Aims Rational prescribing is the therapeutically sound and cost-effective use of drugs (World Health Organisation, 2008). Criteria lists are well-established as research tools in rational prescribing for adults but only one tool has been developed for paediatric practice. The French ‘POPI’ criteria comprise 105 inappropriate prescriptions or omissions. Having revised these to align with UK clinical guidelines, our aim was to assess the usefulness of the modified POPI criteria by evaluating their relevance to UK paediatric patients.

Methods This study was a single centre prospective observational study of 400 paediatric patients (0–18 years) in a children’s emergency department and two paediatric wards. The only exclusion criterion was lack of parent/carer consent. Diagnoses, symptoms and prescriptions were recorded and checked against the modified POPI tool.

Results Patient age ranged from 3 days to 17 years. The median number of prescriptions per patient was 2.5 (range 0–26). 343 patients attended with at least one clinical indication in the tool. 255 were in the category of Pain or Fever; the next most frequent were Nausea, Vomiting or Gastro-oesophageal Reflux (n=123), Cough (n=77) and Diarrhoea (n=36). 29 cases had one or more inappropriate prescriptions or omissions; these related to Pain or Fever (n=25), Bronchiolitis (n=3), Nausea, Vomiting or Gastro-oesophageal Reflux (n=1), and Asthma (n=1) and comprised 20 omissions and 12 inappropriate prescriptions.

Conclusions The modified POPI criteria detected 32 inappropriate prescriptions or omissions. However, 7/21 clinical categories were not relevant to any patients in the study. Furthermore, a number of frequent presentations are absent from the criteria, including sepsis (n=36), viral-induced wheeze (n=35) and lower respiratory tract infections (n=26). In 27 cases, it was not possible to determine objectively whether or not an inappropriate prescription had occurred, highlighting complexity and subjectivity within the criteria. This study demonstrates the potential for a criteria list to act as a useful tool in studying rational prescribing for children. However, it also highlighted a number of limitations that must be resolved in order to develop an effective paediatric rational prescribing tool.