Digital Health and Artificial Intelligence for Public Health Systems in Africa: A Narrative Review

Authors: Abdi, Yusuf Hared; Abdullahi, Yakub Burhan; Abdi, Mohamed Sharif; Bashir, Sharmake Gaiye; Ahmed, Naima Ibrahim

Source journal: Health science reports (2026-11-01)

DOI: 10.1002/hsr2.73522

Abstract

BACKGROUND AND AIMS: Digital health and artificial intelligence (AI) are promoted as ways to strengthen African public health systems, but the evidence is fragmented across technologies, settings, and levels of rigor. This review synthesized evidence on mobile health (mHealth), telemedicine, and AI in Africa to examine what works, where and under which conditions, and the ethical, regulatory, and equity implications. METHODS: Narrative review guided by the SANRA criteria. PubMed/MEDLINE and Google Scholar were searched for literature published from January 2010 to May 2025, supplemented by gray literature from WHO, Africa CDC, and the African Union; quantitative claims were re-verified against original sources in October 2026. Evidence was organized using a framework linking three technology tiers to health system building blocks and appraised by study design and, for AI, by validation stage. RESULTS: The most consistent evidence concerned text messaging: in African randomized trials, one-way SMS improved appointment attendance (odds ratio 2.03) and two-way SMS improved medicine adherence (risk ratio 1.14). Programs that reached national scale were government-led and aligned with existing information systems, and in Rwanda were effective only alongside wider health system support. Telemedicine evidence was largely descriptive; AI evidence consisted mainly of development, and accuracy studies; computer-aided tuberculosis detection had pooled sensitivity of 0.87 and specificity of 0.74 in studies at high risk of bias, and patient-outcome evidence was scarce. Although 38 of 47 WHO African Region Member States reported national digital health strategies, only 9 of 34 publicly available strategies were rated strong. Connectivity, electricity, training, data protection, and donor dependence were barriers, distributed unequally by education, location, and gender. CONCLUSION: Simple, interoperable, government-led tools have the strongest evidence of benefit at scale in Africa. AI requires prospective, locally validated evaluation of patient outcomes, and investment should prioritize governance, interoperability, inclusion, and sustainable domestic financing.

Read the full text

Related topics

Related open access articles

Abstract and bibliographic data from PubMed (U.S. National Library of Medicine). Copyright of the abstract belongs to the original publisher or authors.