Is Africa Facing an Unprecedented Challenge in Transforming Data into Assets and Equitable Benefits?

Significant emerging pathogens and an unprecedented increase in communicable and non-communicable diseases in Africa have characterised the 21st century1. The increased prevalence of risk factors, resulting from weak economic conditions, the threat of climate change, and population growth, will exacerbate health conditions and inequalities. These threats are disproportionately impacting multiple determinants of health and augmenting the pressure on healthcare systems and public health communities across African countries.

Africa, often referred to as the cradle of genetics, is the second-largest continent, spanning approximately 30,365,000 km2, and the second-most populous continent with 1.4 billion inhabitants2. With its varying public health challenges and socio-political contexts, one would expect that the high-quality health (human and animal), environmental, and biomedical data generated across the African continent would serve as a strategic treasure and valuable assets for advancing data-driven public health solutions, economic growth, and data-sharing partnerships.

Harnessing this wealth of data holds immense potential for enhancing governance, strengthening disease surveillance systems, and improving diagnostic and therapeutic capabilities3,4. Furthermore, it can catalyze innovative research tailored to the needs of public health practitioners and policymakers, thereby contributing to more responsive and evidence-based decision-making.

Despite this potential, the healthcare sector across Africa continues to suffer from significant underutilization of available data. This gap undermines the generation of actionable knowledge, impedes infrastructure development, and limits capacity-building efforts. This becomes even more complicated with the abundance of different forms of health and biomedical-related data, its inherent complexity, and fragmentation, as well as issues related to data security and analytical capacity. These factors collectively render much of the data intractable, thereby constraining its utility for evidence-based decision-making.

Nevertheless, Africa has witnessed a growing interest in leveraging data science, an interdisciplinary field focused on extracting meaningful insights from complex datasets, to address unique challenges and unlock new opportunities5,6. The convergence of data science with epidemiology and public health surveillance represents a paradigm shift in the way health data is generated, analyzed, and utilized. In the African context, this integration holds immense potential to strengthen health systems, improve disease control, and support evidence-based policymaking.

It is from the above background that this paper delves into the need for leveraging data on health decision-making and policy implementation in Africa. It further explores the intricate relationship between the burgeoning field of data science, Africa’s persistent infrastructural deficiencies, and the continent’s ongoing healthcare transformation challenges. It advocates the power of data science, of which Africa must invest in strengthening the entire data value chain, a process involving intelligent data collection, ensuring open data-driven strategies are responsive to local contexts and needs, as well as translating data-driven decision-making.

Overview of data-driven decisions for healthcare

Advancing healthcare necessitates data-driven decision-making, with essential players—policymakers, healthcare providers, and administrators, as well as data scientists—relying on actionable insights to enhance health systems and achieve objectives6. Routine health information systems are the key to generating the statistics and indicators necessary for planning and performance assessments, resulting in improved patient care, real-time monitoring, and strengthened healthcare systems to achieve universal health coverage7. Additionally, data-driven strategies can streamline resource allocation, reduce healthcare expenses, and bolster health outcomes. This translates to a healthier, more productive workforce, leading to increased productivity, economic growth, and human satisfaction. It can also help identify potential areas of waste and thus propose strategies to mitigate the waste. Furthermore, such strategies catalyze innovation and entrepreneurship, creating new job opportunities and invigorating economic activity within the healthcare and technology sectors8.

In the past decades, health scientists have been promoting the notion of data-driven decisions and encouraging decision-makers to use the wealth of data typically available to them to make better and more informed decisions. Leaders are encouraged to use historical or comparable data to estimate a sound model, identify repeated patterns, and then apply this data knowledge to guide their decision-making. Even though such data-driven decisions were being applied on a small scale, the COVID-19 pandemic exposed the limitations of traditional decision-making models and highlighted the transformative power of data-driven models9,10. For instance, during the early stages of the pandemic in 2020, many African countries adopted social distancing and handwashing measures based on public health principles. However, as data on COVID-19 transmission became available, a shift occurred. Analyzing trends in daily infections, several countries transitioned to lockdown strategies to drastically reduce transmission rates11,12. This data-driven pivot illustrates the power of real-time data analysis in informing public health interventions. Beyond COVID-19, the Ebola outbreak in Sierra Leone (2014-2016) provides another compelling example. Contact tracing, a cornerstone of Ebola control, traditionally relied on manual methods. However, researchers implemented a data-driven approach using mobile phone network data to identify and track contacts of infected individuals more efficiently13. The complex nature of the healthcare crisis, therefore, offered a critical opportunity to use data as a vital tool, enabling more informed and effective responses to save lives. In Ghana, the government, relying on COVID-19 impact dataset gathered by the Ministry of Health, implemented the zipline drone policy, which allows delivery of medical supplies to remote and difficult to access areas in Ghana14. This policy successfully led to the delivery of over 500,000 medical supplies, saving millions of lives across the country. Another data-driven policy is observed in the Democratic Republic of Congo (DRC), where the Ministry of Health launched the community health strategy to improve access to quality health care in remote areas through community care sites called Site de Soins Communautaire (SSCs).

