OBJECTIVES
Since April 1999, we have set-up an infectious disease consultation in the emergency unit of the University Hospital in Nice. Unjustified antiobiotherapy is often initiated. We therefore conducted a survey to study the motives and validity of antibiotic prescriptions.
METHOD
This prospective study was conducted in two phases. The first consisted in asking the emergency physicians prescribing antibiotics to fill-in a questionnaire giving information on the diagnosis established and the antiobiotherapy proposed. In the second phase, the diagnoses and corresponding treatments were submitted to 4 experts who assessed the acceptability of the diagnoses and the antibiotics prescribed. The experts only had access to the clinical and para-clinical data available. Moreover, their therapeutic judgement was based on previously published consensuses.
RESULTS
The 6-month survey collected 117 questionnaires that could be analysed. The rate of error in diagnosis was of 33% (39/117). Thoracic x-rays could not be interpreted in 11% of cases. In single variant analysis, factors of erroneous diagnosis were due to its interpretation by an internist, the diagnostic category of "broncho-pulmonary infections" and the lack of documentation. In multi variant analysis, only the lack of documentation was related to erroneous diagnosis (OR = 5.5; IC 95% (2.03; 15.30), p < 0.0002). The rate of antibiotherapy not adapted to the diagnosis made by the physician was of 32% (37/117). In 24 cases the modalities of the prescription were incorrect and in 13 cases the prescription was unjustified. Only the status of the prescriber (internist) was statistically associated with an antibiotherapy not adapted to the diagnosis (OR = 2.2; IC 95% (0.93; 5.26), p < 0.05).
CONCLUSION
Unjustified antibiotherapy in an emergency unit is generally due to erroneous diagnosis of infection. The lack of documentation and inexperience of the prescribers appear to be the two elements contributing to unjustified antibiotherapy.