Twenty-four studies were included: seven cluster RCTs, four RCTs, two non-randomised controlled trials, one randomised crossover trial, two interrupted time series studies, one controlled pre–post study, and seven uncontrolled pre–post studies. Duration of follow-up varied with a median of seven months (range: two to 72). Both sample sizes and study quality were highly variable.
When considering both primary and secondary outcomes, clinical decision support systems significantly improved practitioner performance in 15 out of 24 studies including: one of three studies that assessed impact of the display of costs, one of the two studies that assessed impact of the display of previous test results, six of the 10 studies that examined use of reminders, one of the two studies that used display of previous test results and two of the seven studies that assessed impact of the display of recommendations.
There were no significant differences between treatment groups in the four studies which assessed the impact of test cancellation or delay on potential adverse events.