One challenge in public health decision-making in Africa lies not only in the underutilization of existing data but also in the poor quality and relevance of the data. For example, while data may be available for neglected tropical diseases (NTDs), their lack of application in policy and program development can lead to continued negative health outcomes15. Also, the SSCs established in DRC faced challenges as the data used to implement the policy was outdated, relying on 1984 census data, which did not reflect some of the modern challenges in health. Strengthening the capacity for producing timely, accurate, and context-relevant data and analysing it, and fostering collaboration between researchers and policymakers can address this challenge, ensuring data translates into actionable and evidence-based insights.

Another challenge in Africa specifically involves the misalignment between data collection and public health needs. Data systems are also geared towards generating national level estimates and not so much on data that can be used locally by sub-national decision makers, by healthcare managers and by healthcare workers to improve patient care. Furthermore, while data is meticulously recorded, its focus often aligns with the interests of funders, potentially neglecting crucial areas of national concern. This disconnect can hinder informed decision-making and impede the development of effective public health programs. Addressing this issue requires strong national leadership, prioritizing data collection aligned with critical public health needs, and engaging stakeholders in the process16.

Overall, implementing and maintaining a data-driven approach in healthcare systems will require recruiting and retaining the required expertise. Training the workforce, including data scientists, will require strong partnerships between governments, industry, and academia. It is very encouraging to witness the response of the scientific community, such as the Harnessing Data Science for Health Discovery and Innovation (DS-I Africa), Data Science Africa, and Deep Learning Indaba, and various governments toward closing the digital skills gap by investing in digital and AI literacy.

Unmet needs to turn African biomedical data into treasures and valuable assets

African data has been underrepresented and marginalized in global health research, and this needs to be corrected17. The continent has seen an influx of data companies like IBM, Oracle, and Microsoft, whose headquarters remain in the Global North. Many African businesses and governments now rely on data centres provided by these companies for cloud storage. Although some of these data centres are in Africa, these companies remain uninterested in data localization since their data centres in Africa remain an extension of the data farms outside the continent. This inevitably entrenches inequities in global data science for health and promotes unfavourable power relations that disadvantage African scientists.

In the healthcare systems in Africa, there are several levels of decision-making, which most often have the political elite making decisions that affect the entire healthcare system and persons who utilize its services18. At the different levels of decision-making, there are several requirements of capacities that can influence decision-making involving the use of data to formulate policies6,17,19. Data-driven decision-making in healthcare policy in Africa is influenced by other factors such as funding availability and cultural beliefs that can inform decisions that may have consequential outcomes for all.

Despite several funded international groups, consortia, and initiatives working on health projects to solve existing health needs, the underlying political economy for decision- and policy-making efforts remains largely uncoordinated among governments, research, private, and public institutions19. This incoherence remains a major block to facilitating data usage to enable appropriate decision-making and regulation. To overcome this, there needs to be substantial investment in digital open science and FAIR (Findable, Accessible, Interoperable, Reusable) data infrastructure that is sharable, secure, and technologically advanced. Crucially, this infrastructure should be designed for secure and shareable data exchange, maximizing its utility for the African population. The anticipated benefits - improved public health outcomes, increased service efficiency, and enhanced clinical care - justify these investments.

Whilst effective data-driven decision-making hinges on the transformative power of data science, artificial intelligence (AI) and machine learning (ML) playing a significant role, the state of AI and data governance in Africa is quite sobering considering only eight African countries have complete AI policies or guidelines, another fifteen have no laws on AI with the remaining are at various stages of their strategy, policy, and guidelines development. Of note, robust yet flexible data protection laws need to be adopted and implemented. The recent AU AI Strategic Plan will guide an African approach to AI governance and may provide the interconnected position needed to implement good AI and data governance across borders and jurisdictions.

A major challenge lies in overcoming the tension between traditional cultural practices of data secrecy and the agile data-sharing models necessary for effective data science and innovation. There is an unmet need to bridge the culture of secrecy in data collection and sharing to agile data-sharing partnership models and advance evidence-based healthcare policy across the continent. Nevertheless, by addressing these challenges and harnessing the power of data, Africa can unlock a future where data-driven healthcare solutions improve health outcomes and advance equitable access for all.

