Engagement is a necessary condition to test audit and feedback design features: results of a pragmatic, factorial, cluster-randomized trial with an embedded process evaluation

Table 4 Mean number of CNS-active medications prescribed across trial arms at 6 months post-intervention: pairwise comparisons

Comparison	Least square mean difference	95% CI for least square mean difference		p
Arm 2 vs. arm 1	− 0.007	− 0.056	0.041	0.761
Arm 3 vs. arm 1	− 0.001	− 0.047	0.045	0.964
Arm 4 vs. arm 1	− 0.030	− 0.080	0.020	0.232
Arm 3 vs. arm 2	0.006	− 0.038	0.051	0.779
Arm 4 vs. arm 2	− 0.023	− 0.072	0.026	0.355
Arm 4 vs. arm 3	− 0.029	− 0.076	0.017	0.217

ISSN: 1748-5908