Objective High rates of antibiotic prescribing are an increasing worldwide concern, in terms of increased antimicrobial resistance and associated burden to health services. However, there remains a lack of paediatric population-based pharmacoepidemiological research. The aim of the current study is to determine trends in systemic antibiotic prescribing in Irish children.

Design Retrospective analysis of routinely collected data.

Participants Data were obtained from the General Medical Services (GMS) claims database. This represents patients who are unable to pay for medical services and accounts for about one third of Irish children. Data were obtained from 2004 to 2008, for about 271 000 children per year aged ≤15.

Main Outcome Antibiotic prescribing behaviour in a paediatric population.

Results Prescribing rates of systemic antibiotics were compared across years, age (0–4, 5–11, 12–15) and gender and are presented with CI and significance values (using negative binomial regression analysis). In 2004, 631/1000 population (95% CI 628 to 634) received at least one antibiotic prescription compared to 578/1000 (95% CI 575 to 581) in 2008. However, the overall trend across the 5-year period did not show a significant change (IRR 0.99, 95% CI 0.96 to 1.0, p=0.23). No significant difference was observed between males (620/1000, 95% CI 618 to 622) and females (631/1000, 95% CI 629 to 633; IRR 1.02, 95% CI 0.97 to 1.06, p=0.46). However, prevalence rates significantly decreased with age. Children in the 0–4 age group were significantly more likely to receive an antibiotic prescription (827/1000, 95% CI 824 to 829), relative to the 5–11-year old (553/1000, 95% CI 552 to 556) and 12–15 years (489/1000, 95% CI 487 to 492), with all p<0.001. The net ingredient cost of antibiotics increased from €2.3 million in 2004 to €2.8 million in 2008. The five most commonly prescribed drugs were as follows: amoxicillin (298/1000, 95% CI 297 to 299), co-amoxiclav (296/1000, 95% CI 295 to 296), cefaclor (135/1000, 95% CI 134 to 136), clarithromycin (54/1000, 95% CI 53 to 54) and phenoxymethylpenicillin (53/1000, 95% CI 52 to 53). The overall rates observed here are significantly higher than those reported elsewhere in Europe. For example, the Netherlands (178/1000), Denmark (328/1000) and Scotland (142/1000) consistently report low antibiotic prescribing rates. The choice of agents also differed between countries.

Conclusions Prescribing rates remained stable over the 5-year period. Although the rates observed here are higher than European comparisons, these results should be considered in the context of the GMS population. Nevertheless, the results suggest the possible overuse of antibiotics within the GMS population and the potential benefit from interventions to reduce prescribing.