The review included 34 trials. The number of included participants was unclear.
Decision aids significantly increased knowledge (ES 0.67, 95% CI 0.40 to 0.94; 18 comparisons) compared to control groups. When five studies (six comparisons) with a waiting list control were excluded, the effect size was 0.50 (95% CI 0.27 to 0.73). Decision aids significantly reduced anxiety compared to usual care (ES -0.30, 95% CI -0.53 to -0.08; three studies). No significant heterogeneity was observed.
In six of nine comparisons, patients randomised to decision aids had significantly less decisional conflict compared with the usual practice group. When data was pooled across five comparisons, significant heterogeneity was present and there were no significant differences between groups.
Across the three studies that measured knowledge, there were significant improvements in the decision aid group (ES 0.50, 95% CI 0.31 to 0.70). There were no significant differences between groups in for anxiety or decisional conflict. No significant heterogeneity was observed for these analyses.
Treatment preference/intention was measured in three comparisons and was significantly different in only one comparison. Three studies reported patients' role in decision making and this was increased in all three studies in the decision aid group. Satisfaction with decision making was significantly increased in three of four studies.
Comparison of different decision aids:
In studies that measured knowledge, there were no significant differences in screening studies. In prevention/treatment studies, more intensive decision aids had a significant effect on knowledge (ES 0.33, 95% CI 0.10 to 0.56). There were no differences between different types of decision aid for anxiety or decisional conflict.