Antibiotic choice and duration associate with repeat prescriptions in infective asthma exacerbations

Publication date: Available online 28 August 2018

Source: The Journal of Allergy and Clinical Immunology: In Practice

Author(s): Marie Stolbrink, Laura J. Bonnett, John D. Blakey

Abstract

Background

Patients with asthma who present with lower respiratory tract infections (LRTI) often receive antibiotics. There is uncertainty about the need for and consequences of antibiotic administration.

Objective

Characterise the demographics of and antibiotic prescriptions for adult asthma patients with LRTI and investigate factors associated with repeat antibiotic courses.

Methods

We analysed prescriptions of antibiotics for LRTI in UK primary care from 2010 to 2015 using the Optimum Care Database. The primary outcome was a second antibiotic prescription for a LRTI code within 14 days of index prescription, as a proxy of initial treatment failure. A model for repeat prescriptions was derived using uni- and multivariable logistic regression analysis.

Results; We assessed 28,289 cases with complete datasets, 6.5 % of which received a second antibiotic course. Amoxicillin and clarithromycin respectively were used most commonly as index and second agents. Most frequent course length was 7 days for both index and repeat prescriptions.

Multivariable analysis demonstrated that age, index antibiotic and duration, smoking status, location, number of consultations and oral steroid courses in previous year were significantly associated with repeat prescription. The derived model predicted the binary outcome adequately (Cox-Snell R² 0.012; area under curve 0.62, 95 % CI 0.61-0.63). Co-morbidities, vaccinations, asthma treatment and exacerbation number were only significant in univariable analysis.

Conclusion; The current index prescribing preference of 7 days of amoxicillin correlated to fewer repeat courses. Baseline asthma treatment was not associated with risk of further prescriptions. Antibiotic administration in older patients with a smoking history could be a target for future studies.