Hi Monika
Welcome to the RAMESES discussions.  I'll have a first go at answering some of your questions and then the others can jump in and add to it or correct me where I go wrong. 

Firstly, I'm curious about how you are conceptualising your study in realist terms.  You talked about "identifying the perceived characteristics of guidelines that affect uptake of recommendations" - are you conceptualising "perceived characteristics" as part of a mechanism (ie perceptions of characteristics influence responses/decision-making about uptake) or as part of the context for implementation?  What is it that constitutes an outcome? (A realist configuration is structured in terms of context-mechanism-outcome and realist research questions/topics broadly reflect this construct.) 

Secondly, you asked about including "techniques borrowed from other analysis techniques such as those from qualitative methods".  I've got a number of responses to this:

a) realist synthesis isn't a technique, it's a logic of enquiry within which one uses whatever analytic techniques are appropriate to the particular task, so there's no problem 'in principle' with importing techniques from other approaches - so long as they are consistent with, and applied in a way that is consistent with, a realist approach;

b) realist synthesis is sometimes described as being one of the school of qualitative approaches - in that it doesn't rely on statistical aggregation to achieve its findings in the same way as meta-analysis does - so there's no contradiction there; 

c) the realist construct of 'mechanisms' can be applied at different levels of abstraction, from very fine grained to quite abstract. You talk about wanting to 'interpret specific attributes of guideline recommendations that may facilitate uptake'. At that (finegrained) level, the key issue for a realist analysis would be finding out how people responded to those specific attributes of the guidelines - it's their response to the attribute and the decision-making that follows from it that constitutes the fine-grained mechanism.  So from that perspective, I'm not sure why you'd need any other analytic method.  On the other hand, I don't know what reciprocal translation analysis is so perhaps it would be very useful! 

Thirdly, you ask about theory saturation. I'm not clear enough about how you would use "a codebook of definitions" to identify whether there was anything new in the other 200 articles.  The particularly realist 'trick' I think lies in being able to a) identify mechanisms and b) identify the contexts in which particular mechanisms do and don't fire.  Unless your codebook of definitions was so long as to include all possible words to describe aspects of all mechanisms and all relevant features of context, I don't quite understand how it would help to determine whether you'd reached saturation.  

Finally you ask about whether the typology (framework of categories of attributes) and the codebook would be useful to others. I don't have a view on those particular resources because I don't work in that area or know what research is being done, but in principle, I'm in favour of tools and resources being made available in case someone else finds them useful.  On the other hand, I think realist findings in terms of CMO for guideline uptake could potentially be useful for research across a range of domains in which guidelines are used (widely diverse fields of practice), and potentially in relation to a range of other sorts of products that people want to see taken up (research and evaluation findings, for example!)   

Cheers
Gill