Marco Ajelli

This study quantifies the impact of the distributions of the latent and infectious periods on epidemic dynamics. Using an individual-based modeling framework that incorporates transmission in social settings such as households, schools, and workplaces, we compare two model parameterizations of the latent and infectious periods: one based on the analysis of viral-load data (VLP) and a traditional one based on geometric distributions (GP). We analyze these through the lens of a “push-pull” mechanism, where competition for susceptible individuals favors earlier transmission events while longer infectiousness increases the opportunity for a transmission event to take place. Our findings show a clear decreasing trend in the latent period, infectious period, and generation time as the reproduction number increases, with more pronounced differences under GP compared to VLP. Moreover, timing and magnitude of the epidemic peak was highly dependent on the parameterization, varying more than 50% between the models. Our findings highlight that the average value and the shapes of the latent and infectious period distributions are fundamental drivers of epidemic spread. By providing insights into how these distributions shape epidemic spread, we provide a framework to refine model projections through data-driven parameterization.

Forecasting influenza hospitalizations informs public health preparedness, yet questions remain about which types of forecasts best guide action. We evaluate categorical trend forecasts, which communicate probabilities of upcoming increases or decreases in epidemic trajectories, submitted to CDC’s FluSight Forecasting Challenge between Fall-2024 and Spring-2026. Teams submitted probability distributions over five categories describing direction and magnitude of week-over-week changes in laboratory-confirmed influenza hospital admissions. We assessed performance using Ranked Probability Skill Score, Brier Skill Score, and measures of forecast-observation agreement. Most models outperformed an equal-probability baseline; the FluSight ensemble ranked among the top three in the 2024–25 and 2025–26 seasons. Forecasts were most accurate during stable periods and least during periods of rapid change, with most models underestimating observed trends. Conclusions were robust to choice of scoring metric and reference model. These results support categorical trend ensembles as an approach to communicating infectious disease forecasts that may inform public health decision-making.

Locally acquired arboviral infections are an increasing public health concern in the United States. Aedes aegypti and Culex quinquefasciatus , vectors of dengue virus and West Nile virus, respectively, are established in regions where local transmission has been reported. Seasonal variation in mosquito abundance affects vector density, human-vector contact, and arbovirus transmission risk. The aim of this study is to investigate seasonal variations in Ae. aegypti and Cx. quinquefasciatus population dynamics in Miami-Dade County, Florida; Maricopa County, Arizona; and Clark County, Nevada. Monthly relative abundance, average mosquitoes collected per trap-night, normalized abundance ratios, and seasonal-trend decomposition were used to evaluate species- and location-specific seasonal patterns. Mosquito population dynamics differed by species and location. In Miami-Dade County, the two species showed seasonal turnover, with Cx. quinquefasciatus peaking during winter and spring and Ae. aegypti peaking during summer. In Maricopa and Clark counties, both species peaked primarily between August and September. Species dominance also differed by site, with Ae. aegypti more abundant than Cx. quinquefasciatus in Maricopa County and Cx. quinquefasciatus more abundant than Ae. aegypti in Clark and Miami-Dade counties. Seasonal decomposition showed that Ae. aegypti peaked earlier in Miami-Dade County than in Maricopa and Clark counties, whereas Cx. quinquefasciatus showed a spring peak in Miami-Dade County and bimodal seasonal patterns in Maricopa and Clark counties. These results suggest that mosquito population dynamics are species- and location-specific and support the use of local surveillance data to guide the timing of mosquito control and arbovirus preparedness.

Conventional mosquito surveillance typically relies on contemporaneous data, making it challenging to anticipate future vector surges. To support proactive vector management, this study evaluates a multi-model forecasting framework designed to generate probabilistic 1- to 4-week-ahead forecasts of Aedes aegypti relative abundance per trap night. The framework was validated using multi-year surveillance data across four US jurisdictions spanning varied environments (from subtropical to temperate and arid). We found that an ensemble approach aggregating statistical and machine learning models generally achieved the best performance across all locations and forecast horizons. Relative forecast performance improved as the forecast horizon extended from 1 to 4 weeks ahead. The most challenging data to forecast were primarily restricted to low mosquito activity periods or atypical population peaks with unusual timing or magnitude. While full integration into routine vector management workflows represents a long-term process requiring operational adaptation, this work advances forecasting research and establishes a baseline for translating these approaches into real-time applications for public health authorities, with downstream effects in mitigating the risks of mosquito-borne diseases.

Mosquito-borne diseases are a major public health issue. The USA has experienced mosquito-borne disease outbreaks, including West Nile virus, Zika, dengue, and malaria, highlighting widespread risk across the country. In the USA, exposure to mosquito vectors primarily occurs outdoors. Individuals experiencing unsheltered homelessness are particularly vulnerable to mosquito-borne diseases due to their increased exposure to mosquito vectors, limited access to healthcare, and high prevalence of underlying conditions. This review aims to examine the literature on mosquito-borne disease infections among individuals experiencing homelessness in the USA between 1999 and 2024. Our search strategy used Boolean methods on PubMed, Scopus, and Web of Science to select relevant publications. Two queries were used: the first query focused on individuals experiencing homelessness and mosquito-borne diseases, and the second one focused on epidemiological analyses of mosquito-borne disease outbreaks in the USA. Only studies reporting at least one infection among individuals experiencing homelessness were included. Our search resulted in the identification of 347 studies. After screening them, eight were ultimately included in this review. Half of the studies identified West Nile virus among individuals experiencing homelessness, and the other half reported cases of malaria. These studies show that individuals who spend more time outdoors, including individuals experiencing homelessness, may be at a greater risk of MBDs compared to the general population. However, the full extent of mosquito-borne disease exposure among individuals experiencing homelessness remains unclear. With climate change and globalization, there is a consensus in the literature showing an increasing risk of mosquito-borne disease transmission in the next decades, and we have already started to witness this increasing trend. Moreover, the USA is experiencing a steep increase in the overall nationwide homeless population. Therefore, we argue that developing policies aimed at providing homeless populations with housing during mosquito-borne disease outbreaks could be instrumental in protecting this segment of the population and promoting health equity.

