A climate-informed dengue transmission model with Bayesian decision support.

BackgroundAs the arboviral diseases dengue, chikungunya and Zika emerge in the Americas, so does the need for sustainable vector control policies. To successfully achieve mosquito control, joint efforts of both communities and governments are essential. This study investigates this important, but by-and-large neglected topic.MethodsIn June and July 2015, a cross-sectional mixed methods study applying a survey questionnaire (response rate of 82.5%; n = 339), in-depth interviews (n = 20) and focus group discussions (n = 7; 50 participants) was performed in Curaçao. The study was designed based on an integrated theoretical framework of the Health Belief Model and the Theory of Planned Behaviour.ResultsParticipants showed a good knowledge of, and a high-level performance of mosquito breeding site control (MBSC) practices. Personal protection against mosquitoes (e.g. topical repellents) was perceived as relatively less effective thus practiced to lower extent compared to MBSC practices (i.e. larval source management). A lower intention to perform MBSC was independently associated with: (i) satisfaction on governmental MBSC (P = 0.012); (ii) barriers to perform MBSC practices, i.e. ‘Government doesn’t control other breeding sites’ (P = 0.005), ‘Don’t know how to control breeding sites’ (P = 0.041), and ‘a mosquito does not transmit dengue’ (P = 0.016), (iii) attitudes towards MBSC (P = 0.001) and self-efficacy (person’s perceived ability to act) to perform MBSC (P = 0.002). Mixed-methods evidence highlights three possible ways of improving community participation in MBSC. First, it highlights the need for ongoing media coverage, targeting (i) communities’ perceptions on transmission routes of dengue and chikungunya, and (ii) presence of car tires in yards. Secondly, it shows that promotion of governmental activities in MBSC can enhance MBSC of communities, if people develop a sense of responsibility to perform MBSC at their own properties. Thirdly, this study describes the presence of key persons in communities, who could be engaged in mosquito control policies to improve MBSC in neighbourhoods.ConclusionThis study reveals gaps between policy and communities’ lived realities. These gaps might be overcome with the proposed interventions, resulting in a higher performance of MBSC in the community in Curaçao. Furthermore, this study shows how interdisciplinary mixed methods research can provide important, comprehensive, and in-depth insights to inform mosquito control policies.

Characterizing persistent malaria transmission that occurs after the combined deployment of indoor residual spraying (IRS) and long-lasting insecticidal nets (LLINs) is critical to guide malaria control and elimination efforts. This requires a detailed understanding of both human and vector behaviors at the same temporal and spatial scale. Cross-sectional human behavior evaluations and mosquito collections were performed in parallel in Magude district, Mozambique. Net use and the exact time when participant moved into each of five environments (outdoor, indoor before bed, indoor in bed, indoor after getting up, and outdoor after getting up) were recorded for individuals from three different age groups and both sexes during a dry and a rainy season. Malaria mosquitoes were collected with CDC light traps in combination with collection bottle rotators. The percentage of residual exposure to host-seeking vectors that occurred in each environment was calculated for five local malaria vectors with different biting behaviors, and the actual (at observed levels of LLIN use) and potential (i.e. if all residents had used an LLIN) personal protection conferred by LLINs was estimated. Anopheles arabiensis was responsible for more than 74% of residents’ residual exposure to host-seeking vectors during the Magude project. The other four vector species (An. funestus s.s., An. parensis, An. squamosus and An. merus) were responsible for less than 10% each. The personal protection conferred by LLINs prevented only 39.2% of the exposure to host-seeking vectors that survived the implementation of both IRS and LLINs, and it differed significantly across seasons, vector species and age groups. At the observed levels of bednet use, 12.5% of all residual exposure to host-seeking vectors occurred outdoor during the evening, 21.9% indoor before going to bed, almost two thirds (64%) while people were in bed, 1.4% indoors after getting up and 0.2% outdoor after leaving the house. Almost a third of the residual exposure to host-seeking vectors (32.4%) occurred during the low transmission season. The residual bites of An. funestus s.s. and An. parensis outdoors and indoor before bedtime, of An. arabiensis indoors when people are in bed, and of An. squamosus both indoors and outdoors, are likely to have sustained malaria transmission throughout the Magude project. By increasing LLIN use, an additional 24.1% of exposure to the remaining hosts-seeking vectors could have been prevented. Since An. arabiensis, the most abundant vector, feeds primarily while people are in bed, increasing net use and net feeding inhibition (through e.g. community awareness activities and the selection of more effective LLINs) could significantly reduce the exposure to remaining host-seeking mosquitoes. Nonetheless, supplementary interventions aiming to reduce human-vector contact outdoors and/or indoors before people go to bed (e.g. through larval source management, window and eave screening, eave tubes, and spatial repellents) will be needed to reduce residual exposure to the outdoor and early biting An. funestus s.s. and An. parensis.

