Selection of Relevant Studies

A two-stage screening process consisting of a title and abstract scan and a full-text review will be used to determine the eligibility of articles. Both stages will follow the same process, where every article will be independently reviewed in pairs and the results will be documented on the spreadsheet. At the end of each round, the ratings will be compared and resolved by two reviewers or a third reviewer, when consensus is not achieved. Any ambiguities regarding the eligibility of a citation (or article) will be flagged and discussed.

The citations will be assessed for relevance based on a title and abstract scan. To be relevant for full-text review, articles must report on an existing ODL initiative for nurses and midwives in Kenya, Uganda, and Tanzania. The full-text review form will ask reviewers to assess each article using the following questions: Does the article describe/discuss the ODL initiative/intervention/program for nursing and midwifery education in Tanzania, Kenya, and Uganda?

This review is inclusive of all types of literature, thus including commentary articles, technical reports, case studies, and empirical studies employing all types of methodologies (ie, qualitative, quantitative, and mixed methods) and study designs. Viewpoint articles on how ODL education programs should be implemented outside of the context of an existing program will be excluded.

The criteria will be piloted by the reviewers to refine and establish a common understanding of the inclusion criteria. About 10% of the selected citations from a single database will be independently reviewed by four reviewers to establish interrater reliability (IRR). The results of the review will be compared, and the IRR will be calculated. The threshold for IRR is set at an average Cohen kappa of 0.70, indicating substantial agreement [31,32]. The pilot will be run again if the threshold is not met. If met, the remaining articles will be divided and assigned to two sets of pairs for independent review. These adjustments to the inclusion-exclusion process are appropriate, as they provide the team with opportunities to become familiar with the data and to reduce workload [32-34]. Regardless of the IRR outcome, a meeting about the process will be held to compare the results, resolve disagreements, and troubleshoot the challenges that arise during the title-abstract review process.