Table 6.

Summary of study results

Study (author, year)	Implementation strategy	Level 1: change in reaction/attitudes/beliefs	Level 2: learning	Level 3: behaviour
Randomised controlled trial
Beynon 2012 [46]	Basic 3-page policy brief	High-quality ratings Opinion about evidence strength or intervention effectiveness varies by health topic	–	Less likely to source other information and research related to the topic than control
	Basic 3-page policy brief plus an expert opinion piece	High-quality rating Opinion about evidence strength or intervention effectiveness varies by health topic. Increased intention to send policy brief to someone else and tell someone about key messages	–	Less likely to source other information and research related to the topic than control. Trend towards intentions persisting to actions. More likely to send policy brief to someone else
	Basic 3-page policy brief plus an un-credited expert opinion piece	High-quality rating Opinion about evidence strength or intervention effectiveness varies by health topic	–	Less likely to source other information and research related to the topic than control
Dobbins 2009 [9]	Tailored, targeted messages	–	–	Improved use of public health policies and programs compared to control
	Tailored, targeted messages plus a knowledge broker	–	–	Addition of knowledge broker potentially reduced use of public health policies and programs. However, improvements may have occurred in organisations with low research culture
Non-randomised controlled trial
Brownson 2007 [47]	Workshops, ongoing technical assistance, and digital resources	Change in whether heard of recommendations and attended training. Less likely to report state legislators were supportive of physical activity interventions. No change in other outcomes from baseline	All knowledge and skill measurements improved. Change larger for local than state health department decision-makers in every category except methods in understanding cost. The largest change related to attitudes	Improvement in self-reported individual adapted health behaviour change. No difference in other behaviour change outcomes