Mild Adverse Events of Sputnik V Vaccine Extracted from Russian Language Telegram Posts via BERT Deep Learning Model

BackgroundThere is a limited amount of data on the COVID-19 vector vaccine Gam-COVID-Vac (Sputnik V) safety profile. Previous infodemiology studies showed that social media discourse could be analyzed to assess the most concerning adverse events (AE) caused by drugs.ObjectiveWe aimed to investigate mild AEs of Sputnik V based on a participatory trial conducted on Telegram in the Russian language. We compared AEs extracted from Telegram with other limited databases on Sputnik V and other COVID-19 vaccines. We explored symptom co-occurrence patterns and determined how counts of administered doses, age, gender, and sequence of shots could confound the reporting of AEs.Materials and MethodsWe collected a unique dataset consisting of 11,515 self-reported Sputnik V vaccine AEs posted on the Telegram group, and we utilized natural language processing methods to extract AEs. Specifically, we performed multi-label classifications using the deep neural language model BERT “DeepPavlov”, which we pre-trained on a Russian language corpus and applied to the Telegram messages. The resulting AUC score was 0.991. We chose symptom classes that represented the following AEs: fever, pain, chills, fatigue, nausea/vomiting, headache, insomnia, lymph node enlargement, erythema, pruritus, swelling, and diarrhea.ResultsThe results of the retrospective analysis showed that females reported more AEs than males (1.2-fold, Pβ = .05 per year, PThe Lancet. However, the AEs reported in Telegram posts were consistent (Pearson correlation r=.94, P=.02) with those reported in the Argentinian post-marketing AE registry.ConclusionAfter receiving the Sputnik V vaccination, Telegram users complained about pain (47%), fever (47%), fatigue (34%), and headache (25%). The results showed that the AE profile of Sputnik V was comparable with other COVID-19 vaccines. Examining the sentinel properties of participatory trials (which is subject to self-reporting biases) could still provide meaningful information about pharmaceutics, especially if only a limited amount of information on AEs is provided by producers.