Experimental and quasi-experimental designs for evaluating guideline implementation strategies.

The choice of study design for guideline implementation studies will determine the confidence with which the observed effects can be attributed to the interventions under study. In general, cluster randomized trials, of which there are different types, provide the most robust design. However, the use of these designs has implications for the power, conduct and analysis of studies. Wherever possible, designs allowing head-to-head comparisons, which incorporate baseline measures of performance, should be used.