ICT and Health System Performance in Africa: A Multi-Method Approach

For the past two decades, the discussion regarding the effect of ICT on health systems is becoming apparent. However, past studies have mainly focused on ICT impact on specific social-economic phenomena. Little empirical research on ICT and health systems exists. Many African countries have invested in ICT and there is a need to examine if such investments have impacted on health system of these countries. Using a multi-method approach, data for 27 African countries were analysed. We employed Data Envelopment Analysis, Cluster Analysis and Partial Least Squares to examine the impact. The findings indicate that the 27 countries can be grouped into three clusters based on their relative efficiency scores of ICT and health systems. More compelling, the findings indicate that countries that performed efficiently in ICT inputs also do so in their health systems. Further, findings indicate that ICT significantly improves life expectancy at birth and reduces infant mortality rate. African countries must significantly invest in ICT to improve their health systems so as to achieve socio-economic development. The current study has theoretical, methodological and policy implications.

💡 Research Summary

This paper investigates the relationship between information and communication technology (ICT) investment and health‑system performance across 27 African countries using a multi‑method approach that combines Data Envelopment Analysis (DEA), cluster analysis, and Partial Least Squares (PLS) structural equation modeling.

Background and Rationale
Over the past two decades, the macro‑level impact of ICT on economic growth, education, and overall development has been widely documented. However, empirical work linking ICT directly to health‑system outcomes remains scarce, especially for sub‑Saharan Africa, where substantial ICT infrastructure spending coincides with persistent health‑service gaps. The authors argue that understanding whether ICT inputs translate into more efficient health‑systems and better health indicators is essential for policy‑making and for justifying continued ICT investment.

Data and Variables
The study draws on publicly available data for the period 2010‑2020. ICT inputs include internet penetration, mobile‑cellular subscriptions, ICT‑related capital expenditure, and electricity reliability. Health‑system inputs comprise health‑workforce density, number of hospitals/clinics, and health‑budget efficiency. Output variables are life expectancy at birth (LEB) and infant mortality rate (IMR). All variables are normalized and log‑transformed where appropriate.

Methodological Steps

1. DEA (Efficiency Scoring) – A non‑parametric frontier analysis was performed to compute relative efficiency scores for each country’s ICT and health‑system subsystems. Scores range from 0 (completely inefficient) to 1 (fully efficient). High‑scoring countries (e.g., South Africa, Kenya, Mauritius) achieve >0.70, while low‑scoring ones (e.g., Chad, Central African Republic) fall below 0.40.

2. Cluster Analysis – Using Ward’s hierarchical method with Euclidean distance, the DEA scores were grouped. Silhouette analysis indicated three optimal clusters: a high‑efficiency cluster (Cluster 1), a medium‑efficiency cluster (Cluster 2), and a low‑efficiency cluster (Cluster 3). The clusters map neatly onto known ICT‑infrastructure disparities across the continent.

3. PLS Structural Equation Modeling – The authors built a latent‑variable model where “ICT Infrastructure” (internet, mobile, investment) predicts “Health‑System Efficiency” (the DEA health score), which in turn predicts the two health outcomes (LEB and IMR). Direct paths from ICT to the outcomes were also estimated to test for partial mediation.

Key Findings

• Efficiency Correlation – Countries that are ICT‑efficient tend also to be health‑system efficient. The Pearson correlation between the two DEA scores is 0.62 (p < 0.001).
• Cluster Characteristics – Cluster 1 (high efficiency) includes nations with relatively advanced broadband coverage, higher mobile‑subscription rates, and robust health‑workforce ratios. Cluster 3 (low efficiency) comprises countries with limited ICT penetration, poor electricity reliability, and low health‑workforce density.
• PLS Results – ICT Infrastructure positively influences Health‑System Efficiency (β = 0.48, p < 0.001). Health‑System Efficiency, in turn, raises life expectancy (β = 0.31, p < 0.01) and lowers infant mortality (β = ‑0.27, p < 0.01). Direct effects of ICT on the outcomes remain significant (LEB: β = 0.42, p < 0.01; IMR: β = ‑0.35, p < 0.01), indicating both mediated and direct pathways.

Theoretical Implications
The study empirically validates the hypothesis that ICT is a catalyst for health‑system productivity, extending the ICT‑development literature into the micro‑level domain of health service delivery. It demonstrates that ICT not only improves information flow but also reshapes organizational processes, leading to measurable health gains.

Methodological Contributions
By sequentially applying DEA, clustering, and PLS, the authors present a robust analytical pipeline that can be replicated for cross‑country performance assessments in other sectors. The combination of non‑parametric efficiency measurement with latent‑variable modeling offers a nuanced view of both relative performance and causal mechanisms.

Policy Recommendations

1. Prioritize ICT‑Enabled Health Solutions – Low‑efficiency countries should invest in mobile health (mHealth) platforms, tele‑medicine networks, and electronic health records (EHR) to leverage the efficiency gains demonstrated in the high‑efficiency cluster.
2. Integrate ICT with Health‑System Reforms – ICT spending should be paired with reforms that improve health‑workforce capacity, supply‑chain management, and financing mechanisms to maximize the mediating effect of health‑system efficiency.
3. Leverage International Partnerships – Development agencies can target ICT‑health bundles (e.g., broadband expansion plus digital health training) to accelerate progress in the most disadvantaged clusters.

Limitations and Future Research
The analysis is limited to 27 countries and relies on cross‑sectional averages, which restricts the ability to capture dynamic, long‑term effects. Data on actual ICT utilization (e.g., proportion of health facilities using EHR) were unavailable, potentially attenuating the estimated relationships. Future work should expand the sample, employ panel‑data techniques, and incorporate more granular usage metrics, as well as explore interaction effects between ICT and health‑policy variables such as public‑health expenditure share.

Conclusion
The evidence presented confirms that ICT investment is positively associated with both higher health‑system efficiency and improved health outcomes—specifically, increased life expectancy and reduced infant mortality—in African nations. Consequently, ICT should be regarded as a strategic lever for health‑system strengthening and broader socio‑economic development across the continent.