King's College London

Research portal

Diarrhoea and preadmission antibiotic exposure in COVID-19: a retrospective cohort study of 1153 hospitalised patients

Research output: Contribution to journalReview articlepeer-review

Bilal Akhter Mateen, Sandip Samanta, Sebastian Tullie, Sarah O'Neill, Zillah Cargill, Gillian Kelly, Ewen Brennan, Mehul Patel, Mohammad Al-Agil, James Galloway, James Teo, Debbie L Shawcross, Alexandra J Kent, Bu'Hussain Hayee

Original languageEnglish
Article numbere000593
JournalBMJ open gastroenterology
Issue number1
Early online date6 Sep 2021
Accepted/In press15 Mar 2021
E-pub ahead of print6 Sep 2021
Published6 Sep 2021

Bibliographical note

Funding Information: those of the NHS, the NIHR or the Department of Health and Social Care. The funders who support the individual authors had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests JT received research support and funding from Innovate UK, Bristol Myers Squibb, and iRhythm Technologies, and holds shares

King's Authors


OBJECTIVE: The aims of this study were to describe community antibiotic prescribing patterns in individuals hospitalised with COVID-19, and to determine the association between experiencing diarrhoea, stratified by preadmission exposure to antibiotics, and mortality risk in this cohort.

DESIGN/METHODS: Retrospective study of the index presentations of 1153 adult patients with COVID-19, admitted between 1 March 2020 and 29 June 2020 in a South London NHS Trust. Data on patients' medical history (presence of diarrhoea, antibiotic use in the previous 14 days, comorbidities); demographics (age, ethnicity, and body mass index); and blood test results were extracted. Time to event modelling was used to determine the risk of mortality for patients with diarrhoea and/or exposure to antibiotics.

RESULTS: 19.2% of the cohort reported diarrhoea on presentation; these patients tended to be younger, and were less likely to have recent exposure to antibiotics (unadjusted OR 0.64, 95% CI 0.42 to 0.97). 19.1% of the cohort had a course of antibiotics in the 2 weeks preceding admission; this was associated with dementia (unadjusted OR 2.92, 95% CI 1.14 to 7.49). After adjusting for confounders, neither diarrhoea nor recent antibiotic exposure was associated with increased mortality risk. However, the absence of diarrhoea in the presence of recent antibiotic exposure was associated with a 30% increased risk of mortality.

CONCLUSION: Community antibiotic use in patients with COVID-19, prior to hospitalisation, is relatively common, and absence of diarrhoea in antibiotic-exposed patients may be associated with increased risk of mortality. However, it is unclear whether this represents a causal physiological relationship or residual confounding.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454