Objectives : The aim of this study was to assess the association between general practitioners' (GPs') characteristics and the volume of second-choice antibiotics for acute respiratory tract (RT) episodes by GPs. Methods : Morbidity and antibiotic prescription data originated from the Second Dutch National Survey of General Practice (DNSGP-2). GPs' characteristics, including professional activities and views on RT symptoms and antibiotics, were measured by a written questionnaire. Multiple regression was carried out to assess associations between possible determinants and volume of second-choice antibiotic prescriptions. Results : In ∼39% of acute RT episodes antibiotics were prescribed, with one-quarter being second-choice antibiotics, relatively more frequently in lower than in upper RT episodes: 30 versus 19%. GPs who were more frequently consulted by patients with RT episodes (β = 0.29; 95% CI 0.13-0.41), who labelled RT episodes more as diagnoses than as symptoms (β = 0.27; 95% CI 0.15-0.42), who less frequently used national GP guidelines (β = -0.17; 95% CI -0.31 to -0.03) and who were more inclined to prescribe new drugs (β = 0.26; 95% CI 0.13-0.40), prescribed more second-choice antibiotics. Conclusions : Given the growing number of prescriptions of second-choice antibiotics, it is important to implement professional guidelines in daily practice, while training in being reluctant to prescribe new drugs and being alert to the marketing activities of pharmaceutical companies should be started in the medical curriculum.