The European Centre for Disease Prevention and Control (ECDC) has marked 18 November 2008 as the first of the proposed annual European Union (EU) Antibiotic Awareness Days in an attempt to improve the use of antibiotics by families and professionals.

Antibiotics are the most common medicines given to children. In the UK, there are around 6 million antibiotic prescriptions for children each year, the majority given unnecessarily for viral upper respiratory tract infections.1 There is clear evidence linking antibiotic resistance to prescribing, with a possible threshold effect beyond which high prescribing can select resistant isolates.2 Factors such as duration of treatment, dose and class of selected antibiotic3 are important in the resistance chain, along with host susceptibility and pathogen transmissibility.4 Children are an excellent environment for the selection of resistant bacterial pathogens after recent antibiotic use, particularly in day care. As recent studies have suggested, the individual child can host resistant pathogens after antibiotic use for up to 3 months, which is sufficient to sustain a high level of antibiotic resistance in the general population.5 If the amount of antibiotics prescribed in children could be reduced, selection and transmission of resistant strains should occur less often, with an effect on the general population. Is Europe responding to this challenge?

Surveillance of antimicrobial consumption in European children is currently surprisingly poor and not systematic. European organisations such as ESAC (European Surveillance of Antimicrobial Consumption) and EARSS (European Antimicrobial Resistance Surveillance System) collect some excellent data on antibiotic consumption and resistance in the adult population of Europe, but provide very limited paediatric country-specific and age-specific data. In adults, there is very significant variation in antibiotic use across Europe. In a study recently published by ESAC, total outpatient antibiotic use in 2003 varied by a factor of 3 between the country with the highest (31.4 daily defined dose (DDD)/1000 population in Greece) and the country with the lowest (9.8 DDD/1000 population in the Netherlands) use.6 In a retrospective analysis that we have undertaken using the IMS Health database, we also recorded significant differences in community paediatric prescribing among five European countries between 2002 and 2005.7 In this study, Spain had an average prescribing rate of 1581 prescriptions/1000 children, and UK had the lowest rate with 608 prescriptions/1000 children. Similar results were reported in a review of published surveys of antibiotic prescribing in the outpatient paediatric population in Europe and North America between 2000 and 2005.8 This study showed a severalfold difference between the lowest-prescribing European country, the Netherlands (200–400 prescriptions/1000 children/year), and a high-prescribing country such as Italy (900–1300 prescriptions/1000 children/year).

Antibiotic class-specific trends also vary among the European countries. In general, low-prescribing countries tend to use narrower-spectrum antibiotics compared with high-prescribing countries. In a recently published study, it was found that children in Denmark not only receive low rates of antibiotics, but penicillin prescribing is increasing, while the use of macrolides is decreasing.9 Spanish children, on the other hand, receive larger volumes of broad-spectrum antibiotics, with oral co-amoxiclav, cephalosporins and macrolides being the most common antibiotics in order of frequency.10

Centralised paediatric resistance data are also not available in Europe, but several studies have noted the strong relationship between antibiotic overuse and resistance. In a study published on behalf of EARSS, southern European countries with high prescribing rates showed the highest levels of Pneumococcus not susceptible to penicillin (up to 30%), and the same applied for erythromycin with a north–south gradient.11 Similar results were obtained in another study comparing resistance patterns of common bacterial isolates in paediatric patients in France, Germany, Italy and North America.12 Again resistance was higher in countries with high prescribing rates (France, Italy and USA).

