A Cochrane review [Abstract] 1 included 45 studies: 14 RCTs, one controlled before and after study and 31 interrupted time series (ITS) studies. Almost all the included studies (44/45) compared the effectiveness of PEM to no intervention. One single study compared paper-based PEM to the same document delivered on CD-ROM.
Based on seven RCTs and 54 outcomes, the median absolute risk difference in categorical practice outcomes was 0.02 when PEMs were compared to no intervention (range from 0 to +0.11). Based on three RCTs and eight outcomes, the median improvement in standardised mean difference for continuous profession practice outcomes was 0.13 when PEMs were compared to no intervention (range from -0.16 to +0.36). Only two RCTs and two ITS studies reported patient outcomes. In addition, 54 outcomes were re-analyzed from 25 ITS studies, using time series regression and observed statistically significant improvement in level or in slope in 27 outcomes. From the ITS studies, improvements were calculated in professional practice outcomes across studies after PEM dissemination (standardised median change in level = 1.69).
Comment: The quality of evidence is downgraded by inconsistency (heterogeneity in interventions and outcomes) and study quality.
Primary/Secondary Keywords