Timothy Bates


Media is loading

Document Type


Publication Date

Summer 8-20-2021


COVID-19 (Disease) -- United Kingdom, Coronavirus, Computational modeling, Communicable diseases -- Epidemiology -- Mathematical models


In the early months of the COVID-19 pandemic, it was reported that some antibiotics were prescribed as a remedy for viral treatment and prophylaxis based on non-randomized, uncontrolled short clinical trials. A major antibiotic consulted being Azithromycin; a broad-spectrum macrolide selected based on its immunomodulatory effects in chronic inflammatory lung diseases, with a seasonal prescription increase of 21.5% in March 2020 compared to March 2019.

To analyze the effect and possible antimicrobial resistance impact of the pandemic on Azithromycin prescription across general practices in the United Kingdom (UK), this study uses a time series decomposition modeling method to compare a predictive forecast of an uninfluenced prescription period with the actual prescription dataset during the pandemic. Similar studies utilize critical observation of trend and dataset analysis using data visualization.

The comparison model shows significant prescription spike than expected in March and April due to the pandemic, with conclusion that initial excessive exposure of patients to antibiotics dispensation possible to cultivate resistant gut microbes in the UK population. The model also shows a significant decrease in Azithromycin prescription than expected late into the pandemic, possibly influenced by the numerous randomized clinical trial reports of antibiotics ineffectiveness in COVID-19 treatment and/or prophylaxis.

This study puts the essence of standardized clinical decisions in focus and why the government and scientific institutes need to ensure that only well studied experimental procedures are issued clearance.

Persistent Identifier

Oluwasegun- Podcast Transcript.docx (15 kB)
altREU Podcast Transcript

Oluwasegun- Abstract.docx (13 kB)
altREU Abstract