Abstract
Data science health research promises tremendous benefits for African populations, but its implementation is fraught with substantial ethical governance risks that could thwart the delivery of these anticipated benefits. We discuss emerging efforts to build ethical governance frameworks for data science health research in Africa and the opportunities to advance these through investments by African governments and institutions, international funding organizations and collaborations for research and capacity development.
Similar content being viewed by others
Introduction
Data science is poised to revolutionize healthcare and research by enabling the development of novel interventions and groundbreaking strategies derived from high-quality and efficient analyses of the huge datasets derived from the activities of our multifaceted lives. Data science applications use high-performance computational infrastructure to process massive datasets from personal, public, and commercial sources including healthcare systems, smartphones, shopping records, social media postings, and wearable devices. Using novel, complex, and occasionally opaque algorithms, data scientists generate new insights and generalizable knowledge. Examples of data science applications include combination of diverse data streams to develop bio-preparedness, monitoring, and response strategies for infectious diseases outbreaks in human health and in agriculture1. Other examples include the use of Geographical Information System (GIS) data to map spatial variations in the determinants, incidence, prevalence, and outcomes of disease, and the response of healthcare systems2,3. Several countries also utilize data science to monitor and evaluate multi-sectoral progress towards meeting the UN Sustainable Development Goals (SDG), reduce fraud and corruption, identify fake pharmaceutical products, and improve supply chain management to prevent stock-outs4,5,6,7,8,9,10,11. More recently, Large Language Models (LLM) have captured public imagination with the release of tools like GPT-4. LLMs are rapidly being used to transform multiple industries and disciplines12,13,14,15,16,17. With these capabilities, data science is rapidly transforming the landscape and touching vast areas of human endeavors.
Data science health research is the novel application of data science methods and technologies for systematic generation, collection, processing, storage, management, analyses, visualization, interpretation, and communication of health-related data to develop generalizable knowledge and generate actionable insights. Examples of data science health research include integration of data from multiple omics technologies to generate insights about biological mechanisms and diseases’ pathways, and identify novel therapeutic and preventive opportunities18. Other examples include mining electronic health records for precision medicine19 and analysis of medical images for computational histopathology and radiology to improve diagnoses, among other uses20,21. In public health, data scientists are transforming practice through the application of high-level computational methods to population-level datasets typically used in public health to advance precision public health practice and research22. Examples of public health data science research include mining social media data for early detection, analyses of trends, education, planning and implementation of health systems responses to infectious and non-communicable diseases8,9,11,23, and application of myriad novel technologies and algorithms for precision public health22. Data science health research therefore presents huge opportunities for the application of novel methods and transformative technologies that would solve many healthcare challenges facing African people today and enable wider availability of high quality and cost-efficient health services.
Africa, potentially, has the most to gain from implementation of data science for health care and research. With a population expected to reach 2.5 billion people or ~25% of the world’s population by 2050, data science technologies would enable African countries to leapfrog legacy healthcare systems and technologies, and dramatically transform lives on the continent24,25. Even though Africa currently constitutes 17% of the world’s population, it bears 25% of the world’s disease burden, has only 3% of the world’s healthcare workers, and 2% of global health research output26. This is due to limited infrastructure, lack of trained personnel, poor funding, economic and social instability which hinder access to clinical and preventive services26. Global public health emergencies such as emerging and re-emerging infectious diseases epidemics and climate change pose more challenges to African countries than the rest of the world. African countries would therefore require innovative data science tools and strategies to overcome these challenges.
Substantial gaps exist in the representation of people from Africa in the datasets currently used to build data science models and applications27. This underrepresentation renders data science models and algorithms unstable and potentially inaccurate in African populations27. Without dedicated and focused efforts at remediation, persistence of this data science equity gap would worsen and this portends dire consequences for data science health research in African populations.