Infectious disease spread is a multiscale process composed of within-host (biological) and between-host (social) drivers and disentangling them from each other is a central challenge in epidemiology. Here, we introduce VIBES, a multiscale modeling framework that explicitly integrates viral dynamics based on patient-level data with population-level transmission on a data-driven network of social contacts. Using SARS-CoV-2 as a case study, we analyze three emergent epidemic properties, namely the generation time, serial interval, and presymptomatic transmission. First, we established a purely biological baseline, thus independent of the reproduction number ( R ), from the within-host model, estimating a generation time of 6.3 d for symptomatic individuals and 43.1% presymptomatic transmission. Then, using the full model incorporating social contacts, we found a shorter generation time (5.4 d at R = 3.0) and an increase in presymptomatic transmission (52.8% at R = 3.0), disentangling the impact of social drivers from a purely biological baseline. We further show that as pathogen transmissibility increases ( R from 1.3 to 6), competition among infectious individuals shortens the generation time and serial interval by up to 21% and 13%, respectively. Conversely, a social intervention, like isolation, increases the proportion of presymptomatic transmission by about 30%. Our framework also estimates metrics that are challenging to obtain empirically, such as the generation time for asymptomatic individuals (5.6 d; 95%CI: 5.1 to 6.0 at R = 1.3). Our findings establish multiscale modeling as a powerful tool for mechanistically quantifying how pathogen biology and human social behavior shape epidemic dynamics as well as for assessing public health interventions.

Understanding human mobility changes during epidemics is critical for predicting disease spread and planning interventions. However, capturing fine-scale dynamics is challenging. This study analyzed high-resolution human mobility patterns in Shanghai, China, during the 2022 SARS-CoV-2 Omicron BA.2 outbreak using large-scale anonymized cellular signaling data. We investigated mobility shifts across five distinct epidemic phases (pre-outbreak, targeted interventions, citywide lockdown, targeted lifting, and reopening) stratified by age, sex, and travel purpose. A comprehensive evaluation of four gravity and four radiation spatial interaction models was conducted to assess their ability to explain the observed mobility patterns under varying demographic and behavioral conditions. Population size and distance were found to be primary drivers of mobility, with notable variations across demographic groups and travel purposes. During the lockdown, mobility significantly decreased, particularly for social-related trips and the working-age population, while the effect of distance was substantially higher. Although mobility volumes recovered post-lockdown, a larger effect of distance persisted, implying long-lasting behavioral changes. Our comparative analysis showed that while several variants of gravity and radiation models captured overall patterns effectively, their performance was context-dependent, varying significantly across epidemic phases, population subgroups, and travel purposes. These findings highlight the importance of integrating different mobility models to capture the complex human mobility picture by different population groups during an epidemic outbreak. Overall, this study advances our understanding of behavioral adaptations during crises, enhancing preparedness and response planning.

Human contact patterns are a fundamental determinant of respiratory pathogen transmission, yet nationally representative post-pandemic data for the United States are limited. We present Epistorm-Mix, a 2024 probability-based online survey designed to be nationally representative by age, sex, race/ethnicity, household income, census region, and language. Respondents reported all person-to-person contacts from the preceding day, including the contact’s age and the setting (household, school, workplace, or community). We quantified contact numbers across demographic and social characteristics and used generalized additive models to test adjusted differences. We constructed age-stratified contact matrices and their setting-specific counterparts, benchmarking them against widely used synthetic matrices to simulate the spread of an epidemic of a respiratory pathogen. We found an average of 7.4 contacts per day, with significant heterogeneity across the population. Contact rates were highest among teenagers (15-19 years) and lowest among older adults (60+ years). In-person attendance at school or work was a major driver, resulting in 2-3 times more contacts than remote participation. We also identified key socioeconomic and demographic group heterogeneities: the number of contacts generally increased with household income, and Non-Hispanic (NH) Black and NH Asian individuals reported statistically significant fewer total contacts than NH White individuals. We found strong assortative mixing by age and demographic group with markedly distinct contact patterns across different social settings (households, schools, workplaces, and the community). While the study’s age-mixing patterns are broadly comparable to international findings, the identified demographic heterogeneities reflect social structures unique to the US, underscoring the need for country-specific data. Epistorm-Mix provides a nationally representative portrait of post-pandemic US contact patterns and serves as an open-access resource for modeling and public health planning.

Mosquito control strategies based on the mass release of modified males, such as genetically modified mosquitoes (GMMs), aim to suppress wild populations by impairing reproduction. Evaluating these interventions requires resource-intensive field trials, but a lack of standardized implementation practices, particularly regarding release ratios of modified males to wild female mosquitoes and trial timing, has led to variable outcomes. This study's objective is to propose a modeling tool for the "in-silico" simulation of trial designs before field implementation. To this aim, we developed an agent-based model of mosquito population dynamics. As a case study, we calibrated the model using 2019-2023 Aedes aegypti surveillance data from Miami-Dade County, Florida, and compared two GMM trial designs as illustrative examples. Our results show that depending on the implementation choices (e.g., trial start date and duration, release ratio), trials yield highly variable outcomes. For example, changing the start date while fixing all other implementation details can lead to effectiveness between 50% and 90%. Our findings suggest that "in-silico" simulation is a valuable tool for improving trial protocol design, allowing stakeholders to test strategies and reduce outcome uncertainty before committing to a fieldwork experiment.