Long-term effects of an educational seminar on antibiotic prescribing by GPs: a randomised controlled trial

Abstract

Background: High levels of outpatient antibiotic use remain observed in many European countries. Several studies have shown a strong relationship between antibiotic use and bacterial resistance.

Aim: To assess the long-term effect of a standardised educational seminar on antibiotic prescriptions by GPs.

Design and setting: Randomised controlled trial of 171 GPs (of 203 initially randomised) in France.

Method: GPs in the control group (n = 99) received no antibiotic prescription recommendation. Intervention group GPs (n = 72) attended an interactive seminar presenting evidence-based guidelines on antibiotic prescription for respiratory infections. The proportion of prescriptions containing an antibiotic in each group and related costs were compared to the baseline up to 30 months following the intervention. Data were obtained from the National Health Insurance System database.

Results: In the intervention group, 4-6 months after the intervention, there was a significant decrease in the proportion of prescriptions containing an antibiotic from 15.2 ± 5.4% to 12.3 ± 5.8% (-2.8% [95% CI = -3.8 to -1.9], P<0.001). By contrast, an increase was observed in controls from 15.3 ± 6.0 to 16.4 ± 6.7% (+1.1% [95% CI = +0.4 to +1.8], P<0.01), resulting in a between-group difference of 3.93% ([95% CI = 2.75 to 5.11], P<0.001). The between-group difference was maintained 30 months after intervention (1.99% [95% CI = 0.56 to 3.42], P<0.01). Persistence of the intervention effect over the entire study period was confirmed in a hierarchical multivariate analysis.

Conclusion: This randomised trial shows that a standardised and interactive educational seminar results in a long-term reduction in antibiotic prescribing and could justify a large-scale implementation of this intervention.

Figure 1

Design of the randomised controlled trial (EBM = evidence based medicine, R= randomisation).

Figure 2

Effects of an educational meeting on GP prescriptions. Absolute change from 2005 to 2007 compared to baseline in (a) proportion of prescriptions containing an antibiotic and related cost and (b) proportion of prescriptions containing a symptomatic drug and related cost. Data are expressed as mean ± SEM. Asterisks indicate the level of statistical significance for intra-group comparisons (compared to baseline) (*P<0.05, **P<0.01, ***P<0.001). P-values represent between-group differences (control group versus intervention).

Figure 3

Comparison of the effect of two educational strategies: evidence based medicine subgroup (evidence-based medicine subgroup) and evidence-based medicine plus problem-solving strategies subgroup. Absolute change from 2005 to 2007 compared to baseline in (a) Proportion of prescriptions containing an antibiotic and related cost, (b) Proportion of prescriptions containing a non-antibiotic symptomatic drug and related cost. Data are expressed as mean ± SEM. Asterisks indicate the level of statistical significance for intra-group comparisons (compared to baseline) (*P<0.05, **P<0.01, ***P<0.001). P-values represent between-group differences (control group versus intervention).