Data collection processes

Task a: Describe how IPD were requested, collected and managed, including any processes for querying and confirming data with investigators. If IPD were not sought from any eligible study, the reason for this should be stated (for each such study).

Text a: The EBCTCG Secretariat requests IPD from the principal investigator or the relevant research group for each eligible study for each review. This request includes the standard data format, describing each variable and suggesting coding scheme. However, the necessary data are accepted in any format. The IPD are processed by the Secretariat to check for internal consistency, missing data and to confirm that the randomization process appears to have been conducted appropriate. Queries are raised and, where possible, resolved by correspondence with the responsible researchers. The results to be used for each study are shared with the responsible researchers in advance of their inclusion in the published reports.

Task b: If applicable, describe how any studies for which IPD were not available were dealt with. This should include whether, how and what aggregate data were sought or extracted from study reports and publications (such as extracting data independently in duplicate) and any processes for obtaining and confirming these data with investigators.

Text b: The published EBCTCG meta-analyses are usually restricted to the IPD and are not supplemented by aggregate data from trials for which IPD are not available. This is partly because it is not possible to incorporate such data into all the analyses, which might lead to inconsistencies in the results. Details of the identified trials for which IPD were not available are provided and, in some circumstances, illustrative analyses might be performed (in particular for discussion at the meetings of the EBCTCG, or the Steering Committee) in which aggregate data from the trials without IPD would be included. These aggregate data would usually have been extracted from published reports of the relevant trials.