Analysis of online posts reveals impact of disease symptoms
Patients, caregivers concerned about efficacy of treatments
An artificial intelligence (AI)-powered analysis of more than 11,000 posts on social media or in online community groups found that myasthenia gravis (MG) patients use these platforms primarily to discuss the negative impact of their disease symptoms and their feelings about MG treatments. Most of the posts expressed negative emotions, according to a sentiment analysis.
Focus groups of MG patients and caregivers further revealed concerns about the efficacy of current and new treatments at preventing MG crises, potentially life-threatening complications of MG marked by severe breathing problems.
“Patients and care partners are often negatively impacted by MG symptoms and value convenient and fast-acting treatments that control symptoms with minimal side effects,” the researchers wrote. “A greater understanding of these preferences and perspectives may encourage more personalized treatment plans and improved patient experiences in MG,” they wrote.
Findings from the analysis were reported in the study, “Patient and care partner perspectives and preferences related to myasthenia gravis treatment: A qualitative study,” published in Health Science Reports.
High symptom, treatment burdens
MG is an autoimmune condition in which the immune system mistakenly attacks proteins that are required for proper nerve-muscle communication. It is characterized predominantly by symptoms of muscle weakness and fatigue that can be variable and unpredictable.
Treatment for MG typically focuses on controlling disease symptoms, either by suppressing the activity of the immune system or by improving nerve-muscle communication. However, because MG is highly variable, there’s no single treatment that’s best for all patients.
“Due to the high symptom and treatment burden in myasthenia gravis (MG), understanding patient and care partner perspectives and preferences is crucial,” the researchers wrote.
The researchers sought to understand the perspectives of MG patients through AI-driven analysis of social media and online community posts or through detailed analysis of structured interviews from focus groups.
In the AI-driven analyses, researchers identified 11,554 posts from key Inspire online communities and open social media platforms such as Reddit, Instagram, X (formerly known as Twitter), and Facebook. All posts were available publicly and originated from 8,321 MG patients or caregivers in the U.S.
Researchers used AI and machine learning to categorize posts based on their content and additional natural language processing, another AI technique, to analyze the sentiment, or emotional content, of the posts. Posts often contained content about more than one topic.
The most commonly discussed topics of the analyzed posts were MG symptoms (65% of posts), treatments (65%), and disease burden (58%). Other common topics included insurance coverage of MG treatments (31%), MG diagnosis (28%), and disease progression (20%).
Of the 7,563 posts discussing MG symptoms, the most commonly mentioned symptom was difficulty speaking, which was referenced in 22% of the posts. Other commonly mentioned symptoms included muscle weakness (19%), difficulty eating (15%), and difficulty breathing (15%).
The sentiment analysis revealed the majority of the posts expressed negative emotions. When broken out into five negative emotions — anger, fear, disgust, surprise, and sadness — fear emerged as the most frequent negative sentiment, being identified in 69% of the posts.
Symptoms, treatments invoke fear
Of the 9,904 posts discussing treatments, researchers found that 29% of the posts expressed negative sentiments. These negative sentiments were often related to adverse events (26%), unstable disease control (19%), and treatment inertia (8%).
“Patient sentiment toward MG symptoms was generally negative, often associated with feelings of fear,” the researchers wrote. “This sentiment may be linked to unresolved MG symptoms and the adverse impact symptom fluctuations can have on patients’ physical well-being and overall quality of life.”
More than half (57%) of the posts on corticosteroids, a common standard-of-care treatment for MG, expressed negative sentiments, mainly related to fear or side effects, such as weight gain.
Other forms of treatment, such as nonsteroidal immunosuppressants and thymectomy, were viewed negatively in a much smaller proportion of posts (8% for immunosuppressants and 16% for thymectomy).
Researchers also performed focus groups analyses to assess how MG patients and their care partners felt about MG treatments, including current and new ones. These analyses counted with the participation of 12 MG patients and three care partners.
When asked about the most important attributes of current or recent treatments, focus group participants emphasized the importance of avoiding or preventing MG crises. Controlling MG symptoms and reducing side effects from treatment were identified as other key attributes to take into account.
Similarly, when asked about new or emerging MG treatments, 87% of the participants emphasized they expected these therapies to reduce the likelihood of MG crises. A large proportion of respondents (80%) also noted they anticipated these newer medicines would be able to work for longer periods of time.
Treatment convenience was also an important factor, with 60% of respondents stating that treatments that were administered intravenously, or directly into the bloodstream, were the least preferred options.
When asked about the most important factors to consider about the next possible MG treatment, all respondents said they preferred a medication that had worked fast, and less frequent dosing was viewed as a key deciding factor.
“Full consideration of patient preferences may help better navigate the unpredictability of MG, and is an opportunity to improve patients’ overall well-being, enhance patient satisfaction, and optimize treatment decisions,” the researchers wrote.
About the Author
