This study identified the main topics people discuss about multiple sclerosis (MS) on Twitter and the sentiment linked to each. We analysed English-language tweets using Latent Dirichlet Allocation for topic modelling and SentiStrength for sentiment analysis. Conversations fell into four categories, and those about the burden of the condition and related chronic conditions were the most negative.
What was the context of our study?
Social media could be a valuable source of support for people with MS. However, there was little evidence on which MS-related topics are discussed on social media, or the sentiment attached to them.
What did we do?
We extracted English-language tweets about MS and used Latent Dirichlet Allocation to identify the main topics. An iterative inductive process grouped these into recurring themes, and SentiStrength was used to analyse the sentiment of the tweets.
What did we find?
The topics fell into four categories:
- Related chronic conditions: Conditions that often accompany MS.
- Condition burden: The day-to-day impact of living with MS.
- Disease-modifying drugs: Treatments and their effects.
- Awareness-raising: Campaigns and advocacy.
Tweets about condition burden and related chronic conditions carried the most negative sentiment.
Why is this important?
Combining topic and sentiment analysis reveals which themes may represent the greatest burden for people affected by MS. This knowledge can help improve the support and therapeutic approaches offered to them.
Reference
Giunti G, Claes M, Dorronzoro Zubiete E, Rivera-Romero O, Gabarron E. Analysing Sentiment and Topics Related to Multiple Sclerosis on Twitter. Studies in Health Technology and Informatics. 2020;270:911–915. https://doi.org/10.3233/SHTI200294