Drug use in non-hospitalized children.

The knowledge of the characteristics of the use of drugs in non-hospitalized children is of paramount importance since clinical trials are seldom (and sometimes should not be) performed in such patients. The source of drug prescriptions varies a lot: 'self-administration' through parents accounts for as much as formal physician's orders or pharmacist's counselling. Data can be collected through the whole drug chain: drug prescription--drug dispensing--drug attainment--drug consumption. Both qualitative and quantitative data are to be collected in order to draw a complete picture of drug use in non-hospitalized children. Longitudinal collection methods, as well as time-point or cross-sectional studies can be carried out. The particular role of each of these methods of performing a drug utilization study in children are analysed and discussed.