2018 Palliative Care: Can an AI-Based App Help Manage Cancer-Related Pain?
A study of 112 patients with metastatic solid tumors found that the use of an artificial intelligence (AI)-based smartphone app reduced both the severity of patients’ reported pain and hospital admissions. After an 8-week period, patients who used the AI-powered app to monitor and address pain experienced a 20% reduction in the severity of pain and had nearly 70% lower risk of pain-related hospital admissions than patients not using the app. These findings will be presented by Kamdar et al at the upcoming 2018 Palliative and Supportive Care in Oncology Symposium (Abstract 76).
“There is a significant shortage of palliative care providers, which will only worsen in the future as our population ages,” said lead study author Mihir M. Kamdar, MD, Associate Director of the Division of Palliative Care and an interventional pain physician at Massachusetts General Hospital. “This is one of the reasons why technology solutions to help manage palliative care challenges, such as cancer pain, are so important.”
Research shows that up 70–90% of patients with advanced cancer experience pain due to the disease, directly affecting their quality of life. According to the researchers, this is one of the first mobile apps to utilize both patient-reported outcomes and AI clinical algorithms to significantly decrease pain and reduce overall inpatient hospitalizations in patients with cancer-related pain.
About the Study
The app, named ePAL, was designed and studied as part of a collaboration between Partners HealthCare Pivot Labs, the Massachusetts General Hospital Division of Palliative Care, and the Massachusetts General Hospital Cancer Center. Fifty-six patients in the study were assigned to use the ePAL app, while 56 were assigned to usual care. Patients using ePAL received alerts on their smartphones with daily pain management tips and were prompted to submit their pain levels three times a week. The AI in ePAL was able to distinguish urgent from nonurgent pain and provide appropriate patient-facing education in real time. If cancer pain was severe or worsening, the app connected patients to their clinicians for care.
Pain was reported on a scale of 0 to 10, where 10 is the worst pain imaginable. A drop of one point in self-reported pain was considered a significant reduction. If the pain level was severe, the app alerted a nurse who responded within an hour. If the pain was moderate, the app asked the patient about their pain and then electronically gave the patient tailored educational feedback, such as information on how to better manage bothersome medication side effects.
Questionnaires were given to all participants at the start, midpoint, and end of the 8-week trial to assess attitudes about each patient’s overall treatment and their general anxiety.
Results
All patients had similar pain scores at the start of the study. The average pain level of about 4 did not change from the beginning to the end of the trial for people receiving usual care.
Pain severity and negative attitudes toward cancer pain treatment decreased significantly for those using the app compared to controls. Anxiety scores increased for those using ePAL compared to controls. The researchers believe that simply asking about pain may induce anxiety in some people. On the other hand, people reporting pain via the app more than twice a week did not experience an increase in anxiety (the anxiety level was 6.6 at both the beginning and end of the trial).
Over 8 weeks, ePAL users had 40% fewer inpatient hospital admissions compared to controls (n = 15 vs n = 25, P = 0.048).
“It’s significant that patients who used the app had significantly fewer hospital admissions without an associated increase in outpatient clinical burden,” said Kamal Jethwani, MD, Senior Director of Pivot Labs at Partners Healthcare and a senior investigator on the study, in a statement. “These findings suggest that integrating innovations like mobile technology and AI could have a real impact on patient well-being, resource utilization, and cost of care.”
Next Steps
As a next step, the authors plan to develop an even more robust AI-telehealth platform and study it in other clinical settings.
“We’re especially interested to see if this type of novel technology can be helpful in areas where access to palliative care is limited,” said Dr. Kamdar. “Our hope is to use innovation and technology to extend the reach of palliative care to those who need it most.”
This study was funded by a grant from the McKesson Foundation.
Disclosure: Dr. Kamdar has stock and other ownership interests and a consulting or advisory role with Amorsa Therapeutics.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