African researchers, institutions, governments, and the private sector are already using data science for research, discoveries, and preventive and clinical care28,29. Most of these uses involve novel applications or extensions of current healthcare expertise and technologies. Examples of data science applications already in use in Africa include teleradiology and telepathology, patients’ navigation and clinical decision support (CDS) tools, integration of genomics data into public health and clinical care, and cancer screening29,30,31,32,33. However, most of these applications were designed, developed, tested, and validated outside Africa. They may not have been adequately evaluated in African populations and may be insensitive to local contexts and health priorities28,34,35. They may therefore primarily benefit people outside the continent. Given the novel opportunities being created by data science applications, it is critical to develop and implement technologies that are relevant and adapted to the contexts in which they would be used in Africa.
Investment in data science health research infrastructure in Africa
Several initiatives are being implemented in Africa to develop data science health research capacity, build infrastructure, implement training programs, organize scientific conferences, and engage in international collaborations that would empower African institutions to generate datasets, develop and apply data science models, and close the data science gap between Africa and high-income countries (HIC). In 2022, the NIH Common Fund awarded 20 grants worth $74.5 million in the “Harnessing Data Science for Health Discovery and Innovation in Africa (DS-I Africa)” program to accelerate data science health research in Africa. The projects being implemented by the DS-I Africa program include a Coordinating Center, an Open Data Science platform, seven training programs, four ethical, legal, and social implications (ELSI) projects, and seven research projects. (Table 1)36.
The DS-I Africa initiative builds on the infrastructure previously developed by programs such as the $176 million Wellcome Trust and NIH-funded Human Heredity & Health in Africa (H3Africa) program37. H3Africa built new collaborations among scientists, developed genomics research infrastructure, and created publicly available governance and ethics policies for the African genomics research ecosystem38,39,40. Data science conferences and training programs are also proliferating in Africa including the Data Science Africa—an AI and Data Science Research Group at Makerere University, Uganda, the multi-country African ML and AI organization— Deep Learning Indaba, the School for Data Science and Computational Thinking at Stellenbosch University in South Africa, the African Institute for Mathematical Sciences Centre of Excellence in Cameroon and the African Center of Excellence in Data Science in the University of Rwanda. These programs are critical to generating data that will close the data science gap in Africa and enrich global resources for data science health research.
Developing a comprehensive framework for the governance of data science health research across Africa
In contrast with other disciplines where data science is also rapidly advancing, health research already has established frameworks and infrastructure for ethical governance. Substantial investments by the US National Institutes of Health (NIH), UK Wellcome Trust, the European Union through the European-Developing Countries Clinical Trials Partnership (EDCTP), African governments and institutions have significantly expanded African health research ethics infrastructure in the past few decades41,42,43,44,45. Despite these investments, there remain many unresolved challenges including concerns about quality of informed consent, data ownership, data sharing, benefit-sharing, privacy, autonomy, exploitation, and weak governance46,47,48,49. Recent examples of these challenges include controversies about community benefit and data sharing during COVID-19 pandemic research and unauthorized use of DNA samples from African populations to develop a DNA genotyping microarray chip38,50. Given the methods and technologies used in data science health research, its potential to exacerbate preexisting health research ethics problems and generate new ones are quite substantial.
Research consortia like H3Africa developed policies on samples and data sharing, biorepositories, publications, collaborations, and commercialization38. They also provided training for researchers and ethics committees. For example, H3Africa’s publication policy gives African researchers protected time to analyze and publish before their data becomes publicly available. This protection, which is designed to accommodate the infrastructural and personnel challenges faced by African researchers, enables them to frame the narrative about their research and advance their research priorities. Other policies require that the funded studies should focus on African health priorities, be led by African researchers, and that African institutions should be the primary recipients of the research grants, even when they collaborate with international institutions. These are meaningful practices that the emerging data science health research programs should emulate and further develop.
