Researchers from OSF Healthcare and their partners have developed an artificial intelligence algorithm that can predict the upcoming week’s work of each cancer navigator for their existing patients. . They also created a second machine learning model to allocate new patients to navigators in specialties for better workload balance.
WHY IT IMPORTANT
Patient navigators can help health systems reduce time from diagnosis to treatment, but programs are often underfunded, resulting in heavy workloads.
OSF Healthcare, run by the Sisters of the Third Order of Saint Francis, has 10 acute care and five critical access hospitals, with 2,084 licensed beds in Illinois and Michigan. Peoria, Illinois health system cancer navigators provide education, advocacy, and support to cancer patients and help facilitate their journey of care.
Dr. Jonathan Handler, senior fellow for innovation at OSF Healthcare, said in the health system announcement that innovative approaches to employee retention – “our greatest asset”.
To achieve greater workload equity for their CPN team, OSF partnered with the nearby University of Illinois College of Medicine, the University of Illinois at Urbana-Champaign, and the Western University Feinberg School of Medicine. North in a retrospective study to develop and test a machine learning method. the algorithm’s ability to better perform random tasks and generate fairer CPN workloads within a specialization.
The researchers note that a navigator in one specialty may have more jobs than others in the same major at a given time, they say.
“CPNs do not transfer their existing patients to other CPNs when the workload becomes overwhelming,” the researchers wrote in their report published in JCO Clinical Cancer Informatics in May.
“They want to retain their patients for the laudable purpose of maintaining a consistent relationship between the patient and the CPN. Thus, the algorithm uses the only available leverage to balance the workload: new patient distribution.”
They used a three-year data set compiled from electronic health records, including demographics, cancer type, and prior healthcare usage, to assess prior workloads. Here are 13 specialized CPNs operating at the largest hospital of the health system. According to the report, the dataset contains 273,057 records, including 13,033 unique patients.
The researchers then built three supervised regression models, each built from one of the most popular and successful open source machine learning libraries. The third step is to develop a delivery model that can minimize the difference between those navigators in next week’s workload.
“Dozens of input features were used to make each week’s prediction for every patient,” they said.
“Our program seeks to maintain the relationship between the patient and the CPN, so the only consistent restriction that applies is allocation to ensure that the patient remains on the CPN originally assigned during the course of the study. the whole time they attended the conference.”
In addition to their retrospective simulation analysis, the researchers also compared the predictive information distribution with the random distribution and evaluated workload differences among navigators within the same oncology subspecialty. letters. They note that expected patient needs, navigator experience, and current workload are not considered in OSF’s current CPN workload decisions.
They say that the predictive information-based model achieved fairness on significantly larger workloads than the random distribution.
“To our knowledge, this work may represent the first description of an automated, algorithm-driven approach to CPN workload balancing,” the researchers said.
“Optimization has been applied to healthcare staff and patient allocations in other healthcare areas, but this is often applied to cases rather than individuals.”
The plan is to integrate the OSF Community Connect tool, a platform that automates workflows, and test its effectiveness before the planned opening of the OSF Cancer Institute in 2024, OSF announced last week. .
TREND TO BIGGER WOMAN
Globally, AI is being used or developed to address unprecedented levels of burnout in the healthcare workforce.
Software companies and healthcare IT developers use machine learning to solve a number of healthcare tasks, from transcribing audio or video, to addressing administrative and provisioning inefficiencies. insights into patients and patient populations – all aimed at improving patient outcomes and outcomes, while not overloading the healthcare workforce with performance professionals.
For example, UCLA Health System is using algorithms to make its nursing workload fairer.
Nurse informatics have developed a machine learning model that generates real-time workload scores for each paramedic, said Meg Furukawa and Stesha Selsky, nurse informatics at UCLA Health. dozen.
The nurse in charge uses the generated score that is visible to all nursing staff to aid in decision making, and they can adjust their workload or request additional staff as needed.
Before HIMSS23, they said Healthcare IT News that the ML model is based on existing patient chart information and other nursing documentation from electronic health records and other systems. Furukawa noted that their guiding principles were to create a tool that would not increase the administrative burden and would rely on data generated from existing workflows.
Furukawa and Selsky say that by working collaboratively, UCLA Health’s use of AI by nursing staff has helped achieve more equitable nursing resources and patient assignments.
“We had bedside nurses involved, certainly, from the beginning, we had nursing leadership, we had workload severity champions as a part of the project to really give us input and feedback, and help us develop the tool and validate the tool along the way,” explains Furukawa.
ON PROFILE
“Our cancer navigators are dedicated and their workloads can be overwhelming at times,” said OSF’s Handler. “They never want to swap patients, so they trade themselves off, working overtime and sacrificing their own health to help patients,” he said. We hope our system can address those workloads and improve their work-life balance.”
Andrea Fox is the senior editor of Healthcare IT News.
Email: [email protected]
Healthcare IT News is a publication of HIMSS Media.
