General practitioners' (GPs') prescriptions of antibiotics have shown large variations and may not always be rational.
We analysed GPs'prescriptions and use of microbiological diagnostics in Viborg County during a 6-month period in 1992 based on Danish Health Service data.
In a logistic regression model we tried to identify potential predictors for a high prescriber of antibiotics, i.e. the GPs with the highest number of prescriptions per patient (upper quartile).
Two categories were calculated for the predictor variables, dividing the distribution by the median value.
The most liberal GP wrote 15 times as many prescriptions for antibiotics per patient as the most restrictive GP.
A strong predictor for high prescribing of antibiotics was the number of prescriptions for other drugs per patient (odds ratio (OR) 12.3,95% Cl : 2.8-54.4] after adjustment for age and sex.
High use of throat swabs was a strong negative predictor of high prescribing of antibiotics (OR 0.2,95% Cl : 0.1-0.8) while high use of cultures (OR 2.4,95% Cl : 0.8-6.9) and of urinary susceptibility tests (OR 3.1,95% Cl : 1.1-9.3) were positive predictors.
The GP's general attitude to pharmacotherapy seems important for antimicrobial chemotherapy, and if use of antibiotics should be reduced, targeted strategies should be aimed at high prescribers.
Mots-clés Pascal : Médecin généraliste, Fréquence, Prescription médicale, Antibiotique, Distribution, Microbiologie, Diagnostic, Homme, Epidémiologie
Mots-clés Pascal anglais : General practitioner, Frequency, Medical prescription, Antibiotic, Distribution, Microbiology, Diagnosis, Human, Epidemiology
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 97-0523496
Code Inist : 002B02S02. Création : 13/02/1998.