Sentiment Analysis of Digital Conversations Related to Myasthenia Gravis by Race/Ethnicity
Nizar Souayah1, Caroline Brethenoux2, Alyssa DeLuca2, Jacqueline Pesa3, Louis Jackson3, Zia Choudhry3, Patrick Fuery2, Rosario Alvarez2, Laura González Quijano2, Alex Lorenzo2, Ashley Anderson4
1Department of Neurology, Rutgers University, New Jersey Medical School, Newark, NJ, 2Human Dot Plus, 3Janssen Scientific Affairs, 4Houston Methodist - Department of Neurology
Objective:

To describe sentiments, barriers and drivers related to myasthenia gravis (MG) through analyzing conversations by race/ethnicity.

Background:

MG is a rare, autoantibody neuromuscular disease characterized by muscle weakness and fatigue, which imposes profound patient burden. Digital conversations can provide unprompted insights into perceptions and highlight areas of greatest concern to MG patients. There is a lack of data focusing on race/ethnicity.

Design/Methods:

US-based public domain patient conversations focusing on MG and posted within topical sites, message boards, social networks, and blogs from August 2022-23 were mined.  Content contributors were patients self-identified as White, Black, Hispanic, or Asian within the conversations or on public profiles. Advanced search techniques and artificial intelligence powered algorithms were used to extract/organize data by topics. Natural language processing was conducted to identify sentiments, mindsets, and drivers/barriers towards treatment.

Results:

13,163 conversations were extracted and segmented by patient race/ethnicity.  Among 1,678 conversations by Black patients, 67% were negative led by “misdiagnosis problems” (23%) and "impact on life (23%). “Uncertain” mindsets were more frequent among Black and Hispanics than White and Asian patients (47% and 55% vs. 39% and 40%). Frequent themes related to “barriers to treatment” were lack of efficacy and side effects for all groups. "Level of relief", "duration", and "symptom relief" were the most frequent themes within the efficacy category for all groups. Cost/insurance was a more dominant theme for Black and Hispanic than Asian and White patients (15% and 18% vs. 4% and 5%).

Conclusions:

Conversations by race/ethnicity that may indicate relevant differences in the experiences of people living with MG. Black and Hispanic patients more frequently discuss misdiagnosis, which is often a challenge in the MG patient journey. Additionally, conversations highlighting cost/insurance issues may point to health disparities for persons of color. This data informs management strategies personalized to individual patients’ needs and priorities.

10.1212/WNL.0000000000205774