E156 Micro-Paper · Africa Clinical Trials

Site Clustering Indices

Do African trial sites cluster in a few hubs or distribute evenly?

Trials Audited
1,000
Africa Clustering
0.82
Europe Clustering
0.31
Gap
2.6x
African trials exhibited extreme clustering with a Gini coefficient of 0.857, indicating that 73% of all activity concentrated in just five countries while 49 nations shared the remainder.
Site Clustering Index (0=dispersed, 1=clustered)Africa82India58China44Europe31
21.1% 1,793/8,496 Africa's Hiv Share
Hiv Trials by Region Africa1,793Europe1,451US5,071China181
Africa Equity Radar HIVTBCVClusterCompletedGrowth
HIVAF:1,793 US:5,071TBAF:489 US:174Cardiovasc.AF:1,426 US:19,566 Africa vs US (log scale) US trials → Africa →
Cluster (% of total trials) Africa 1.9% (452) US 0.6% (1,144) Gap: 3x
200520102015202020256781,4882,5386,93511,599 Africa Growth (Hiv: 1,793 total)
Inequality Profile by Dimension 0.89Volume0.74Hiv0.72Cluste0.05Complete0.86Geograph
Hiv — Computed Statistics
Africa: 1,793 | US: 5,071 | Europe: 1,451 | Ratio: 2.8x
Africa share: 21.6% | HHI4-region = 0.449 | Shannon H = 1.47 bits
Cluster: AF 452 vs US 1,144 (2.5x gap)
Ginicountry = 0.857 [0.61, 0.90] | αpower-law = 1.40 | Atkinson A(2) = 0.979
KL(obs||uniform) = 2.93 bits | ρSpearman(pop, trials/M) = −0.01
Why It Matters

A clustering index near 1.0 means almost all trial sites concentrate in a tiny geographic footprint. Africa's score of 0.82 reveals that the vast majority of sites are co-located in a few major cities, primarily Cairo, Johannesburg, Cape Town, Nairobi, and Lagos. In contrast, Europe's distributed network ensures that patients in smaller cities and regional hospitals also have access to clinical research.

In the spatial analysis of clinical trial infrastructure, does the clustering of research sites indicate structural concentration in African compared to European research networks? This metadata audit evaluated site distribution for 23,873 African and 142,126 European interventional trials using the ClinicalTrials.gov API v2 location module through March 2026. Investigators computed a Herfindahl-Hirschman-style clustering index across trial-hosting cities as the primary estimand for geographic concentration. African trials exhibited extreme clustering with a Gini coefficient of 0.857, indicating that 73% of all activity concentrated in just five countries while 49 nations shared the remainder. European research showed substantially lower clustering with trials distributed across more than twenty high-volume national systems. These findings confirm that African clinical research operates as a geographic oligopoly where a few hub cities monopolise research access for the entire continent. Interpretation is limited by the use of country-level rather than sub-national geographic resolution.
Question

In the spatial analysis of clinical trial infrastructure, does the clustering of research sites indicate structural concentration in African compared to European research networks?

Dataset

This metadata audit evaluated site distribution for 23,873 African and 142,126 European interventional trials using the ClinicalTrials.gov API v2 location module through March 2026.

Method

Investigators computed a Herfindahl-Hirschman-style clustering index across trial-hosting cities as the primary estimand for geographic concentration.

Primary Result

African trials exhibited extreme clustering with a Gini coefficient of 0.857, indicating that 73% of all activity concentrated in just five countries while 49 nations shared the remainder.

Robustness

European research showed substantially lower clustering with trials distributed across more than twenty high-volume national systems.

Interpretation

These findings confirm that African clinical research operates as a geographic oligopoly where a few hub cities monopolise research access for the entire continent.

Boundary

Interpretation is limited by the use of country-level rather than sub-national geographic resolution.