There is a critical need to develop solutions to the continent’s most pressing health and healthcare system problems through a robust ecosystem of new partners from the academic, government, and private sectors. Importantly, the African continent has a range of rich biodiversity, genetic diversity, and climate change, driving critical human adaptation and health impacts, thereby there is an unmet need to turn African health and biomedical data into assets that will be a state of business. Furthermore, focusing on African health and biomedical data space will not only boost our understanding of genetics and disease etiology but also contribute fundamentally new insights into the drivers of global health. African FAIR data asset is perfect for tackling the challenge of understanding the complex interplay of environment, climate change, and genetic etiology to disease and health.

Data-sharing partnership models in Africa

The health data ecosystem in Africa is structured as a multi-tiered system that reflects the flow of health information from the grassroots to global institutions (Table 1). This hierarchical structure shows the flow from the Community level to Health Facilities, District level, Provincial/Regional level, National level, and Global level. The structure provides a strong foundation to advance data-sharing partnerships in Africa by: (1) Enabling Interoperability: Standardized data collection tools (e.g., DHIS2) across levels facilitate seamless data exchange between institutions and countries, (2) Supporting Local-to-Global Integration: Data flows upward from communities to global agencies, ensuring that local realities inform international health strategies, (3) Enhancing Trust and Accountability: Clear governance structures and defined roles at each level foster transparency and mutual accountability among partners, (4) Promoting Equity and Responsiveness: Context-sensitive data use ensures that shared insights are relevant to diverse socio-political and health contexts across the continent, (5) Driving Innovation: Partnerships can leverage this ecosystem to co-develop tools, analytics, and platforms that address shared challenges like disease surveillance, health financing, and emergency response.

Table 1 Hierarchical structure of the health data ecosystem in Africa
Full size table

Currently, the most utilized health information and management system in Africa is the District Health Information Management System (DHIMS2), which has been implemented in most countries (Table 2). Since its implementation, for example, decision-making concerning malaria, HIV, and TB has been significantly influenced by data collected through the DHIMS2 platform. In Ghana, for example, it was previously assumed that providing antimalarial medication would help reduce anemia in pregnant women and therefore decrease the prevalence of low birth weights in children20. However, analyzing data from DHIMS2 showed that there is a missing factor, as the anti-malarial medication did not decrease low birth weight or reduce maternal anaemia. This insight informed the National Malaria Elimination Program of the Ghana Health Service that anemia is not the only cause of low birth weight, and therefore, a new policy in pregnancy will be required21.

Table 2 Key decision-making tools in health care and policy in Africa
Full size table

In addition, in Malawi, based on routine health data aggregated in the District Health Information System (DHIS2), a league table to rank and compare performance among health facilities has been introduced at the district level20. In South Africa, where DHIS2 has been in implementation since 1996, tools have been developed to improve outcomes in the reduction of mother-to-child transmission of HIV and increase antiretroviral therapy rollout.

Most African institutions and governments still do not have fair share partnerships with third parties. These institutions and governments do not realize the benefits of their data assets through adopting a win-win data-sharing partnership model and potentially maximizing wider partnerships with other African and international societies, for financial returns, innovative and equitable global health research, and improved healthcare systems.

Data sharing in a partnership model can enhance data-driven innovation by generating more research, development, and economic growth (Fig. 1).

Fig. 1: Different elements that can make data contribute to data-driven decisions and their impact.
figure 1

The data-sharing partnership model should deliver benefits to patients, society, and across the continent and the world. Secure data environments (SDEs) are data storage and access platforms that uphold the highest levels of privacy and security. This is also an analytics and research platform from which no raw data can be moved, and any processed data that is extracted should go through strict review committees. The SDEs can overcome several current issues of transparency, trustworthiness, privacy, cross-border, public mistrust, and data security, enhancing and enabling partners to bring facilities, expertise, and funding. In doing so, combined with data assets within trustworthiness, SDEs can potentially lead to crucial innovation that transforms patient care and public health community service. Once SDEs have been implemented, entering data-sharing partnership agreements should define how data assets will be created, deployed, maintained, tracked, and shared between partners. It is worth noting that a data sharing partnership agreement (DSPA) must reflect value sharing that prioritizes the benefits of patients through improved, for example, clinical outcomes, optimised medications, improved public health, and SDEs services. In addition to this, the DSPA should also consider all applicable legal, privacy, regulatory, and security obligations and exclusive arrangements. The key goal of DSPA must be to promote innovative data solutions by aligning, simplifying, and incentives procedures. In addition, establishing DSPA through SDEs can potentially support improvement in patient pathways, clinical decision-making, discovering and developing new interventions, and targeted therapeutics. Given the potential need for African data assets, the continent needs to shift to a secure data environment to allow innovation to take place at a more rapid pace. Africa is the best place to invest in developing health data quality, infrastructure, and digital capabilities that will leverage its health data assets for transformative health, scientific, and economic impact.