The COVID-19 pandemic underscored the importance of developing holistic frameworks that incorporate a variety of external factors into infectious disease modeling, in order to guide public health responses. While traditional models often focus on isolated aspects of disease dynamics, this thesis proposes a unified infectious disease ecosystem framework that captures the complex interplay of non-pharmaceutical interventions (NPIs), pharmaceutical interventions (PIs), human mobility patterns, weather conditions, and pathogen variants. By bridging compartmental models such as SEIR (susceptible–exposed–infected–recovered) with machine learning techniques, this research addresses critical gaps in the literature and advances both theoretical understanding and practical applications in pandemic modeling. The thesis is structured around four interconnected research objectives, each contributing to the unified framework. First, it quantifies the waning effectiveness of NPIs over time by introducing a modified SEIR model, SVEIC-NLC, which embeds Newton’s Law of Cooling (NLC) to represent NPI fatigue; the model is validated across multiple countries. Second, it explores the behavioral heterogeneity induced by vaccination certificate (green pass) policies through the SEIQRD^2 model, demonstrating how these policies influence group dynamics and transmission rates. Third, it examines the critical role of human mobility in disease transmission by integrating mobility data with NPIs and PIs in a hybrid SEIR-neural network framework, NeuralSEIR, to reveal its influence on pandemic progression. Finally, the thesis develops SEIR-VRNN, a comprehensive model that simultaneously incorporates NPIs, PIs, human mobility, and weather conditions using a variational recurrent neural network (VRNN), enabling robust forecasting and scenario analysis. By synthesizing these contributions, the thesis advances a unified knowledge framework for infectious disease modeling that integrates dynamic external factors, balances predictive accuracy with interpretability, and provides actionable insights for public health planning. Validated using real-world data from diverse regions, the proposed models demonstrate scalability, adaptability, and relevance for managing current and future pandemics. This work lays the foundation for a new generation of integrative models capable of addressing the multifaceted challenges of infectious disease dynamics.

The spread of resistant bacteria in hospitals is an increasing problem worldwide. Transfers of patients, who may be colonized with resistant bacteria, are considered to be an important driver of promoting resistance. Even though transmission rates within a hospital are often low, readmissions of patients who were colonized during an earlier hospital stay lead to repeated introductions of resistant bacteria into hospitals. We developed a mathematical model that combines a deterministic model for within-hospital spread of pathogens, discharge to the community and readmission, with a hospital–community network simulation of patient transfers between hospitals. Model parameters used to create the hospital–community network are obtained from two health insurance datasets from Germany. For parameter values representing transmission of resistant Enterobacteriaceae, we compute estimates for the single admission reproduction numbers RA and the basic reproduction numbers R0 per hospital–community pair. We simulate the spread of colonization through the network of hospitals, and investigate how increasing connectedness of hospitals through the network influences the prevalence in the hospital–community pairs. We find that the prevalence in hospitals is determined by their RA and R0 values. Increasing transfer rates between network nodes tend to lower the overall prevalence in the network by diluting the high prevalence of hospitals with high R0 to hospitals where persistent spread is not possible. We conclude that hospitals with high reproduction numbers represent a continuous source of risk for importing resistant pathogens for hospitals with otherwise low levels of transmission. Moreover, high risk hospital–community nodes act as reservoirs of pathogens in a densely connected network.