The rapid evolution of data science methods, utilization of complex algorithms, and huge datasets obtained from a variety of sources under uncertain consenting procedures particularly challenges the current model of ethical review of health research51,52. When data science health research projects are conducted within single institutions, e.g., computational histopathology of diagnostic biopsies, sufficient ethical oversight can be provided by the institution’s health research ethics committee. However, even in these situations, institutional health research ethics committees may lack sufficient knowledge, expertise, and experience in evaluating the ethical dilemmas that may accompany these studies and struggle to provide adequate ethical review and oversight53. In addition, the methods and technologies of data science often run counter to the established principles and practices of ethics review of health research thereby creating situations that may be beyond the capacity of individual ethics committees to resolve54,55,56. In such cases and in others where data science health research is being conducted in multiple institutions within the same country, collaborations between the ethics committees or centralization of ethics review, for example, by national health research ethics committees may be required to provide ethical oversight57. National health research ethics committees can constitute standing review committees that, in addition to chartered members, may also include local and international experts as ad-hoc members who can provide ethical oversight for complex data science health research within national boundaries57. This centralization of ethical review enables efficient utilization of scarce data science health research ethics expertise and improves the efficiency of ethical review57,58. Despite these innovations, even national health research ethics committees are susceptible to some of the problems affecting institutional health research ethics committees including lack of resources, lack of independence, poor funding and lack of efficacy, albeit to a lesser degree44,45,53.
What should be next on the agenda for data science health research ethics in Africa?
Improve institutional and national health research ethics governance infrastructure
Despite the tremendous investments in recent decades, the capacity, resources, and infrastructure for ethical oversight of health research in Africa remains weak and poorly resourced. A surge of data science health research projects would significantly strain and may overwhelm the system. The major responsibility for building and maintaining national and institutional health research ethics infrastructure rests with African governments and local institutions. Information showing how research significantly boosts the intellectual and economic capital of institutions and countries, and are engines for growing local and national economies may encourage more investment in research infrastructure, including ethical review in Africa. Many African institutions built their current health research ethics programs to support local investigators involved in collaborative international research projects. Research sponsors should incentivize development of local data science health research ethics capacity by linking progress in this domain with new research funding59. This would be highly impactful and motivate significant institutional response.
Research ethicists should engage with their local research and data science communities to better understand data science health research methods and projects, and jointly develop ethical governance frameworks that build on existing research ethics oversight infrastructure. Well-funded, well-designed, and sufficiently long training programs that have enabled African countries to avoid widespread egregious harm to research participants despite recent growth in the volume and sophistication of health research on the continent, are also needed for data science health research ethics42. These would improve the knowledge of research ethicists about data science health research and that of data scientists about health research ethics, build local capacity that would enable local ownership and sustainment of training programs, and support the conduct of research into contextual data science health research ethics in Africa while contributing to the global health research ethics discourse.
Develop culturally and resource-level appropriate national laws, guidelines, and regulations, and the infrastructure for enforcement
Many African governments are rushing to enact laws similar to the European Union’s General Data Protection Regulation (GDPR) and modifying them for their environments60,61. The major challenges with GDPR and similar data protection laws include lack of sufficiently explicit frameworks for enforcement, complexity of certain provisions, a focus that is often insufficient for the nuances of data science health research62,63. Other suggestions for ethical governance of data science include giving participants ownership of their digital selves or using blockchain technologies to protect digital privacy and securely share data64,65. These approaches are highly technological, expensive and are not resource-level appropriate in the African health research setting66.
Research and training consortia are developing novel policies, ideas, and implementation strategies for ethical regulation of data science health research67,68. These consortia must engage frontline stakeholders in different research environments and serve as petri-dishes for experiments into ethical regulation of data science health research. Governments and their agencies also have major roles to play in engendering and maintaining public trust, accountability, and support that are required to sustain public engagement in and support for data science health research.
Existing health research ethics regulations already have the essential elements for the ethical oversight of all types of health research. While data science health research includes novel methods and technologies, these do not abrogate the foundations, principles, and practices of modern health research ethics. African countries can quickly introduce sufficient oversight of data science health research by adding to or modifying existing regulations.
Develop and implement enforceable multinational regulations
Most data science health research sponsors, principal investigators, and scientists belong to international or commercial organizations that may not have local offices in Africa and may not be subject to national laws, guidelines, and regulations58. This poses significant problems for oversight and accountability. Multilateral agencies including the United Nations and its organs, governments, advocates, bioethicists, and researchers have conducted multiple consultations and stakeholders’ meetings leading to issuance of guidelines on the use of data science in healthcare, research, and policy69,70. These guidelines call for development of multinational frameworks for data science health research to prevent egregious harm and maximize data science’s benefits to global populations52. To ensure relevance and implementation, these multinational agencies should work with African institutions to develop the mechanisms for enforcement of these principles, model laws, guidelines, and regulations for ethical oversight of data science health research across national borders. They should rigorously engage a broad range of stakeholders including those whose voices are typically drowned out in global discourse. Innovations in virtual meeting technologies should enable cost-effective, frequent, and sustained global engagement of stakeholders.
