Improving Transparency, Reproducibility, and Ethical Considerations for Evidence Generated in Fragile Settings of Cameroon Using the TREE Approach

Article type
Authors
Okwen P1, Sherifa-Jamil N1
1Effective Basic Services (eBASE) Africa, Bamenda, North West Region, Cameroon
Abstract
"Background
Ensuring ethical considerations including protecting research participants and research team is important for research evidence to be fairly generated and used. In fragile settings (including conflict, pandemics, natural disasters, and other emergencies) ensuring these principles are met following international standards is challenging. eBASE Africa gained interest in the 3ie TREE template for transparent, reproducible, and ethical considerations for evidence (TREE) generation. It sought to test and review its effects on its IRB review process.
Objective
To improve transparency, reproducibility, and ethical considerations for evidence generation in conflict and fragile settings in Cameroon
Methods
Studies submitted by organizations in Europe, North America, and Africa for research in 6 regions of Cameroon. We used the 3ie TREE template to review 6 studies conducted in settings affected by conflict, cholera, and COVID-19.
We categorized findings under 10 ethical standards recognized by the TREE template (36 points). We further categorized these into Transparency (11 points), Reproducibility (8 points), and Ethical Considerations (36 points). We also considered Transparency and Reproducibility as ethical considerations standards but calculated them independently. All IRB members (6) read each research protocol. Two reviewers independently reviewed each research protocol and presented findings by categories to the IRB. IRB deliberated and prepared a final document for the research team which included recommendations, independent anonymized reviews, and IRB conclusion.
Results
We evaluated 6 studies with a total of 70,392 participants. TREE scores by categories were: Transparency 56% [Range: 36%-82%], Reproducibility 46% [Range: 0% to 88%], and Ethical Consideration 48% [range: 38% -64%]. The overall mean TREE score was 49% [Range: 0% -88%]. We reported an initial Kappa score between reviewers of 0.9 [R 0.8-1.0]. Based on IRB recommendations, research teams were able to improve on TREE by 33%.
Discussion/Conclusion
The TREE template though not developed for IRB/scoring, has been used innovatively by the eBASE IRB to guide ethics process and ensure key ethical issues are considered. The rigorous, transparent, and systematic process of the TREE approach allows for identification of potential risks for research participants and teams and has the potential to improve uptake of ethical review boards.
"