Changes in penicillin prescribing in general practice.

In the present study, we have analysed changes from 1979 to 1983 in the use of phenoxymethyl penicillin and broad spectrum penicillin in the treatment of sinusitis in general practice. The changes are related to the doctors' seniority and to their practice conditions. All GPs in five counties were invited to register all prescriptions of antimicrobials during two periods, from 19 to 23 March 1979 and from 21 to 25 March 1983. In 1979 and 1983, 574 and 625 GPs participated, constituting 67% and 66%, respectively, of all the GPs in the five counties. A total of 15,863 prescriptions of antimicrobials was registered. In all, 1,567 patients with sinusitis were included in the study. Log-linear analyses were performed in order to take into consideration covariation in the explanatory variables and changes in the composition of the material between 1979 and 1983. The probability that a patient with sinusitis was treated with phenoxymethyl penicillin was reduced between 1979 and 1983 (odds ratio = 0.84), while the probability for treatment with broad spectrum penicillin was increased (odds ratio = 1.18). The log-linear analyses revealed no relationship between the doctors' seniority, the practice form or localisation, and changes in use of phenoxymethyl penicillin or broad spectrum penicillin.