MYPAINPAL: A NEW-TAKE ON THE TRADITIONAL PENPAL WHO HELPS IN REPORTING AND MANAGING PEDIATRIC CHRONIC PAIN

DIGITAL MEDICINE & WEARABLE TECHNOLOGY

Author: Jonathan Hong

Status: Project Concept


Background
15-30% of school-age children suffer from pediatric chronic pain and current methods in assessing this pain to inform treatment are limited in their methodology [1][2]. Popular models based on 0-10 or no pain-worst pain scales have been shown to be inconsistent with younger patients, and are limited by only assessing pain severity, and often only in a clinic/hospital setting [3]. Written pain diaries are often used to assess the social, physiological, and emotional facets of pediatric chronic pain between follow-ups; however, its unstructured nature makes it difficult for patients to report consistently and thoroughly and for physicians to efficiently utilize and interpret [2]. Currently there exists no pediatric pain assessment and reporting model that addresses the multi-faceted nature of pain in a continuous, engaging, and educational manner outside of the clinical setting.

Methods
myPainPal is a mobile application that creates dialogue between the user and a simulated “pal” who also suffers from chronic pain as the intended user does and employs natural language processing (NLP) to analyze the discourse. The pal will be pre-programmed to periodically notify and report its pain episodes to the user. The report is intentionally designed to report the multiple dimensions of pain including intensity, location, duration, emotional factors, and other contextual information of interest to a care provider. Afterwards, the user will be prompted to select from a series of pain-relieving therapies to guide their pal through pain management. The user will be asked by the pal to report their pain to the pal whenever they experience pain so that they can “navigate and help each other through their chronic pain together.” When the user opens the application to report their pain, the pal will ask a series of guided questions regarding the above-mentioned dimensions of pain. The user’s vocal responses will then be transcribed via dictation. NLP, text analytics, and a specifically designed pediatric pain medical taxonomy are applied to the dictated text in order to extract keywords and themes signifying severity of pain, social factors, etc. These are then either quantified on a scale individual to the user, based on relative changes in comparison to previous reports, or categorized. The application then generates a concise report of the pain episodes that occurred between follow-ups for the healthcare provider.

Conclusion
myPainPal has vast applications in improving currently limited chronic pain assessment techniques in a range of diseases from juvenile arthritis to sickle cell disease to migraines. The application’s natural language processing and dialogue-based model educates the patient on how to report pain, empowers them to have more ownership over their pain-reporting, and allows the provider to gain more comprehensive and individualized insights into the often underreported symptoms and causes of pediatric pain.

1. Slover. R, Neuenkirchen, G., Olamikan, S., & Keng, S. Chronic Pediatric Pain. Advances in Pediatrics. 2010; 57(1), 141-162.
2. Manworren, R. & Stinson, J. Pediatric Pain Measurement, Assessment, and Evaluation. Seminars in Pediatric Neurology. 2016; 23(3), 189-200.
3. Avian, A., Messerer, B., Meissner, W., Sandner-Kiesling, A., Kammel, J., Labugger, M., Weinberg, A., & Berghold, A. Using a worst pain intensity measure in children and adolescents. Journal of Advanced Nursing. 2017; 73(8), 1873-1883.