A group of clinicians, informaticists and AI specialists at Renton, Wash.-based Providence collaborated to create ProvARIA to organize inbox messages based on content and urgency.
Medical assistants and caregivers at Providence were grappling with a surge in electronic messages, which had tripled since the onset of the COVID-19 pandemic. In response, Providence sought to create an alternative to address this challenge, according to a Nov. 10 news release from Microsoft.
ProvARIA works by giving priority to emails from patients reporting severe or worrisome symptoms over a query about clinic hours. It uses a natural language processing engine to organize the messages and a tailored user interface integrated into its EHR system to facilitate the triage of categorized messages.
The natural language engine is built on OpenAI's GPT architecture and operates within Providence's Microsoft Azure cloud.
Providence has piloted the program at four clinics and was able to achieve a 35% improvement in turnaround time, the release said.
The new system introduced efficiencies, enabling clinics to manage messages with fewer medical assistants.
With the support of ProvARIA, Providence medical assistants now handle 5,000 messages daily, spanning 145 clinics and involving 650 providers.