Author’s response to reviews Title: Preliminary testing of the reliability and feasibility of SAGE: A system to measure and score engagement with, and use of research in health policies and programs Authors:

the implications section in the discussion, and the conclusion. We emphasise that the results point to the possibility that SAGE is a feasible tool to use in practical settings, may possess adequate reliability, and allow independent coders to score in a manner consistent with experts in this field, however, these conclusions are highly tentative, and further research with a larger sample is necessary. 6. The sample size is limited, which is acknowledged in the limitations. However, it is concerning that there were only 2 non-experts compared with the 6 experts. For the 2 nonexperts, were there any analyses conducted to understand their own intrarater reliability? We have made efforts to emphasise the tentative nature of conclusions due to the very small sample size, in the limitations, but also the conclusion and abstract. Unfortunately, because this was not the original aim of the study, we did not conduct analyses of the intra-rater reliability of the non-experts (which presumably would involve instructing the non-experts to repeatedly rate the same documents, and examine if they show consistency in how they score these same documents). 7. Minor comment: the non-experts were called 2 different things (research officers and nonexperts); the experts were called experts and 'knowledge translation experts'. It would seem that the non-experts were not non-experts in research, rather they were non-experts in knowledge translation. It is unclear whether the expert vs non-expert distinction is pertaining to knowledge translation or to use of research in health policies. There should be consistency of language and perhaps an alternative title for the research officers could be developed. This is correct, that the non-experts were in fact non-experts in knowledge translation. We no longer use the term non-experts or research officers, and uniformly refer to these participants as independent coders (see participants section on page 7). We define the independent coders as those that had relevant qualifications but were not researchers in the field of evidence-informed health policy.