AI tool shows ‘clinically significant impact’ on completion of cancer screening
Click Here to Manage Email Alerts
Key takeaways:
- AI-driven navigation tool nearly doubled completion of colonoscopies among patients who did not keep their initial appointment.
- Tool to be evaluated in other screening programs.
Researchers at Montefiore Einstein Comprehensive Cancer Center have developed an AI-based patient navigation tool that may increase completion of colonoscopies among individuals who canceled or did not show for an initial appointment.
An analysis of the quality improvement initiative’s impact, scheduled for presentation at ASCO Annual Meeting, revealed a near doubling of completed colonoscopies among no-shows after engagement with the bilingual patient navigation tool named MyEleanor.
“Our AI virtual navigator demonstrated high feasibility through high levels of patient acceptance and engagement, with a clinically significant impact on patient reengagement with cancer screening,” Alyson Moadel, PhD, deputy director of community engagement and cancer health equity at Montefiore Einstein Comprehensive Cancer Center, said during a presentation.
The Montefiore Einstein patient population in the Bronx primarily comprised those from historically underserved communities (44% Hispanic, 39% Black). These same patients face the largest disparities in treatment outcomes because many do not receive routine screening procedures, which results in later-stage diagnosis.
The institution developed the automated care assistant to aid its patient navigation staff with follow-up on those who missed their appointments. Available in English and Spanish, MyEleanor provided patients with procedure reminders and offered live transfers to clinical staff for rescheduling, all while evaluating for patterns that identify barriers to access.
Study researchers evaluated the tool’s impact on colonoscopy completion rates, patient volume, factors serving as barriers to undergoing the procedure and any characteristics predictive of patient engagement with the AI-based tool.
Between April and December of 2023 Montefiore Einstein evaluated MyEleanor in 2,400 adults who did not keep their original colonoscopy appointment as part of their quality improvement national project.
Patient volume for colonoscopy increased by 36% during the study period.
Over half (57%) of patients engaged with MyEleanor; of these, 58% accepted the navigation tools’ live chat offer to reschedule appointments.
Researchers also observed a near-doubling of the colonoscopy completion rate among initial no-shows, from 10% to 19%.
Study participants identified transportation issues as the largest impediment to completing colonoscopy; other factors included mistrust of medical providers and fear of the findings. The results highlight “some of the psychosocial nuances related to patient engagement,” Moadel said.
Researchers next plan to assess the tool for patient satisfaction and feasibility in other cancer screening programs.
Perspective
Back to TopChristopher G. Cann, MD
Completing screening colonoscopies is essential for identification of precancerous or cancerous lesions, with the hope of preventing the development of colorectal cancer or detecting cancer at an early stage before it potentially spreads to other sites within the body.The rate of colon cancer screening is suboptimal, with less than two-thirds of the US population receiving up-to-date screening. Lower screening rates are seen among patients with lower levels of education, those who live in poverty, those who lack health insurance and people for whom English is a second language.
This study evaluated the innovative approach of using an artificial intelligence-based patient navigator to help improve disparities in colon cancer screening in an underserved patient population. The study showed use of an AI technology resulted in a nearly twofold increase in completed colonoscopies among those who initially missed their screening colonoscopy and helped to identify significant barriers to completing screening colonoscopies The study shows significant promise of how the medical community can use AI to identify barriers to care, optimize patient communication and improve patient outcomes. Further research is still needed on how to optimally integrate AI into medicine at large.
Perspective
Back to Top
Fumiko Chino, MD
Colonoscopy is both cancer screening and cancer prevention. Identification and removal of polyps can prevent colon cancer from forming over the next 10 years.
Compared with non-Hispanic white patients, both Black and Latinx Hispanic patients are diagnosed with colorectal cancer at younger ages and are more likely to be diagnosed with advanced -stage disease. This diagnosis gap contributes to disparate cancer outcomes, as cancers found later are harder to treat and harder to cure.
The conversational AI navigator was successful at engaging patients and essentially doubled the completion rate in this population of patients who had previously no-showed for colonoscopy.
This is a quality improvement project and not a randomized controlled study. The benefits will need to be continuously evaluated over time and rigorously tested in other populations.
Overall, this quality improvement initiative is a truly innovative means of increasing cancer screening. It leverages AI technology to optimize the outreach capacity to vulnerable populations. The work delivers on the promise of technology to deliver better, more efficient, more equitable cancer care. I anticipate that it will ultimately improve cancer outcomes.
Healio | Women in Oncology Peer Perspective Board member
Collapse
Moadel AB, et al. Abstract 100. Presented at: ASCO Annual Meeting; May 31-June 4, 2024; Chicago.
Click Here to Manage Email Alerts