This paper presents a unified analytical and computational framework for the study of typhoid fever transmission dynamics governed by a Caputo fractional-order compartmental model of order κ∈(0,1]. The population is stratified into five epidemiological classes, namely susceptible (S), asymptomatic (A), symptomatic (I), hospitalised (H), and recovered (R), and the governing system explicitly incorporates asymptomatic transmission, treatment dynamics, and temporary immunity with waning. The use of the Caputo fractional derivative is motivated by the well-documented existence of chronic asymptomatic Salmonella Typhi carriers, whose heavy-tailed sojourn times in the carrier state are naturally encoded by the Mittag–Leffler waiting-time distribution arising from the fractional operator. A complete qualitative analysis of the fractional system is carried out: the basic reproduction number R0 is derived via the next-generation matrix method; local and global asymptotic stability of both the disease-free equilibrium E0 (when R0≤1) and the endemic equilibrium E* (when R0>1) are established using fractional Lyapunov theory and the LaSalle invariance principle; and the normalised sensitivity indices of R0 are computed to identify transmission-amplifying and transmission-suppressing parameters. Existence, uniqueness, and Ulam–Hyers stability of solutions are established via Banach and Leray–Schauder fixed-point arguments. To complement the analytical results, a fractional physics-informed neural network (PINN) framework is developed to simultaneously reconstruct compartmental trajectories and identify unknown biological parameters from sparse synthetic observations. PINN embeds the L1-Caputo discretisation directly into the training residuals and employs a four-stage Adam–L-BFGS optimisation strategy to recover five trainable parameters Θ = {ϕ,μ,σ,ψ,β} across three fractional orders κ∈{1.0,0.95,0.9}. The estimated parameters show strong agreement with the true values at the classical limit κ=1.0 (MAPE=2.27%), with the natural mortality rate μ recovered with APE≤0.51% and the transmission rate β with APE≤3.63% across all fractional orders, confirming the structural identifiability of the model. Pairwise correlation analysis of the learned parameters establishes the absence of equifinality, validating that β can be reliably included in the trainable set. Noise robustness experiments under Gaussian perturbations of 1%, 3%, and 5% demonstrate graceful degradation (MAPE: 0.82%→3.10%→7.31%), confirming the reliability of the proposed framework under realistic observational conditions.

Spatiotemporal transmission dynamics for influenza disease in a heterogenous environment

This study establishes a novel time-delay fractional SEIHR infectious disease model to investigate the effects of saturated incidence rates and time delays on different populations, including susceptibles, infected individuals, recovered individuals, and latent infected individuals. First, the existence and boundedness of the model’s solutions are verified, confirming its well-posedness. Subsequently, the existence of equilibria is analyzed, and the impact of parameter variations on the system is explored by examining the equilibria ϵ0 and ϵ∗, as well as the basic reproduction number R0. Additionally, the global dynamics of the equilibria are further analyzed using the Lyapunov method, while Hopf bifurcation theory is applied to explore the conditions under which the system shifts from stability to oscillatory behavior. Numerical simulations further validate the theoretical analysis, showing that time-delay effects significantly influence the system’s responsiveness and the rate of disease transmission. Moreover, when the time delay τ crosses the critical threshold τ0, the system exhibits periodic oscillations. By predicting periodic fluctuations and incorporating memory effects and persistent influences, we can better control epidemics, emphasizing the importance of time-delay adjustments and enhancing the system’s biological realism.

Huanglongbing (HLB) is a bacterial disease that affects citrus trees worldwide. We present an innovative approach for identifying optimal control and risk measures for HLB in citrus orchards. Our method is based on a mathematical model that incorporates the number of roguing trees and a logistic growth model for the dynamic of the Asian Citrus Psyllid (ACP), the primary vector for HLB transmission. We derive an expression for: (1) the basic reproduction number R0; (2) the final size for the number of roguing trees; and (3) the transmission risk. The above let us propose a difference map equation that assesses this final size with a low computational cost. We use this difference map in an evolutionary algorithm to identify the most effective combination of control parameter values for reducing HLB transmission, including the timing and frequency of roguing and the use of insecticides. In this sense, we propose two control strategies, which we called tree-centered and vector-centered.

