverage.

Ingredients

Read the original