An REA method was used to identify and synthesise existing evidence relating to the two overarching issues set out above. Following a pilot phase, searches were conducted of peer-reviewed journal databases and sources of grey literature, using an agreed set of search terms and inclusion criteria. The searches took place during December 2016. The resulting literature was prioritised, and over 200 of the most relevant and robust studies were subsequently coded, analysed and synthesised for this report. Their methodological rigour was measured using a quality assessment (QA) tool. The outcome of this process informed the relative weight that each study was given in the final report. Additional studies are referred to in several places to provide further context.