Reduce digital inequity and increase volume and diversity of African datasets
Data science relies on large repositories of data generated by individuals as they engage with the healthcare system, during activities of daily living, and participation in research71. Healthcare data may derive from electronic health records, surveillance data, diseases’ registries, etc., while other datasets may be derived from wearable devices and other digital footprints. Large-scale genomic, transcriptomic, proteomic, and other omics research projects are also generating huge amounts of data for data science health research. Despite interventions like H3Africa and DS-I Africa, more of these data are still being generated in HIC compared to LMIC27,72. The resulting digital data inequity is pervasive and growing worse73. Digital health innovations also contribute to the widening inequities because of the “inverse care law” which postulates that well-resourced individuals are more likely to be aware of and utilize digital health interventions74. Unaddressed, these inequities will lead to severe and adverse health outcomes for majority of the world’s population75. Urgent, sustained, large-scale efforts are required to reverse this trend for the sake of equity and justice.
Multi-level interventions guided by frameworks for digital health equity such as the digital determinants of health (DDOH) would be useful for identifying the barriers and facilitators, and guide meaningful interventions to increase the volume of digital health data generated in Africa73. General investments in healthcare systems, implementation of electronic health records, improvements in diseases’ registries, and broader utilization of digital systems will increase the amount of digital data generated by African populations. Additional systematic interventions that are similar to but substantially larger than programs like H3Africa and DS-I Africa are needed to ensure that African countries, at a minimum, keep up with the high volume of omics and other data types being generated in HIC for data science.
Reduce and eliminate algorithmic bias, data colonization, and extractive research
Data science technologies produce algorithmic bias by replicating and reinforcing societal biases that benefit or disadvantage certain individuals or groups. This results in structural, racial and ethnic biases in the HIC where most data science technologies are developed76. These algorithmic biases coupled with the lack of equity and diversity in the foundational datasets used to develop, train, and validate data science algorithms lead to algorithmic deprivation, discrimination, and distortion77,78,79,80,81. Other concerns, particularly with respect to data science in Africa, are data colonization and extractive research82. Biased and inequitable algorithms lead to ethically, socially, politically, and economically undesirable outcomes in health research and health care, and can negatively affect perceptions of fairness, acceptability, and trust in applications derived from data science health research. These have the potential of denying populations that are most in need, the benefits of data science health research76. These harms are unpredictable and may not be remediable post hoc, they therefore require vigorous and robust attention a priori51,52,76.
Many approaches have been recommended to reduce or eliminate algorithmic bias in data science health research79,81,83,84,85. These include improving the diversity of data scientists through targeted capacity building programs, creation and implementation of guidelines and policies, implementation of programs to detect and rectify algorithmic bias, training data scientists on health research ethics52,85,86,87. These interventions require long-term commitments that go beyond the typical duration of many HIC research grant award mechanisms. They also require strong commitment by national governments, local institutions, and research sponsors. Novel strategies for supporting the development of personnel, resources, and infrastructure for data science health research in Africa that are aligned with clear goals and objectives, rather than just utilizing frameworks developed and used in the substantially different health research ecosystems of HIC are desperately needed.
Given the scope of data science health research, its potential to improve health outcomes, engender more equitable research participation, reduce marginalization, and utilize heterogeneous data types, all stakeholders must be urgently engaged in development of the most efficacious governance frameworks for it.
References
Keshavamurthy, R., Dixon, S., Pazdernik, K. T. & Charles, L. E. Predicting infectious disease for biopreparedness and response: a systematic review of machine learning and deep learning approaches. One Health 15, 100439 (2022).
Tanser, F. C. & le Sueur, D. The application of geographical information systems to important public health problems in Africa. Int. J. Health Geogr. 1, 4 (2002).
