Fluoroquinolones to treat uncomplicated acute cough in primary care: predictors for unjustified prescribing of antibiotics.

BACKGROUND Despite efforts to ensure more accurate prescribing of antibiotics for respiratory tract infections, inappropriate selection of antibiotic treatment remains a big issue. We tried to ascertain which factors best predict the nature of fluoroquinolone prescribing for acute cough in primary care. METHODS Random effects logistic regression models were applied to the baseline prescription data taken from a cluster-randomized controlled trial based on 104 general practitioners (GPs) and 2745 patients. RESULTS Significant predictors for the prescription of fluoroquinolones from both patient and GP data were identified. Predictors from a patient's perspective were the severity of illness {odds ratio (OR) 3.56 [95% confidence interval (CI) 2.45-5.19] P < 0.001}, the duration of illness before seeing the GP [OR 1.09 (95% CI 1.04-1.14) P < 0.020] and the individual patient's age [OR 1.01 (95% CI 1.00-1.01) P < 0.015]. Predictors from the GP's perspective were extent/lack of specific vocational training [OR 3.10 (95% CI 1.54-6.22) P < 0.001], status as a general internist [OR 2.00 (95% CI 1.10-3.70) P < 0.002], the physician's overall antibiotic prescription rate for acute cough [OR 1.02 (95% CI 1.01-1.04) P < 0.001], the duration of illness before contact with patient [OR 0.81 (95% CI 0.69-0.95) P < 0.010] and the severity of illness [OR 0.27 (95% CI 0.12-0.63) P < 0.002]. DISCUSSION Whether a fluoroquinolone is prescribed by a GP seems to be determined not only by the patient's characteristics but also by the GP's vocational training and overall antibiotic prescribing rate. As the prescription of fluoroquinolones for the treatment of acute coughing can rarely be justified, such prescriptions may serve as a quality indicator for antibiotic prescribing in primary care.