Antibiotic prescriptions for upper respiratory tract infections in France and The Netherlands: determinants at the patient and GP level.

Within Europe, France and the Netherlands are extremes when it comes to the prescription of antibiotics: France has the highest volume in the European Union, the Netherlands the lowest. Antibiotic prescribing for upper respiratory tract infections (URTI) is not recommended in both countries. Non-rational prescribing antibiotics is problematic in terms of public health and health care resources. Determinants for antibiotics prescribing are not clear. In this paper we study what are determinants for antibiotic prescription in ‘highusing’ France and ‘low-using’ the Netherlands at both the patient and the GP level. Methods: For France, data were used from 778 GPs and 185 383 patients who had at least one URTI consultation; for the Netherlands data were available on 123 GPs with 23 828 patients with at least one URTI consultation. Data on consultations, prescriptions and patients were available from routine databases in both countries for 2001. In both countries information on GP determinants were collected in an additional written survey. The same binary multilevel analysis was performed with two levels of inference: GP and patient. The outcome variable was having gotten an antibiotic prescription (0 = no; 1 = yes). Results: Both in France and the Netherlands the probability of receiving an antibiotic prescription is largely explained by patient influence (about 70%). However, GP characteristics also clearly influence the chance that a patient receives antibiotics. In both countries patients are more likely to get an antibiotic if they have more severe complaints. However, the results for socioeconomic variables differ between the countries. For example, while in France women and non-employed are less likely to be treated with antibiotics, in the Netherlands women and nonemployed are more likely to be treated. At the GP level an important finding is that both in France and the Netherlands the number of recent visits from pharmaceutical sales representatives is associated with a higher probability of antibiotic prescriptions. Conclusions: Although France and the Netherlands have considerable different levels of antibiotic prescriptions, considerable differences in level of antibiotic prescriptions can be found between patient groups and between GPs in both countries. The influence of the pharmaceutical industry explains differences in antibiotic prescriptions between GPs and their patients on a similar way in France and Netherlands. (aut. ref.