Mercy Launches AI-Driven Texting Program to Prevent Chemo-Related Hospitalizations

November 14, 2023

ST. LOUIS – Chemotherapy is a lifesaving treatment for many cancer patients, but the side effects can lead to hospitalizations. Named The Chen Chemotherapy Model after lead data scientist Jiajing Chen, the new AI-driven texting platform is proving valuable at keeping patients out of the hospital by raising red flags before symptoms become severe.

Chen served Mercy for nearly four years before she lost her own battle to cancer earlier this year. Prior to her diagnosis, Chen used her expertise in statistical modeling, machine learning algorithms and predictive analytics to develop the model that would later help others like her.

Mercy doctor's coat Jiajing Chen

“Jiajing always had a passion for medicine and for helping people,” said Andy Greenwood, Chen’s husband. “She was a medical student in China for a few years before coming to the U.S., then found herself in public health using her analytical and computer skills to help patients.”

The Chen Chemotherapy Model creates a risk score for non-leukemia chemotherapy patients over age 18. Continuously learning, the model predicts the likelihood of outpatient chemotherapy patients experiencing symptoms that may result in hospitalization within 30 days of their chemotherapy treatment. 

Once opted into the smart texting platform, patients receive a daily text for seven days, except weekends and holidays. The text includes a list of symptoms including diarrhea, pain, fever, fatigue, nausea/vomiting or no symptoms.

“Historically, we have not known which patients are having a problem until they presented to the ER or called the office, which was typically late in the day when we had limited options available,” said Dr. Jay Carlson, medical director of Mercy oncology service line. “The Chen Chemotherapy Model and smart texting allows us to proactively manage these patients and identify when they are having problems earlier in the journey. This means they may be able to be treated in the office, recover faster and feel better overall.”

When a patient selects a symptom, they are prompted to rate it and, based on their answers, the information may be escalated to their provider.

Greenwood said this project meant so much to Chen, whose short battle with a rare and aggressive form of cancer interrupted her work. “When she was in the hospital, she used all her strength to make sure her knowledge was passed along to her Mercy colleagues,” Greenwood said. “The project, named for Jiajing, will be a legacy I can share with our children, ages 1 and 4, to help them know their mother.”

Chen won Innovator of the Year at the 2020 ACCC National Oncology Conference for her work on the model, which has been improved with Epic workflow integration and texting technology. The new smart texting program using The Chen Chemotherapy Model is the first of many use cases made possible by the Mercy and Microsoft collaboration

More about Jiajing Chen, her contributions and life’s work
Chen held a Ph.D. in health outcomes research and a master’s in biostatistics and epidemiology from Saint Louis University. Her expertise included statistical modeling, machine learning algorithms and predictive analytics. She chaired a Mercy analytic work group involved in a comparative electronic health record study, known as real-world evidence, leading to Food and Drug Administration approval for a medical device label extension.

Prior to joining Mercy, Chen was a faculty member at the Saint Louis University Department of Health and Clinical Outcomes Research. Her research interests included health and economic outcomes research associated with diabetes, cardiovascular disease, cancer and organ transplants. She had extensive experience working on large national databases including private health insurance administrative data, SEER-Medicare data and the Organ Procurement and Transplantation Network.