Abstract:Purpose: COVID-19 infection may interact with patients' medical conditions or medications. The objective of this study was to identify potential signals of effect modification of adverse drug reactions by statistical reporting interactions with COVID-19 infection (SRIsCOVID-19) in a large spontaneous reporting database. Methods: Data from the US Food and Drug Administration Adverse Event Reporting System through the second quarter of 2020 were used. Three-dimensional disproportionality analyses were conducted to identify drug-event-event (DEE) combinations, for which 1 of the events was COVID-19 infection, that were disproportionately reported. Effect size was quantified by an interaction signal score (INTSS) when COVID-19 was coreported as an adverse event or an indication (INTSSCOVID-19). An SRICOVID-19 exists when the calculated INTSSCOVID-19 is >2. The analyses focused on pandemic-emergent SRIsCOVID-19. Screening for extreme duplication of cases was applied. To assess possible reporting artifacts during the early pandemic as an alternative explanation for pandemic-emergent SRICOVID-19, we repeated the analyses with an additional year of data to gauge temporal stability of our findings. Findings: When examining DEE interactions, 193 emergent SRIsCOVID-19 were identified, involving 44 drugs and 88 events, in addition to COVID-19 infection. Of the 44 drugs recorded, most were immunosuppressant or modulatory drugs, followed by antivirals. Seven drugs (eg, azithromycin) were identified in emergent SRIsCOVID-19 with preferred terms representing off-label use for prevention or treatment of COVID-19 infection. These drugs were in fact repurposed for COVID-19 treatment, supporting assay sensitivity of our procedure. Infections and infestations were the most frequently observed system organ class, followed by the general disorders and respiratory disorders. The psychiatric system organ class had only a few emergent SRIsCOVID-19 but contained the largest INTSSs. Less commonly reported manifestations of COVID-19 (e.g., skin events) were also identified. After excluding DEE combinations that were highly suggestive of extreme duplication, there remained a more robust set of emergent SRIsCOVID-19, which were supported by biological plausibility considerations. Our findings indicate a relative temporal stability, with >90% of SRIsCOVID-19 persisting after updating the analysis with an additional year of data. Implications: The signals identified in the analyses could be critical in refining our understanding of the causality of spontaneously reported adverse drug events and thus informing the ongoing care of patients with COVID-19. Our findings also underscore the importance of undetected report duplication as a distorting influence on disproportionality analysis.

Data Mining Pipeline for COVID-19 Vaccine Safety Analysis Using a Large Electronic Health Record

The Application Framework of Big Data Technology During the COVID-19 Pandemic in China.

A Signal Detection Method for Temporal Variation of Adverse Effect with Vaccine Adverse Event Reporting System Data

Data Mining in Large Frequency Tables With Ontology, with an Application to the Vaccine Adverse Event Reporting System

COVID Vaccine and Cardiovascular Risks - A Natural Language Analysis of Vaccine Adverse Event Reports.

A Computational Method to Quantitatively Measure Pediatric Drug Safety Using Electronic Medical Records

Using a Machine Learning Approach to Monitor COVID-19 Vaccine Adverse Events (VAE) from Twitter Data

A comprehensive analysis of the efficacy and safety of COVID-19 vaccines.

COVID Vaccine and Cardiovascular Signals: An Analysis of Vaccine Adverse Event Reports (Preprint)

Using Vaccine Safety Data to Demonstrate the Potential of Pooled Data Analysis

Comparison of Statistical Signal Detection Methods in Adverse Events Following Immunization — China, 2011–2015

Causal Inference for Estimation of Vaccine Effects from Time-to-Event Data

CVSARRP: A Framework to Predict the Risk of Adverse to Severe Adverse Reactions for 10855 Diseases after COVID-19 Vaccination

Identification of sex-associated network patterns in Vaccine-Adverse Event Association Network in VAERS

Analysis of COVID-19 vaccine-related adverse event reports following COVID-19 vaccine BNT162b2 mRNA-1273 and Ad.26.COV2.S based on the US vaccine adverse event reporting system

Unpacking adverse events and associations post COVID-19 vaccination: a deep dive into vaccine adverse event reporting system data.

Non-parametric Bayesian mixture model to study adverse events of COVID-19 vaccines

Profiling COVID-19 Vaccine Adverse Events by Statistical and Ontological Analysis of VAERS Case Reports

Background incidence rates of health outcomes of interest for COVID-19 vaccine safety monitoring in a US population: a claims database analysis

Potential Signals of COVID-19 as an Effect Modifier of Adverse Drug Reactions

Improved Apriori Method for Safety Signal Detection Using Post-Marketing Clinical Data