The purpose of this investigation was to describe the impact of an early review of
antibiotic prescriptions in a hospital using unsolicited
infectious disease physician (
IDP) counselling, identify areas for improvement and examine factors associated with physicians' non-compliance. The prescriptions of 15 selected
antibiotics from surgical or medical wards were screened daily and reviewed between days 3 and 5 by a single
IDP during an 8-month period to identify those likely needing counselling. Improved
antibiotic use was sought by encouraging ward physicians to withdraw or de-escalate
therapy, promoting oral switch or reducing the
duration of therapy whenever appropriate. Variables potentially associated with
IDP counselling and physicians' non-compliance were tested using bivariate analysis and then entered in a logistic regression model. Among 857 prescriptions analysed, 54.6 % prompted unsolicited counselling, mostly for stopping
therapy (18.8 %), reducing its duration (18.0 %) or de-escalation (13.0 %). Variables independently associated with
IDP counselling included
antibiotic combination (adjusted odds ratio [aOR], 5.27 [95 % confidence interval (CI), 1.80-15.45]; p = 0.002), non-clinically documented
infection (aOR, 4.98 [95 % CI, 2.81-8.82]; p < 0.001) and microbiologically documented
infection (aOR, 2.04 [95 % CI, 1.51-2.75]; p < 0.001). The physicians' compliance rate was 77.3 %. Variables independently associated with physicians' non-compliance to the
IDP recommendation were the surgical speciality of the ward physician (aOR, 1.91 [95 % CI, 1.17-3.12]; p = 0.009) and advice to reduce the
duration of therapy (aOR, 1.88 [95 % CI, 1.12-3.15]; p = 0.017). An unsolicited post-prescription
antibiotic review can be successfully implemented with a high rate of physicians' compliance. Areas for targeting improvement measures include prescriptions in surgical wards and shortening the
duration of therapy.