Decolonisation and Self-Regulation as Alternative Paths to Data Science Health Research Governance in Africa

TL;DR

This paper explores how African communal values and governance can address ethical challenges in data science health research, offering alternatives to traditional consent models.

Contribution

The paper introduces decolonisation and self-regulation as practical frameworks for ethical data science health research governance in African contexts.

Findings

African case studies show communal approval is prioritized over individual consent in data science health research.

Community engagement and advisory boards are key elements in facilitating ethical research practices.

Decolonisation and self-regulation require deliberate action to move beyond theoretical discourse.

Abstract

Data science health research (DSHR) presents new ethical challenges to the traditional model of human subject research, particularly by enabling data processing without the consent of data subjects. Although the current research governance framework makes informed consent a cornerstone of ethical research practices, obtaining individual consent can often be impractical in DSHR. This paper explores the alignment of DSHR with African customary governance and communal lifestyles as a framework for ethical research oversight.

Using a mixed-method approach, this study integrates doctrinal analysis of legal and policy frameworks with case studies from Nigeria, Kenya, Ghana, Uganda, and South Africa. Data were synthesised from peer-reviewed literature, with a focus on initiatives that operationalise decolonised governance.

Data science health research challenges traditional biomedical ethics by enabling data processing without consent, thereby questioning the longstanding principle that informed consent is a prerequisite for ethical research. However, this principle has been widely contested as a universal standard, particularly in African contexts where decision-making is often communal rather than individualistic. Case studies from Nigeria, Kenya, Ghana, Uganda and South Africa illustrate that while informed consent remains a normative requirement, largely to satisfy the expectations of funding bodies, communal approval is paramount. Furthermore, religious and cultural traditions often accommodate forms of paternalistic consent, reinforcing collective decision-making structures.

Given that African societies emphasise communal governance, the ethical challenges posed by DSHR, particularly regarding consent, may be less pronounced in Africa. However, decolonisation and self-regulation are not merely theoretical constructs, but a practical and necessary process that requires deliberate action. Unless African leaders take decisive steps to restructure governance, prioritise self-reliance, and invest in homegrown research and development, the discourse on decolonising DSHR in Africa will remain purely theoretical, lacking the practical implementation needed for real change.

Although informed consent is the criterion for ethical research conduct, data science health research (DSHR) presents novel ethical challenges by enabling data processing without the consent of data subjects. This study introduced the decolonisation and self-regulation framework, a practice that researchers argue is inherent in most African societies, involving a more communal approach to collective decision-making.

Drawing on select case studies from Africa, this study identifies practices such as community engagement, community advisory boards, respect for elders and community leaders as interconnected elements that play a crucial role in facilitating informed consent in research. These practices can be leveraged to strengthen the ethical governance of DSHR across the continent.

To fully harness the vast potential of DSHR, its successful implementation is essential. Decolonisation and the self-regulation framework require deliberate actions. Further research could explore more case studies in diverse African sociopolitical contexts to assess best practices for effective implementation.

Figures