Clusters of African countries based on the social contacts and associated socioeconomic indicators relevant to the spread of the epidemic

Abstract Introduction It is well known that social contact patterns differ from country to country. This variation coincides with significant socioeconomic heterogeneity that complicates the design of effective nonpharmaceutical interventions. This study examined how socioeconomic heterogeneity in selected African countries might be factored in to explain better patterns of social contact mix between countries. Methods We used a standardized contact matrix for 32 African countries, estimated in (Prem et al. in PLoS Comput Biol 17(7):e1009098, 2021). We scaled the matrices using an epidemic model from (Röst et al. in Viruses 12(7):708, 2020). We also analyzed aggregated data from the World Bank country website. The data includes 28 variables; social, economic, environmental, institutional, governance, health and well-being, education, gender inequality, and other development-related indicators that describe countries. Principal component analysis was used to visualize socioeconomic similarities between countries and identify the indicators for maximum variation. The $(2D)^{2} PCA$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msup> <mml:mrow> <mml:mo>(</mml:mo> <mml:mn>2</mml:mn> <mml:mi>D</mml:mi> <mml:mo>)</mml:mo> </mml:mrow> <mml:mn>2</mml:mn> </mml:msup> <mml:mi>P</mml:mi> <mml:mi>C</mml:mi> <mml:mi>A</mml:mi> </mml:math> approach was used to reduce the dimension of synthetic contact matrices for each country to avoid the dimensionality curse. Then, hierarchical aggregate clustering was used to identify groups of countries with similar social patterns, taking into account the country’s socioeconomic performance. Results Our model yielded four meaningful clusters, each with a few distinguishing features. Social contacts varied between groups, but were generally similar within each set. The socioeconomic performance of the country influenced the clusters. Conclusions Our results suggest that integrating socioeconomic factors into social contacts can better explain infectious disease transmission dynamics and that similar interventions can be implemented in countries within the cluster