Full size image

Roadmap for a successful data-driven decision-making strategy for the African health sector

Data science without FAIR data assets will not generate robust science to enhance decision-making. As quality data is generated at different points, leveraging a federated data analysis ecosystem (FDAE), will enable secure and collaborative analysis of data from multiple sources without the need to centralize the data and limit scalability, within SDEs, it is imperative to promote quality data-driven decisions. A federated data analysis ecosystem can also promote data-driven decision-making by enabling organizations to leverage data from SDE and collaborate on analysis without compromising data privacy and security. This distributed approach ensures compliance with FAIR data principles, promoting robust research and reliable decision-making. Additionally, FDAEs can handle large data volumes efficiently, facilitating quicker insights and more agile decision-making processes. While FDAE is crucial, it requires capacity building in data science techniques such as federated learning and secure multi-party computation.

Investment and more efforts need to be coordinated in co-developing open science infrastructure for African health challenges. These efforts must be well coordinated and coherent among governments, researchers, and private and public institutions to facilitate open science and FAIR and secure data infrastructure to enable decision-making and regulation to effectively improve continental data-driven decision-making, implementation, and intervention that positively impact human health. Such efforts and investment shall leverage a multidisciplinary through continental consortia to co-develop a shared and resilient African health data space (AHDS). The AHDS can innovatively undertake the sharing of data for the delivery of healthcare, enhance a certified market for digital and electronic health record systems, and provide an efficient trustworthy, and framework for open science and FAIR health data for innovation, research, policy-making, and regulatory and industry activities across theAfricanUnion (AU) member countries members.

It is worth noting that developing DSPA where data assets are within SDEs on the continent can be crucial to remaining competitive in the global research market and can fasten and consolidate ongoing efforts in harnessing data science for health discovery and innovation (Fig. 1). To successfully influence change, there is also an unmet need to provide leadership to the local governments, nongovernmental organizations, community leaders, healthcare professionals, the domestic and international private/public sectors, universities, training institutions, international organizations and agencies, and faith-related communities, concerning health protection to improve data responsibly and governance to unlock the value of data sharing for health research in Africa.

Encouraging pan-African and global partnerships, and promoting transdisciplinary and cross-domain collaborations, especially those that go beyond academia in co-developing a resilient AHDS and DSPA will (1) enable a systematic benchmark for the state of health data science and health communities, evaluating progress and opportunities over the next several years; (2) highlight the potential of health data science to improve health in Africa and (3) advance the field, discuss opportunities and new trends, exchanging ideas and practices and stimulating new thinking. While the need for health data scientist talent in Africa is greater than ever around the world, the future of work and effort is changing. African health data science must reach its full potential and step up onto the world stage as the next frontier. Importantly, African governments and researchers should act by implementing the key recommendations proposed below.

Firstly, enabling research training of the next generation of highly skilled innovators and industry-ready graduates characterized by (1) Artificial Intelligence (AI) tools/method development and problem-solving skills, (2) understanding trustworthiness, ethical, societal, and legal implications of data science in health research, clinical and industrial settings, and (3) contributing to the African commercial development as well as the exploitation of data-driven decision and solutions for health challenges. Secondly, ensure that the next generation utilizes data, data sharing/standardization/harmonization, data availability, and data governance to address equitable and global health research and enable effective digitalization to benefit patients’ experience, public health, policymaking, and healthcare service providers. Thirdly, to unfold appropriate collaborative frameworks to consider trustworthiness, ownership, and ethical societal implications of data-driven health data science in research, clinical, and industrial settings. Fourthly, Africa requires a hands-on deck to ensure the continent is at the forefront, enabling impact, health innovations, and economic growth that can directly benefit from the data assets within AI and data sciences guaranteed by discoveries and investments. Fifthly, improving and investing in infrastructure and increasing engagement from a variety of academic, government, non-government, private sectors, and international actors will unlock not only African data assets, but also the health data science horizon in Africa.

Conclusion

This paper poses the critical importance of data-driven decision-making in achieving equitable and sustainable health impact. We have highlighted different data elements that can contribute to data-driven decisions in healthcare and shown examples of successes and limitations encountered in implementing and adopting data-driven decisions in healthcare interventions ranging from reducing malaria cases to crisis management of COVID-19. The paper also provides valuable lessons for regions facing similar resource constraints and illustrates how data can be utilized to identify health disparities, monitor progress toward health goals, and ensure that interventions are tailored to respond to pertinent challenges. The paper further advocates for a collaborative global approach, where data sharing and best practices transcend borders, capacity-building programs on data analysis, management, and use, the linkage between researchers and policymakers, putting proper measures to secure data, and fostering an environment for data-driven decisions. The insights gained from African experiences can inform global health strategies, ensuring that they are inclusive, adaptable, and responsive to the needs of diverse populations, thereby promoting equitable health outcomes worldwide.