Back

A case-control study of autoimmune AEFIs following COVID-19 vaccination reported to VAERS

von Csefalvay, C.

2021-07-07 epidemiology
10.1101/2021.07.06.21260074 medRxiv
Show abstract

Autoimmune adverse effects following immunisation (AEFIs) are widely regarded as a chief concern driving vaccine hesitancy. This case-control study seeks to shed light on the true risk of autoimmune AEFIs associated with the COVID-19 vaccine through a case-control analysis of VAERS reports. Reports of autoimmune aetiology were matched with reports of non-autoimmune controls. Statistical analysis reveals that the safety profile of COVID-19 vaccines with regard to autoimmune AEFIs is highly favourable. In particular, neuroautoimmune AEFIs have statistically significant reporting odds ratios below unity (Guillain-Barre syndrome: 0.35, multiple sclerosis: 0.70, transverse myelitis: 0.79), indicating a reduced association of reports of these conditions with the COVID-19 vaccine versus other vaccines. Only three autoimmune aetiologies exceed a ROR of 2.0 and thus present a potential signal. Of these, myasthenia gravis (ROR = 3.90, p < 0.001, 95% CI: 2.63-5.80) may be the result of epidemiological confounding factors not sufficiently controlled by matching, as the population most likely to develop myasthenia gravis was strongly prioritised in the COVID-19 vaccines initial rollout. Immune thrombocytopaenia (ROR = 26.83, p < 0.001, 95% CI: 16.93-42.54) is a clear safety signal, confirming a large number of case reports and studies that indicate a risk of immune thrombocytopaenic events following the COVID-19 vaccine. The lone strong safety signal of immune thrombocytopaenia notwithstanding, this study attests to the safety of the COVID-19 vaccine where autoimmune conditions are concerned. Through quantifying the risk of autoimmune disorders associated with COVID-19 vaccination, this study contributes to a growing body of evidence supporting the safety of such vaccines.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.