This protocol paper outlines a rigorous modified e-Delphi study designed to validate and achieve stakeholder consensus on a novel ethical oversight framework for data science health research in Nigeria. Building on qualitative findings from earlier stakeholder consultations, this study seeks to refine and establish agreement on specific principles, guidelines, and governance mechanisms appropriate for the Nigerian context and applicable to other low- and middle-income countries.

The study addresses the reality that Nigeria’s current ethical guidelines were not designed to handle the complex challenges posed by data science health research, including pervasive data environments, consent for secondary data use, algorithmic decision-making and bias, privacy risks, commercial entity involvement, data colonization, and inequitable benefit-sharing arrangements. The researchers developed a draft guideline incorporating principles such as trust, veracity, and global justice, along with alternative ethical approval mechanisms tailored to data science research contexts.

The modified Delphi approach will engage 65 experts—including health researchers, ethics committee members, data scientists, policymakers, funders, and key opinion leaders—who will iteratively rate and refine 13 core principles covering global justice, algorithmic bias, data governance, and related provisions. Participants evaluate each principle on multiple dimensions including importance, desirability for inclusion in national guidelines, feasibility of implementation, and confidence in practical application using structured rating scales complemented by qualitative feedback. Through multiple rounds of anonymous expert consultation, the study aims to achieve consensus on a comprehensive ethical oversight framework that balances the need for adequate regulation with facilitation of international collaborations and partnerships that prioritize health research goals and innovation in Nigeria and similar contexts.