Validating community concerns of menstrual changes associated with COVID-19 vaccination using a self-controlled case series analysis of real-world data.

Résumé

OBJECTIVES: Stories of menstrual changes occurring post COVID-19 vaccination have abounded, with many affected persons expressing frustration their concerns were not being heard. In an era where misinformation is rampant and can fuel vaccine hesitancy it is imperative to address and validate community concerns. We aimed to investigate evidence of increased menstrual disturbances associated with COVID-19 vaccination.METHODS: We adopted a two-pronged approach; firstly, scrutinising social-media for discussions on menstrual changes associated with COVID-19 vaccination using our deep learning framework VaxPulse. Secondly, we analysed a large de-identified Australian general practice dataset to validate any evidence of increased menstrual disturbance presentations for females aged 15-49 years post-COVID-19 vaccination from 1 January 2021 to 28 March 2023, stratified by vaccine platform (adenovirus vector, mRNA, or protein-subunit). We used a self-controlled case series (SCCS) analysis to determine the relative incidence (RI) with 95 % confidence interval (CI) at six weeks post-vaccination and monitored thereafter until a return to baseline for minimum 2 consecutive weeks.RESULTS: Examining Reddit and Twitter (now X) data, we identified 70,977 posts discussing menstrual cycle irregularities with two prominent peaks since the global COVID-19 vaccine rollout. The SCCS analysis demonstrated increased presentations with a menstrual disturbance diagnosis associated with mRNA COVID-19 vaccination (RI: 1.14, 95 %CI: 1.07, 1.22, P