Stewart, K. et al. Modeling spatial access to cervical cancer screening services in Ondo State, Nigeria. Int J. Health Geogr. 19, 28 (2020).
Georgakopoulos, S. V., Gallos, P. & Plagianakos, V. P. Using Big Data Analytics to Detect Fraud in Healthcare Provision in 2020 IEEE 5th Middle East and Africa Conference on Biomedical Engineering (MECBME 2020). (IEEE, New Jersy, 2020).
Gebremeskel, G. B., Yi, C., He, Z. & Haile, D. Combined data mining techniques based patient data outlier detection for healthcare safety. Int. J. Intell. Comput. Cybern. 9, 42–68 (2016).
Kumar, A., Choudhary, D., Raju, M. S., Chaudhary, D. K. & Sagar, R. K. Combating counterfeit drugs: a quantitative analysis on cracking down the fake drug industry by using Blockchain technology in Proc. 9th International Conference on Cloud Computing, Data Science & Engineering (Confluence 2019). (IEEE, Noida, India, 2019).
United Nations. Big data for sustainable development, https://www.un.org/en/global-issues/big-data-for-sustainable-development (2023).
Arslan, J. & Benke, K. K. Artificial Intelligence and Telehealth may provide early warning of epidemics. Front. Artif. Intell. 4, 556848 (2021).
Judson, S. D. et al. COVID-19 data reporting systems in Africa reveal insights for future pandemics. Epidemiol. Infect. 150, e119 (2022).
Shaw, N. & McGuire, S. Understanding the use of geographical information systems (GISs) in health informatics research: a review. BMJ Health Care Inform. 24, 940 (2017).
Kimera, R. et al. in Leveraging Data Science for Global Health (eds Celi, L. et al.) 329–350 (Springer International Publishing, 2020).
Waisberg, E. et al. GPT-4: a new era of artificial intelligence in medicine. Irish J. Med. Sci. https://doi.org/10.1007/s11845-023-03377-8 (2023).
Akbasli, I. T. & Bayrakci, B. Balancing caution and innovation: exploring the potential of large language models in critical decision-making. Crit. Care 27, 172 (2023).
Bair, H. & Norden, J. Large Language Models and Their Implications on Medical Education. Acad. Med. https://doi.org/10.1097/ACM.0000000000005265 (2023).
Perera Molligoda Arachchige, A. S. Large language models (LLM) and ChatGPT: a medical student perspective. Eur. J. Nucl. Med. Mol. Imaging https://doi.org/10.1007/s00259-023-06227-y (2023).
Qureshi, R. et al. Are ChatGPT and large language models “the answer” to bringing us closer to systematic review automation? Syst. Rev. 12, 72 (2023).
Sorin, V., Barash, Y., Konen, E. & Klang, E. Large language models for oncological applications. J. Cancer Res. Clin. Oncol. https://doi.org/10.1007/s00432-023-04824-w (2023).
Dar, M. A. et al. Multiomics technologies: role in disease biomarker discoveries and therapeutics. Brief. Funct. Genomics 22, 76–96 (2023).
Martínez-García, M. & Hernández-Lemus, E. Data integration challenges for machine learning in precision medicine. Front. Med. 8, 784455 (2021).
Shmatko, A., Ghaffari Laleh, N., Gerstung, M. & Kather, J. N. Artificial intelligence in histopathology: enhancing cancer research and clinical oncology. Nat. Cancer 3, 1026–1038 (2022).
Milam, M. E. & Koo, C. W. The current status and future of FDA-approved artificial intelligence tools in chest radiology in the United States. Clin. Radiol. 78, 115–122 (2023).
Goldsmith, J. et al. The emergence and future of public health data science. Public Health Rev. https://doi.org/10.3389/phrs.2021.1604023 (2021).
Cohen, I. G., Lynch, H. F., Vayena, E. & Gasser, U. Big Data, Health Law, and Bioethics (Cambridge University Press, 2018).
World Health Organization. Atlas of African health statistics 2022: health situation analysis of the WHO African Region. (World Health Organization. Regional Office for Africa, 2022).
Anon. in The Economist (The Economist Group, London, 2020).