Bovine tuberculosis (bTB) is a bacterial and zoonotic disease which is transmitted through consumption of unpasteurized milk and uncooked meat and inhalation of aerosols. In this paper, a deterministic mathematical model is formulated to study the transmission dynamics of bTB in humans and animals. The basic reproduction number R0 is computed to determine the behavior of the disease. Stability analysis shows that there is a possibility for disease-free equilibrium and endemic equilibrium to coexist when R0 = 1. To determine parameters which drive the dynamics of bTB, we performed sensitivity analysis. The analysis shows that the rate at which dairy products are produced, the rate of transmission of bTB from animal to animal, and the rates at which human acquires bTB from infectious dairy products and animals drive the transmission of bTB. However, the disease decreases as the rate of consumption of dairy products decreases. To control bTB, education campaign, inspection of dairy products, treatment of infected humans, and quarantine of infected animals are recommended.

Protecting individuals and households against mosquito bites with long-lasting insecticidal nets (LLINs) or indoor residual spraying (IRS) can suppress entire populations of unusually efficient malaria vector species that predominantly feed...

Background: This study proposes a simplified SIR-type compartmental mathematical model tailored to the HIV epidemic in Mizoram, a small state in India and is one of the highest-prevalence regions in the country. The model includes key processes such as antiretroviral treatment (ART), behaviour-change dynamics, recruitment, and natural mortality.Aims and Objectives: This study emphasizes the need for early treatment initiation and promotion of risk-reducing behaviour to control and curb the prevalence of HIV/AIDS especially in a highly prevalent region like Mizoram and may serve as a practical and flexible resource for policymakers and public health planners to evaluate the effectiveness of early interventions in high-prevalence populations.Materials and Methods: The basic reproduction number R0 is derived using the next-generation matrix method and local as well as global stability conditions for both disease-free and endemic equilibria are established via Lyapunov functions. Numerical simulations demonstrate the impact of transmission rate (β), treatment uptake (γ), and behavior-modification rate (µ) on epidemic outcomes.Results and Conclusion: It is observed that when γ and µ i.e., rate of treatment uptake and behaviour-modification rate respectively are increased effectively, R0 decreases and when R01, the endemic equilibrium remains, indicating persistent transmission in the population. Although µ does not directly reduce transmission, it significantly boosts the recovered class when increased.

Mathematical model with sensitivity analysis and control strategies for marijuana consumption

In this paper, we investigate a Vector‐Borne disease model with nonlinear incidence rate and 2 delays: One is the incubation period in the vectors and the other is the incubation period in the host. Under the biologically motivated assumptions, we show that the global dynamics are completely determined by the basic reproduction number R0. The disease‐free equilibrium is globally asymptotically stable if R0≤1; when R0>1, the system is uniformly persistent, and there exists a unique endemic equilibrium that is globally asymptotically. Numerical simulations are conducted to illustrate the theoretical results.

Estimating risk of long COVID using a Bayesian network-based decision support tool.

ABSTRACTIn this article, the quality of data produced by national statistical institutes and by governmental institutions is considered. In particular, the problem of measurement error is analyzed and an integrated Bayesian network decision support system based on non-parametric Bayesian networks is proposed for its detection and correction. Non-parametric Bayesian networks are graphical models expressing dependence structure via bivariate copulas associated to the edges of the graph. The network structure and the misreport probability are estimated using a validation sample. The Bayesian network model is proposed to decide: (i) which records have to be corrected; (ii) the kind and amount of correction to be adopted. The proposed correction procedure is applied to the Banca d’Italia Survey on Household Income and Wealth and, specifically, the bond amounts are analyzed. Finally, the sensitivity of the conditional distribution of the true value random variable given the observed one to different evidence configurations is studied.