Conducted jointly by UGB and IRD [Île-de-France], this research models bacterial meningitis in Senegal.
The Essentials: Conducted jointly by UGB and IRD [Île-de-France], this research models bacterial meningitis in Senegal.
Papa Massar Niane, researcher at UMMISCO (Université Gaston Berger de Saint-Louis Sénégal).
Thesis defended in 2023 at the doctoral school École doctorale Informatique, télécommunications et électronique de Paris (1992-....).
This research is the product of an international joint supervision between several partner institutions.
Listed in the ABES/STAR network, this thesis meets the rigor criteria of French higher education.
Context and Research Problem
Bacterial meningitis is a major public health challenge, particularly in the Sahel region, notably in Senegal. Meningitis epidemics are influenced by a multitude of factors, including environmental, climatic, and societal elements. This multiscale complexity makes it essential to develop predictive models capable of integrating these various parameters in order to anticipate and manage epidemics effectively. Papa Massar Niane's thesis focuses on this issue and proposes the MenAfriSIM™ model, an innovative multi-agent approach to modelling meningitis transmission.
Methodology
The MenAfriSIM™ model was designed to analyze the spread of meningitis in Senegal, taking climatic and environmental factors into account. One key element of this model is the COefficient of Meningitis Invasion and Development for Environmental eXposure (COMIDEX), which integrates spatialized environmental data derived from remote sensing. This coefficient makes it possible to assess the impact of temperature and dust levels on disease transmission.
MenAfriSIM™ also uses a spatial interaction model to account for inter-urban mobility, a crucial factor in meningitis transmission dynamics. The model was tested on data from the 2012 season, a period marked by a record number of meningitis cases. Evaluating the model made it possible to analyze case variability in relation to demographic and environmental factors.
Key Findings
The results obtained from MenAfriSIM™ indicate that more than 50% of the total variability in meningitis cases is explained by the model, with a coefficient of determination R² of 0.53. In addition, nearly a third of case variability can be attributed to temperature and dust, with an R² of 0.29. These results underscore a strong correlation between the number of meningitis cases and population density, suggesting that the municipalities most affected by the disease are generally located in the most densely populated areas.
Analyses also revealed that northern Senegal presents the most unfavorable climatic and environmental conditions, corresponding to a higher incidence of meningitis cases. The notion of a "meningitis tri-zone" illustrates a risk gradient that decreases from north to south of the country. These observations are corroborated by existing literature on climatic conditions in Senegal and by the exploration of 2013 season data.
Discussion and Outlook
The MenAfriSIM™ model represents a significant advance in understanding bacterial meningitis in Senegal. By integrating environmental data and accounting for spatial interactions, it allows for a more refined analysis of the factors contributing to the spread of meningitis. The emphasis on monitoring the country's northern regions is particularly relevant, given that these areas constitute critical points for meningitis risk.
Integrating the model's results into early warning systems could facilitate a faster response to epidemics. However, further research over an extended period is needed to refine conclusions and develop robust policy recommendations. Accounting for the time it takes meningitis to spread between zones (estimated at 2 to 3 weeks) is also essential for improving public health interventions.
The results of this thesis open up interesting prospects for epidemiological modelling and public health in Senegal. They underscore the need for an integrated approach that combines scientific research, health surveillance, and appropriate health policies to effectively address the challenges posed by bacterial meningitis.
Key Data
- 53%: MenAfriSIM™ explains 53% of the variability in meningitis cases in Senegal.
- 29%: Nearly a third of case variability is attributed to temperature and dust.
Accéder à l'étude complète
Sources and Access
Papa Massar Niane. Modélisation de la méningite bactérienne dans l'interface Environnement-Climat-Société par approche multi-agents : cas d'application au Sénégal. Modélisation et simulation. Sorbonne Université; Université Cheikh Anta Diop (Dakar, Sénégal ; 1957-..), 2023. Français. ⟨NNT : 2023SORUS535⟩. ⟨tel-04457230⟩