Using Twitter Data Analysis to Understand the Perceptions, Beliefs, and Attitudes about Pharmacotherapy Used in Rheumatology: An Observational Study

Abstract

Twitter has become an important platform for disseminating information about rheumatology drugs by patients, health professionals, institutions, and other users. The aim of this study was to analyze tweets related to 16 drugs used in rheumatology, including their volume, content, and type of user (patients, patients’ relatives, health professionals, health institutions, pharmaceutical industry, general press, scientific journals and patients’ associations), and to detect inappropriate medical content. A total of 8829 original tweets were obtained, with a random sample of 25% of the total number of tweets for each drug (at least 100 tweets) analyzed. Methotrexate (MTX) accounted for a quarter of all tweets, and there were significant differences in the proportion of tweets issued according to the type of user. Patients and their relatives mainly tweeted about MTX, while professionals, institutions, and patient associations posted more about TNF inhibitors. In contrast, the pharmaceutical industry focused on IL-17 inhibitors. Medical content prevailed in all drugs except anti-CD20 and IL-1 inhibitors and the most discussed medical topic was efficacy, followed by posology and adverse effects. Inappropriate or fake content was found to be very low. In conclusion, the majority of the tweets were about MTX, which is a first-line treatment for several diseases. The distribution of medical content varied according to the type of user. In contrast to other studies, the amount of medically inappropriate content was very low.

From Web to RheumaLpack: Creating a Linguistic Corpus for Exploitation and Knowledge Discovery in Rheumatology

“…These last data sources have received special attention in recent years in the healthcare sector Nawaz et al [2017]. In fact, patients are prone to use social media websites for health-related purposes Studenic et al [2018], Taik et al [2024], Blackie et al [2023], and to express their feelings Wilson et al [2023], Abbasi-Perez et al [2023], creating what is known as patient-generated data. Finally, accredited health websites provide trustworthy health information which can complement the previous data sources for training specialised AI systems.…”