World Health Organization. World health statistics 2022: monitoring health for the SDGs, sustainable development goals, https://www.who.int/publications-detail-redirect/9789240051157 (2022).
Gao, Y., Sharma, T. & Cui, Y. Addressing the challenge of biomedical data inequality: an Artificial Intelligence Perspective. Annu. Rev. Biomed. Data Sci. https://doi.org/10.1146/annurev-biodatasci-020722-020704 (2023).
Owoyemi, A., Owoyemi, J., Osiyemi, A. & Boyd, A. Artificial intelligence for healthcare in Africa. Front. Digital Health 2, 6 (2020).
Ezugwu, A. E., Oyelade, O. N., Ikotun, A. M., Agushaka, J. O. & Ho, Y.-S. Machine learning research trends in Africa: a 30 years overview with bibliometric analysis review. Arch. Comput. Methods Eng. 30, 4177–4207 (2023).
Bellemo, V. et al. Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study. Lancet Digital Health 1, e35–e44 (2019).
Botwe, B. O. et al. The integration of artificial intelligence in medical imaging practice: Perspectives of African radiographers. Radiography 27, 861–866 (2021).
Guo, J. & Li, B. The application of medical artificial intelligence technology in rural areas of developing countries. Health Equity 2, 174–181 (2018).
Wani, S. U. D. et al. Utilization of Artificial Intelligence in disease prevention: diagnosis, treatment, and implications for the healthcare workforce. Healthcare 10, 608 (2022).
Zou, J. & Schiebinger, L. Ensuring that biomedical AI benefits diverse populations. EBioMedicine 67, 103358 (2021).
Kaushal, A., Altman, R. & Langlotz, C. Health care AI systems are biased. Sci. Am. 11, 17 (2020).
Data Science for Health Discovery and Innovation in Africa. Our Projects https://dsi-africa.org/projects (2023).
Consortium, H. A. et al. Research capacity. Enabling the genomic revolution in Africa. Science 344, 1346–1348 (2014).
de Vries, J. et al. Addressing ethical issues in H3Africa research–the views of research ethics committee members. HUGO J. 9, 1–4 (2015).
Bentley, A. R., Callier, S. L. & Rotimi, C. N. Evaluating the promise of inclusion of African ancestry populations in genomics. NPJ Genom. Med. 5, 5 (2020).
Ozulumba, T. Sustaining breakthroughs in health research in Africa, https://www.nature.com/articles/d44148-021-00124-y (2021).
Millum, J., Grady, C., Keusch, G. & Sina, B. Introduction: the Fogarty International Research Ethics Education and Curriculum Development Program in historical context. J. Empir. Res. Hum. Res. Ethics 8, 3–16 (2013).
Millum, J., Sina, B. & Glass, R. International research ethics education. JAMA 313, 461–462 (2015).
Ndebele, P. et al. Research ethics capacity building in Sub-Saharan Africa: a review of NIH Fogarty-funded programs 2000-2012. J. Empir. Res. Hum. Res. Ethics 9, 24–40 (2014).
Hummel, P., Adam, T., Reis, A. & Littler, K. Taking stock of the availability and functions of National Ethics Committees worldwide. BMC Med. Ethics 22, 56 (2021).
Kohler, J., Reis, A. A. & Saxena, A. A survey of national ethics and bioethics committees. Bull. World Health Organ. 99, 138–147 (2021).
Bedeker, A. et al. A framework for the promotion of ethical benefit sharing in health research. BMJ Glob. Health 7, e008096 (2022).
Staunton, C. & de Vries, J. The governance of genomic biobank research in Africa: reframing the regulatory tilt. J. Law Biosci. 7, lsz018 (2020).
Chaudhry, I. et al. Strengthening ethics committees for health-related research in sub-Saharan Africa: a scoping review. BMJ Open 12, e062847 (2022).
Sudoi, A., De Vries, J. & Kamuya, D. A scoping review of considerations and practices for benefit sharing in biobanking. BMC Med. Ethics 22, 102 (2021).
Moodley, K. et al. Ethics and governance challenges related to genomic data sharing in southern Africa: the case of SARS-CoV-2. Lancet Glob. Health 10, e1855–e1859 (2022).
