A smartphone application utilizing elements of artificial intelligence was associated with improved cancer pain outcomes and a significant reduction in pain-related hospital admissions, according to data presented at the 2018 Palliative and Supportive Care in Oncology Symposium.¹

Results of the randomized controlled trial showed pain severity and negative attitudes toward cancer-pain treatment decreased significantly in patients randomly assigned to the app vs standard palliative care alone. Researchers also observed a significant reduction in the risk of pain-related inpatient admissions among patients enrolled in the intervention arm compared to controls.

We are at an inflection point in our field. We need to think about novel care delivery models to optimize care for patients dealing with cancer-related pain between clinic visits.

— Mihir M. Kamdar, MD

Tweet this quote

“To our knowledge, this is the first mobile app to utilize artificial intelligence and clinical algorithms to significantly decrease pain and reduce overall inpatient hospitalizations in patients with cancer-related pain,” said Mihir M. Kamdar, MD, Associate Director for Palliative Care and Director of the Cancer Center at Massachusetts General Hospital, Boston.

As Dr. Kamdar explained, cancer pain affects up to 90% of patients with an advanced malignancy, resulting in impaired quality of life and increased health-care utilization. In addition, there is a growing disparity between palliative care needs and clinicians available to provide it as the U.S. population ages.

“We are at an inflection point in our field,” said Dr. Kamdar. “We need to think about novel care delivery models to optimize care for patients dealing with cancer-related pain between clinic visits. Although Web- and mobile-based apps have demonstrated benefit, the challenge is to separate signal from noise so that we help patients while not overtaxing an already overburdened palliative care delivery model.”

Study Methodology

For this study, Dr. Kamdar and colleagues sought to determine whether artificial intelligence could be integrated into telemedicine to monitor and distinguish between urgent and nonurgent pain issues and intervene in real time to improve the trajectory of care. According to the researchers, the real-time interventions of the smartphone application, ePAL, were designed to empower patients in the self-management of their pain and selectively involve palliative care at pivotal moments of need, to keep patients from reaching crisis moments.

In addition to its function as a pain-monitoring and intervention tool, said Dr. Kamdar, the app had the ability to send patients daily, tailored educational coaching messages that were adaptive through the course of the study via artificial intelligence. Patients also had an educational library with content including videos of their own providers discussing common barriers they may be facing related to their pain control. In addition, patients had the ability to request nonurgent refills.

“One of the barriers to pain control that we have seen is patients running low on medications and calling at the last minute, so this was one of the features of the app,” said Dr. Kamdar, who noted that providers were also given a weekly summary of the different barriers that patients may be experiencing.

As Dr. Kamdar explained, the purpose of the trial was to determine ePAL’s impact on pain severity, attitudes toward cancer treatment, and health-care utilization in patients with cancer pain. The researchers recruited patients with metastatic, solid tumors from the Cancer Center at Massachusetts General Hospital. Patients with life expectancy less than 2 months, a known history of substance abuse, or significant psychiatric or cognitive comorbidities were excluded from the study.

Patients were then randomly assigned to either a control group (n = 56) that received usual care or an intervention group (n = 56) that used the ePAL app in addition to usual care for 8 weeks. The app assessed pain three times per week, and questionnaires about pain (Brief Pain Inventory), attitudes toward cancer treatment (Barriers Questionnaire II), and general anxiety disorder (GAD-7) were administered at 0, 4, and 8 weeks.

Significant Reduction in Pain Scores

As Dr. Kamdar reported, 39 patients in the intervention arm and 40 patients in the control group -completed the 8-week study, and no significant differences in baseline characteristics were observed. According to Dr. Kamdar, however, patients enrolled in the study were “younger than expected” and relatively well educated. Approximately 75% of patients in both groups had a college education, which is important to consider with this type of technology, said Dr. Kamdar.

Our hope is to build a more robust next generation of ePAL utilizing human-centered design research … to optimize palliative care delivery and enhance patient-provider connectedness.

— Mihir M. Kamdar, MD

Tweet this quote

Results of the study showed a significant decrease in both pain severity and negative attitudes toward cancer-pain treatment for patients using the app compared to controls. Pain severity scores decreased from approximately 4 to 2.99 for patients enrolled in the intervention, said Dr. Kamdar, whereas patients in the control group reported the same pain levels at the start and conclusion of the study. Dr. Kamdar also noted, however, that anxiety scores increased for those using the ePAL app compared to controls.

“An increase in anxiety has been noted in other telehealth interventions, and the increase in this study was mild, but I think this is an important consideration to address for future iterations,” said Dr. Kamdar. He added that patients who used the app more than two times a week did not see an increase in anxiety over time.

In addition, data demonstrated a significant reduction in pain--related admissions. Over the course of 8 weeks, 4 ePAL users had pain-related hospital admissions compared to 20 patients in the control arm. The per-patient risk of pain-related hospital admission was reduced by 69% with this intervention, said Dr. Kamdar.

“Our hope is to build a more robust next generation of ePAL utilizing human-centered design research and harness technology to optimize palliative care delivery and enhance patient-provider connectedness,” Dr. Kamdar concluded. “I’d also love to eventually expand to nonpain symptom management and even move to other illnesses.” ■

DISCLOSURE: This study abstract was sponsored by McKesson Foundation. Dr. Kamdar disclosed a relationship with Amorsa Therapeutics.

REFERENCE

1. Kamdar MM, Centi AJ, Fischer N, et al: A randomized controlled trial of a novel artificial-intelligence based smartphone application to optimize the management of cancer-related pain. 2018 Palliative and Supportive Care in Oncology Symposium. Abstract 76. Presented November 16, 2018.