The association between amoxicillin exposure in infancy and the development of asthma in later childhood in the UK: A retrospective cohort study using Clinical Practice Research Datalink

Date of ISAC Approval: 
05/05/2020
Lay Summary: 
Amoxicillin is commonly prescribed in the paediatric population (children <18 years) for a variety of infections. Asthma is one of the most common childhood diseases worldwide, with approximately 5.4 million patients receiving treatment for asthma in the UK alone. Recently there has been growing interest surrounding the factors that might predispose a patient to developing asthma. Disruption of the normal gut bacteria through the use of antibiotics has been suggested as a factor that might have a role. This study will investigate whether there is an association between the use of amoxicillin during a child's first year of life and development of asthma in later childhood. It will examine differences in the risk of asthma depending on frequency of exposure to amoxicillin during infancy, and it will also determine the average time between the initial exposure to amoxicillin and diagnosis of asthma. Results of this study will add to scientific understanding of the role of the normal bacteria in diseases such as asthma.
Technical Summary: 
This retrospective cohort study will examine the risk of asthma amongst children exposed to amoxicillin between the ages 0 and 11 months ("infancy"), inclusive. The study will also assess changes in the risk of asthma depending on the frequency of exposure during infancy. A time-to-event analysis will be conducted amongst exposed patients who developed asthma between the ages zero to ten years (inclusive), to determine the expected time between initial exposure to amoxicillin (during infancy), and development of asthma in later childhood. All patients who were born between 01 January 2009 and 31 December 2009 will be included in the study. Antibiotic prescribing practices change over time, so including all children born in 2009 will provide the most up-to-date information while allowing ten years of follow-up time. Including all children born in this calendar year will ensure the study is adequately powered. The cohort will be examined to determine children who were and were not exposed to amoxicillin during infancy. All children will be followed up and censored at first recorded asthma code, their 11th birthday, or loss to follow-up (whichever is earliest). Categorical data will be presented in tables, continuous data will be described using appropriate summary statistics (mean/SD). Incidence risk of asthma amongst the exposed, the unexposed and the overall cohort will be calculated. Where possible this will be stratified by sex, age, and other relevant covariates. Logistic regression will compare risk of asthma in the exposed and the unexposed, whilst allowing for adjustment for confounding variables and/or effect modification.
Health Outcomes to be Measured: 
Implied or confirmed asthma (between ages 0-10 years, inclusive).
Collaborators: 

Miss Samantha Lane - Chief Investigator - Drug Safety Research Unit
Dr Abena Amoah - Collaborator - London School of Hygiene & Tropical Medicine (LSHTM)
Dr Debabrata Roy - Collaborator - Drug Safety Research Unit
Ms Estelle McLean - Collaborator - London School of Hygiene & Tropical Medicine (LSHTM)
Miss Samantha Lane - Corresponding Applicant - Drug Safety Research Unit