Metcalf, J. & Crawford, K. Where are human subjects in Big Data research? The emerging ethics divide. Big Data Soc. 3, 2053951716650211 (2016).
Ada Lovelace Institute. Looking before we leap: Ethical review processes for AI and data science research. (Ada Lovelace Institute, Exeter, UK, 2022).
Ferretti, A., Ienca, M., Velarde, M. R., Hurst, S. & Vayena, E. The Challenges of Big Data for Research Ethics Committees: A Qualitative Swiss Study. J. Empir. Res. Hum. Res. Ethics 17, 129–143 (2022).
Mahomed, S. & Labuschaigne, M. L. The evolving role of research ethics committees in the era of open data. S. Afr. J. Bioethics Law https://doi.org/10.7196/SAJBL.2022.v15i3.XX (2023).
Silaigwana, B. & Wassenaar, D. Biomedical Research Ethics Committees in Sub-Saharan Africa: A Collective Review of Their Structure, Functioning, and Outcomes. J. Empir. Res. Hum. Res. Ethics 10, 169–184 (2015).
Bain, L. E., Ebuenyi, I. D., Ekukwe, N. C. & Awah, P. K. Rethinking research ethics committees in low- and medium-income countries. Res. Ethics 14, 1–7 (2018).
National Health Research Ethics Committee of Nigeria. (Federal Ministry of Health of Nigeria, Abuja, 2007).
Leonelli, S. Locating ethics in data science: responsibility and accountability in global and distributed knowledge production systems. Philos. Trans. R. Soc. A: Math., Phys. Eng. Sci. 374, 20160122 (2016).
Terry, R., Littler, K. & Olliaro, P. Sharing health research data? the role of funders in improving the impact [version 2; peer review: 3 approved]. F1000Research https://doi.org/10.12688/f1000research.16523.2 (2018).
Greenleaf, G. & Cottier, B. Data privacy laws and Bills: Growth in Africa, GDPR influence. GDPR Influence 152, 11–13 (2018).
Daigle, B. Data protection laws in Africa: a pan-African survey and noted trends. J. Int’l Com. Econ. 1, 1–27 (2021).
McCall, B. What does the GDPR mean for the medical community? Lancet 391, 1249–1250 (2018).
Irish Council for Civil Liberties. 5 years: GDPR’s crisis point, https://www.iccl.ie/wp-content/uploads/2023/05/5-years-GDPR-crisis.pdf (2023).
Mann, S. P., Savulescu, J., Ravaud, P. & Benchoufi, M. Blockchain, consent and prosent for medical research. J. Med. Ethics 47, 244–250 (2021).
Morley, J. et al. The ethics of AI in health care: a mapping review. Soc. Sci. Med. 260, 113172 (2020).
Kalenzi, C. Artificial Intelligence and Blockchain: How Should Emerging Technologies Be Governed? Front. Res. Metrics Analytics https://doi.org/10.3389/frma.2022.801549 (2022).
Global Health Network. Global Health Data Science, https://globalhealthdatascience.tghn.org/ (2023).
Bridging Gaps in the ELSI of Data Science Research (BridgELSI) Project. 21 May, https://bioethicscenter.net/bridgelsi-project/ (2023).
World Health Organization. Ethics and governance of artificial intelligence for health: WHO guidance. (2021).
World Health Organization. Health Data as a global public good – a call for Health Data Governance 30 September, https://www.who.int/news-room/articles-detail/health-data-as-a-global-public-good-a-call-for-health-data-governance-30-september (2021).
Edwards, T. L. et al. Challenges and Opportunities for Data Science in Women’s Health. Annu. Rev. Biomed. Data Sci. https://doi.org/10.1146/annurev-biodatasci-020722-105958 (2023).
Mills, M. C. & Rahal, C. The GWAS Diversity Monitor tracks diversity by disease in real time. Nat. Genet. 52, 242–243 (2020).
Richardson, S., Lawrence, K., Schoenthaler, A. M. & Mann, D. A framework for digital health equity. npj Digital Med. 5, 119 (2022).
