Application of data science in health research to benefit the African population

A recent Nature Communications study discusses the importance of data science for health research that would significantly benefit the African population.

Study: The promise of data science for health research in Africa. Image Credit: NicoElNino / Shutterstock.com

How has data science revolutionized scientific research?

The application of data science is associated with the processing of massive datasets using high-performance computational infrastructure. These datasets are derived from public, personal, and commercial domains.

Data science has significantly helped research in various fields by aiding in the development of multiple novel interventions and strategies that have considerable social benefits. 

Data scientists process a massive amount of information retrieved from healthcare systems, shopping records, smartphones, social media postings, and wearable devices using novel algorithms. Data analysis helps generate new insights and generalizable knowledge.

Analysis of large datasets has positively helped in bio-preparedness, monitoring, and formulation of response strategies to combat infectious disease outbreaks for both plants and animals. For example, the Geographical Information System (GIS) data is used to map spatial variations in accordance with the incidence, prevalence, and outcomes of diseases. This data is also used to assess the effectiveness of the strategy implemented by healthcare systems post-disease outbreak.

Data science has also been used to reduce fraud and corruption, improve supply chain management to prevent a shortage of products, and detect fake pharmaceutical products. In health research, data science is involved with the systematic collection, generation, storage, processing, management, visualization, analyses, interpretation, and communication of health-related data. This analysis provides actionable insights to prevent or manage a disease outbreak.

Data science in health research in Africa

The implementation of data science in healthcare research would likely solve many challenges faced by the people of Africa. Although Africa constitutes about 17% of the world’s population, it bears 25% of the world’s disease burden.

To further complicate matters, Africa lacks an appropriate number of healthcare workers and infrastructure. These social and clinical adversities can be attributed to the lack of properly trained personnel, economic and social instability, and poor funding.

Data scientists have highlighted that African countries require innovative data science tools and strategies to overcome challenges linked to differential climate and disease manifestations from other parts of the world.

Currently, there remains a significant gap in the dataset that adequately represents the African population. The underrepresented dataset leads to the development of unstable and inaccurate models and algorithms to analyze the African population.

Academic institutions, governments, African researchers, and the public sector actively use data science for discoveries, research, and formulation of strategies to manage infection outbreaks. Notably, many of these data science tools used have been developed and validated outside Africa.

How to improve data science health research in Africa

More infrastructure, training programs, scientific conferences, and international collaborations are needed to aid in the generation of high-quality datasets, which will help develop stable and accurate data science models. These initiatives will effectively close the gap in data science between Africa and high-income countries (HIC).

In 2022, 20 grants worth $74.5 million USD were awarded in the Harnessing Data Science for Health Discovery and Innovation in Africa (DS-I Africa) program. The main objective of this program is to improve data science and health research in Africa.

Prior to the DS-I Africa program, the Human Heredity & Health in Africa (H3Africa) program enabled the establishment of new scientific collaborations to develop genomics research infrastructure. This helped expand the African genomics research ecosystem.

Despite the implementation of strict health research ethics infrastructure by African governments and institutions, many challenges regarding the quality of informed consent, benefit-sharing, autonomy, data sharing, privacy, and exploitation prevail. These unresolved challenges and controversies must be addressed to benefit data science health research.

It is imperative to build and maintain national and institutional health research ethics infrastructure in Africa. Information will support the economic and intellectual capital of institutions and countries.

Research ethicists must bring local research and data science communities together to evolve data science health research strategies. Well-designed, properly funded, and relatively long training programs are crucial for data science health research ethics.

Multilateral agencies, including the United Nations, governments, researchers, advocates, bioethicists, and other stakeholders, must work with African institutions to develop guidelines for data science health research to optimize benefits for the global population. 

Data science technologies have been associated with the production of algorithmic bias by replicating and reinforcing societal biases. Algorithmic biases, combined with the poor equity and diversity in the foundational datasets, cause algorithmic deprivation, distortion, and discrimination. Issues related to data colonization and extractive research must be addressed to reduce biases.

In the future, novel strategies for data science health research must be developed with targeted objectives, which would significantly benefit different health research ecosystems in Africa.