4 key steps toward adopting AI-aided triage at scale
The COVID crisis has occasioned wildly irregular spikes and dips in patient demand for immediate attention all over the world. In response, Harvard researchers at Mass General Brigham have built an AI chatbot that can automatically triage patients whenever they call or show up in disaster-level numbers.
Team members offer the details of their imitable project in the December edition of Healthcare.
Lead author Lucinda Lai, MD, senior author Haipeng (Mark) Zhang, DO, and colleagues started by consulting with leadership at Providence St. Joseph Health in Washington state. As the first COVID-19 hotspot in the U.S., that health system had gained deep experience creating an AI-based online screening and triage tool that could flag patients most in need of speedy care.
Providence screened over 40,000 patients in its first week, the MGB authors note, underscoring Providence’s subsequent efficiency delivering care “on a scale that is otherwise impossible to achieve using traditional clinician-dependent triage pathways.”
Along the way to adopting a similar AI-based solution for their own patient populations, the MGB researchers took a number of steps that have proven key to their own success. The first four of these:
- Using a mobile-responsive, web-based interface to present users with a series of questions about risk factors and symptoms based on CDC guidelines. To these the MGB team added input from MGB infectious-disease experts. The idea was to capture initial risk-assessment findings so as to send the patient to the most appropriate next step.
- Feeding a decision algorithm the answers to those questions so as to arrive at a disposition endpoint. “Complexities of subsequent triage and management decisions (gender, age, pre-existing comorbidities) were deliberately handled by a COVID-19 expert clinician if they were directed to this endpoint according to the initial Chatbot algorithm,” Lai and colleagues write. “The screener was also provided in Spanish, as many of our sickest patients originated from primarily Spanish-speaking populations.”
- Deploying the resulting chatbot via an interactive voice-responsive message. People calling a COVID-19 hotline received this message while waiting on hold. It instructed them to visit the MGB website hosting the chatbot, which supplied COVID information and guided the less ill to conservative self-care measures while stepping the sicker patients through to a live nurse.
- Designing their AI triage tool to screen “enormous numbers of people and differentiate the ‘worried well’ from those warranting additional evaluation,” the MGB authors write. “There was an increased utilization during the period of March–April 2020 reflecting the overall success of the chatbot implemented. Patients needing additional assessment were directed to the hotline clinician to determine where to direct the patient from options in menu of pathway options.
“While the clinician was a critical part of the process, the AI triage tool decreased the burden of work and [now] serves as an exemplar for the integration of automated technologies into human workflows,” the authors write.
Reflecting on how far they’ve come and how fast, Lai et al. point out that AI tools, many as simple as theirs, are everywhere in e-commerce.
However, the implementation of this type of AI at Mass General Brigham and other early hospital adopters “represents a radical change in operational strategy for healthcare.”
“Through AI, patients are able to access prompt, evidence-based advice and direction to the most appropriate care setting,” Lai and colleagues conclude. “This minimizes missed opportunities in an overburdened system. Digital strategy highlights the true humanity of the system by providing the greatest care for the greatest number of people to the greatest of our capabilities.”
The study is available in full for free.
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.