Lyles, C. R., Nguyen, O. K., Khoong, E. C., Aguilera, A. & Sarkar, U. Multilevel determinants of digital health equity: a literature synthesis to advance the field. Annu. Rev. Public Health 44, 383–405 (2023).
Brewer, L. C. et al. Back to the future: achieving health equity through health informatics and digital health. JMIR Mhealth Uhealth 8, e14512 (2020).
Kordzadeh, N. & Ghasemaghaei, M. Algorithmic bias: review, synthesis, and future research directions. Eur. J. Inf. Syst. 31, 388–409 (2022).
Yu, P. K. The algorithmic divide and equality in the age of artificial intelligence. Fla. L. Rev. 72, 331 (2020).
Panch, T., Mattie, H. & Atun, R. Artificial intelligence and algorithmic bias: implications for health systems. J. Glob. Health 9, 010318 (2019).
Thomasian, N. M., Eickhoff, C. & Adashi, E. Y. Advancing health equity with artificial intelligence. J. Public Health Policy 42, 602–611 (2021).
de Hond, A. A., van Buchem, M. M. & Hernandez-Boussard, T. Picture a data scientist: a call to action for increasing diversity, equity, and inclusion in the age of AI. J. Am. Med. Inform. Assoc. 29, 2178–2181 (2022).
Leavy, S. in Proc. 1st International Workshop on Gender Equality in Software Engineering. (eds E. Abraham, E. Di Nitto, & R. Mirandola) 14–16 (Association for Computing Machinery, 2018).
Gwagwa, A., Kraemer-Mbula, E., Rizk, N., Rutenberg, I. & De Beer, J. Artificial Intelligence (AI) deployments in Africa: benefits, challenges and policy dimensions. Afr. J. Inf. Commun. 26, 1–28 (2020).
Park, Y. et al. in Healthcare Information Management Systems: Cases, Strategies, and Solutions (eds Joan M. Kiel, George R. Kim, & Marion J. Ball) 223–234 (Springer International Publishing, 2022).
Dankwa-Mullan, I. & Weeraratne, D. Artificial intelligence and machine learning technologies in cancer care: addressing disparities, bias, and data diversity. Cancer Discov. 12, 1423–1427 (2022).
United Nations Sustainable Development Group. Data Privacy, Ethics and Protection. Guidance Note on Big Data for Achievement of the 2030 Agenda. 19 (United Nations, Geneva, Switzerland, 2017).
Katell, M. et al. in FAT*’20: Proceedings of the 2020 conference on fairness, accountability, and transparency. (eds M. Hildebrant et al.) 45–55 (Association for Computing Machinery, New York, NY, 2020).
Embi, P. J. Algorithmovigilance—advancing methods to analyze and monitor artificial intelligence–driven health care for effectiveness and equity. JAMA Netw. Open 4, e214622–e214622 (2021).
Acknowledgements
This work is supported by the following grants: 1. Bridging Gaps in the ELSI of Data Science Health Research in Nigeria (BridgELSI, U01 MH127693) to CAA and TO. 2. The Maryland Department of Health’s Cigarette Restitution Fund Program (CH-649-CRF) to CAA and SNA. 3. The University of Maryland Greenebaum Comprehensive Cancer Center Support Grant (P30CA134274) to CAA and SNA. The funding agencies did not play any role in the publication.
Author information
Authors and Affiliations
Consortia
Contributions
C.A.A. conceived the study, led the co-authors in writing and revising the manuscript, read all papers and documents referenced in the manuscript to ensure correct interpretation and finalized the version submitted for publication. S.C. co-wrote the paper, revised the manuscript, and approved the final version. S.A., O.M., A.J., C.A., T.O., and S.N.A. co-wrote, read and critiqued several drafts of the manuscript and contributed to development and refinement of the ideas in this paper. They reviewed and approved the final draft of the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Robert Terry, and the other anonymous, reviewers for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Adebamowo, C.A., Callier, S., Akintola, S. et al. The promise of data science for health research in Africa. Nat Commun 14, 6084 (2023). https://doi.org/10.1038/s41467-023-41809-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-41809-2
This article is cited by
-
Health in Africa
Nature Communications (2024)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.