So how is the UK doing? Reasonably well overall, but with worrying recent trends is probably the best summary. In the UK, paediatric community antibiotic prescribing declined by over a third in the late 1990s and early 2000s following an international trend, but has now increased again by almost 10% since 2003. Most antibiotic prescriptions are given to children with non-specific upper respiratory tract infections or for fever with non-specific diagnoses. Narrow-spectrum antibiotics such as amoxicillin and penicillin make up >75% of total antibiotic use (unpublished data), which is reflected in the low resistance rates for key pathogens such as Pneumococcus in the UK.13 14 Unsurprisingly, clinicians also seem to have developed guideline fatigue. There is now good evidence that, for example, otitis media serial guidelines aimed at reducing prescribing appear sadly to have had virtually no effect on prescribing rates.15

We now have the most ambitious prescribing guidelines yet, from the National Institute for Health and Clinical Excellence (NICE), for the management of upper respiratory tract infections in children and adolescents. The recommendation is for a no or delayed antibiotic prescribing policy for five common conditions (acute otitis media, acute sore throat, common cold, acute rhinosinusitis and acute cough).16 The guidance provides strong evidence that such a policy is safe and should be effective in reducing prescribing—in children by about a third, potentially reversing the upward trend and maybe bringing us down to near Dutch levels. Concerns have been raised that reduced antibiotic prescribing might be associated with increased rates of suppurative bacterial infection, for example mastoiditis or empyema.17 The other problem is that, as guidelines not to prescribe for these specific diagnoses become ever clearer, doctors are now switching to non-specific diagnoses to justify maintaining or even increasing prescribing rates for children. Guidelines alone will never be enough to change such entrenched behaviour patterns as prescribing antibiotics for an unwell child. Doctors prescribe antibiotics when they feel it is necessary to do so regardless of guidelines, cost–benefit analysis or risk to drive resistance. Clinical experience and family-based care is very important in this decision-making process. It is of great concern that increasingly children with minor infections are being seen by junior paediatric staff in the evening A&E rush rather than by experienced general practitioners.

In the UK, the Department of Health Expert Advisory Committee on Antimicrobial Resistance and Healthcare Associated Infection (ARHAI) can help by encouraging the development of more formal systems to both monitor hospital and community prescribing as performance indicators, but also to monitor the safety and clinical outcomes of the no-prescribing option. We must be able to convince families, paediatricians and general practitioners that it is both appropriate and safe for their child not to receive an antibiotic. This requires investment in monitoring systems to show that low prescribing rates are safe and best.

There is no doubt that achieving low prescribing rates across Europe is a massive task, which emphasises the complexity that surrounds antibiotic prescribing. Social, economic and medical factors are key in this process. The role of society and the pressure exhibited from parents/carers to prescribe antibiotics is well documented. France and Germany, although similar in terms of demographics and health systems, exhibit very different antibiotic use rates and pathogen sensitivities, which is a reflection of the way the community perceive the use of medicines.18 The significance of racial and ethnic groups has also been documented in studies from the USA, with a strong correlation between parental expectation to receive antibiotics and physicians’ perception of this expectation, which will inevitably lead to a prescription.19 Not uncommonly though, paediatricians overestimate the expectation of the parent/carer to leave with a prescription, leading to a specific diagnosis being made secondary to the decision to prescribe.20

In Europe, we need to develop a centralised surveillance network, which will document antibiotic use and resistance patterns in children at the community and hospital level. Although guidelines are only a small part of the solution, they can help. There is a striking absence of national prescribing guidelines for many EU countries, and unified evidence-based treatment recommendations endorsed by the appropriate European expert societies may be useful. Academically, more work needs to be carried out on identifying what are optimum prescribing rates, particularly with the introduction of the conjugate pneumococcal vaccine across Europe.

Finally, we have to raise public awareness of the risks and benefits of antibiotic use, and the EU antibiotic day is an excellent start.21 Patients and parents are increasingly involved in decision making, and the benefit of self-care in “common” infections has to be emphasised. Experience from some European countries has shown that public education really can work,22 although long-term studies have not been conducted. Any effective campaign has to involve modern communication tools that can reach the entire community in a faster and more efficient way.

There is a long way to go to identify optimal prescribing, monitor efficiently for any evidence of harm, accurately identify high-risk groups of children, and engage the public in prudent antibiotic use. The UK can lead the way in this process and, through international collaborations, assist in identifying and harmonising best practice in the management of paediatric infections across Europe.