Satoko Hori: Utilizing Patient Voices from Social Media in Healthcare

2023/03/20

In recent years, evidence for the optimization of drug therapy has accumulated, making it possible to provide drug therapy tailored to individual patients. In this context, to achieve treatment optimization, it is essential for patients themselves to actively participate in treatment and receive care in accordance with those decisions (a concept known as adherence). It is important for healthcare professionals to accurately identify and address patients' symptoms, side effects, and treatment-related concerns and needs. However, it is known that patients do not communicate much of this information to healthcare professionals. The lack of medical communication and barriers to patients sharing information with healthcare professionals are major challenges hindering the optimization of treatment.

To address these challenges, we established the "Patient Salon" as a third place where people can learn and engage in dialogue about health and medical care, and we have held it monthly for 11 years. Many healthcare professionals want to understand patients' feelings outside the typical clinical relationship, and it has now become a place where both parties gather to understand each other's perspectives and engage in open dialogue on equal footing.

Patients are increasingly sharing their experiences with illness on social media (blogs, SNS, YouTube, etc.). These sources contain candid accounts of their thoughts on illness and treatment, as well as daily life struggles. While patients use social media as a source of medical information, it is difficult to find the specific information they need from the vast amount of data available. Therefore, our lab has aimed to develop and utilize models to identify medically valuable insights for patients from the massive amount of text they generate. Specifically, we are building models using natural language processing to extract information on drug side effects, treatment concerns, QOL, and practical knowledge from text posted by patients on social media and other platforms.

Furthermore, we regularly hold exchanges and discussions between patient groups and lab students. While receiving feedback from a patient perspective, we are exploring ways to use this model to help patients connect with appropriate medical care and social support more easily and at an early stage. To take treatment optimization a step further, it will be desirable to improve patient adherence and design new medication support systems that incorporate the patient's perspective.