To compare the author's prescribing habits for antibiotics in two different work settings and to study the reasons for the observed different trends.
A retrospective database study of computer-based patient records. The author's 55,261 patient contacts during a nine-year period were investigated.
A solo practice in a small isolated fishing village, Bolungarvík, in 1984-1987, and a group practice in the town of Akureyri in 1987-1992, where the people had a free choice of doctors.
The author himself was the subject of the study.
Type and number of prescriptions of antibiotics. The predictive variables were number and type of contacts, number of listed patients, age and sex of the patients, diagnosis, and the use of laboratory investigations.
From 1984 to 1986 the author reduced the number of his prescriptions of antibiotics from 82.9 to 63.8 per 1000 contacts (p