Predictive models of diabetes complications: protocol for a scoping review

Table 6 Review timeline

Stage of the review at this time*	Started	Completed
Preliminary searches	Yes	Yes
Piloting of the study selection process	Yes	Yes
Formal screening of search results against eligibility criteria**	Yes	No
Data extraction	Yes	No
Data analysis	No	No

*Started: June 21, 2018, at Quebec City, Canada; Anticipated completion: December 20, 2020
**The results from the June search of Ovid MEDLINE were screened first. When the search was updated in October 2018, the June search results were subtracted from the October search results, leaving 435 additional references to screen. Also in October 2018, an EMBASE search was performed (see, Appendix 2 Table 5). There was inevitable, incomplete overlap between the results from the Ovid MEDLINE search and the EMBASE search, which includes the MEDLINE database. Altogether, we retrieved 19,491 references from MEDLINE database. To obtain just the references unique to the EMBASE search, the Ovid MEDLINE search results were subtracted from the EMBASE results way of a multi-step de-duplication process using Thompson EndNote, as described in Bramer 2016. The results of this EMBASE search yielded an additional 8020 references. We screened 164 full-text articles from the two databases (128 retained from MEDLINE and 36 from EMBASE). Finally, in June 2020, a search update was done in the two databases and 2661 additional references were found and extracted. We did this update according to Cochrane’s recommendations, i.e., using the dates of entry (rather than the dates of publication)

ISSN: 2046-4053