Evidence synthesis involves combining data from two or more studies that address the same research question to arrive at the most accurate answer. The steps involved are:
- Identifying studies to include,
- Extracting information from those studies, and
- Combining the information from those studies to answer one or more research questions.
The BCIO provides a structured framework for identifying items of information to use in all three of these stages.
- Study identification: Create a list of available entities in the BCIO to use for a) searching and b) study selection. This should be done for all the relevant BCIO upper-level components: outcome behaviour, intervention content, intervention delivery, population, setting, and features of the methods.
- Information extraction: Add further entities to the list in 1. to create a combined list of all the information that needs to be extracted. Then, build a spreadsheet including the data for each study corresponding to each entity. We have created a data extraction template in Excel to get you started. This can be found here.
- Annotation: Annotate the documents using the items identified in the ontology. This can be done using software such as EPPI-Reviewer or you can use this template. Each annotation would normally consist of the ontology item identified in the paper and the text that was used from the paper to identify that entity. Manuals that were used in the HBCP for annotation using component ontologies of the BCIO are in the HBCP Resources folder on Google drive. An Excel file containing all the HBCP annotations of Behaviour Change Techniques is also available in that folder as is a file with the full set of annotations.
- Information combination: For each research question, construct an analysis plan showing precisely how each entity will be used to answer the question and what analyses will be undertaken with those entities.
