CIChristopher Irrgang

Christopher Irrgang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Christopher Irrgang has a total of 140 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Regional heterogeneity in the association between extreme temperatures and emergency hospitalizations in Germany

    by Maria Gueltzow, Risto Conte Keivabu, Christopher Irrgang, Marcus Ebeling

    Background: The association between extreme temperature and emergency hospitalizations may differ by regional socioeconomic deprivation and its intersection with age and urbanization. This study investigates how regional characteristics modify the relationship between temperature and emergency hospitalizations in Germany. Methods: We analyzed the weekly emergency hospitalization rate by German districts (2005-2021) using the German Diagnosis-Related Groups Statistic, linked with high-resolution temperature data from the European gridded daily observational dataset (E-OBS). We fit a fixed effects model and interact regional socioeconomic deprivation with the weekly number of days falling into one of eleven temperature bins, adjusting for seasonality. Analysis was stratified by urbanization and age group. Results: We observe 112,909,391 emergency hospitalizations over the study period. Among rural districts, an additional heat day is associated with 17.6 (95% CI: 12.3–23.1) additional emergency cases per 10,000 in the least-deprived districts, with no change in the most-deprived districts. Among ages 65+, an additional heat day is associated with the largest increase in emergency hospitalizations in medium-deprived compared to least-deprived districts. Conclusions: Our study finds that the temperature-emergency hospitalization relationship varies across German regions, with differences only becoming visible if regions are profiled by several characteristics. This highlights the need for an intersectoral approach to adequately address the differential impact of climate stressors on inpatient care uptake.

  • Open Access

    High-Dimensional Heterogeneity in Heat-Related Health Impacts Across German Districts

    by Maria Gueltzow, Tarik Benmarhnia, Marcus Ebeling, Christopher Irrgang

    Background: Multiple factors can drive heat susceptibility but are typically analyzed individually. This study investigates high-dimensional heterogeneity across multiple domains in the effect of hot days on emergency hospitalizations across German districts. Methods: We obtained emergency hospitalization rates per 1,000 in 3-day windows for the summers of 2005-2023 from the German Diagnosis-Related Groups-Statistic, linked with spatio-temporal temperature information. A hot day was defined as at least one day above the district-level 98th percentile heat index. Multiple characteristics of the demographic, health/social, and environmental domains were selected as effect modifiers. We implemented a generalized random forest, where outcome and propensity scores were modeled with fixed effects estimators to control for seasonality and district-level confounding. We divided the conditional average treatment effects into quartiles to map their spatial distribution and identify key effect modifiers via the variable importance and observed indicator profiles across quartiles. Results: While the average effect of heat on emergency hospitalizations was close to null (-0.3 per 1,000), effects in the lowest and highest effect quartiles ranged from -2.3 to +1.4 per 1,000. Demographic indicators explained the largest share of heterogeneity (34.5%), followed by environmental (14.53%) and health/-social factors (10.28%). The indicator profile for the most susceptible districts was characterized by lower air pollution levels, a larger percentage of vulnerable populations, and higher car density. Conclusion: We highlight that heat susceptibility cannot be summarized by one number per region or be explained by a single effect modifier, which needs to be considered in heat-health warning systems and adaptation planning.

  • Open Access

    Integrating respiratory infection surveillance and temperature data improves all-season mortality reconstruction

    by Junyu Wang, Matthias An der Heiden, Christopher Irrgang

    All-season mortality surveillance can inform public-health planning under climate change, but attribution is complicated by overlapping effects of temperature, respiratory infections and other non-environmental factors. Here we develop a component-based neural-network model for daily all-cause mortality that combines high-resolution temperature and dew-point data, severe acute respiratory infection (SARI) hospitalization incidence, demographic structure and an adaptive mortality baseline. In Germany, inputs included non-COVID SARI and COVID-19-associated SARI. Using 2014-2025 mortality data, the hybrid model reduced daily root-mean-square error to 80.2 deaths, compared with 170.2 for a weather-only model and 117.1 for an infection-only model, and enabled age- and sex-resolved examination of fitted components. Including respiratory-infection indicators reduced the increase in risk assigned by the model to cold exposure, suggesting that temperature-only models may partly assign winter infection-related variation to cold. Component magnitudes represent conditional model decompositions rather than causal estimates of deaths due to SARI or temperature.

  • Open Access

    The extrinsic incubation period for Zika virus: a Bayesian time delay modelling study

    by Anna-Maria Hartner, Lothar H. Wieler, Christopher Irrgang

    The extrinsic incubation period (EIP), defined as the time between a mosquito acquiring a virus and becoming capable of transmitting it, is a key component of arbovirus transmission and varies with temperature. For Zika virus (ZIKV), empirical estimates of EIP are derived from heterogeneous laboratory studies and are typically analysed without accounting for interval censoring inherent in vector competence experiments. We applied Bayesian interval-censored survival models to pooled individual-level observations from eight published studies, comparing 20 candidate models representing four parametric survival distributions and five temperature-response functions in Ae. aegypti and Ae. albopictus . Model performance was evaluated using approximate leave-one-out cross-validation. A lognormal survival model with a quadratic temperature response and study-level random intercept provided the best predictive performance. Median EIP declined non-linearly with increasing temperature, from 59.5 days (95% CrI 27.0–124.0) at 20°C to 8.4 days (3.8–17.0) at 32°C in Ae. aegypti . Across the temperature range examined, estimated EIPs for Ae. albopictus were approximately 1.5-fold longer than those for Ae. aegypti . Credible intervals widened at temperature extremes, reflecting between-study heterogeneity and limited data availability. These results provide a statistical framework for estimating temperature-dependent ZIKV EIP while accounting for censoring and uncertainty, supporting improved parameterisation of mechanistic models of arbovirus transmission. Author summary: When a mosquito feeds on an infected host, the virus must replicate and spread to the mosquito’s salivary glands before it can be transmitted. This delay, known as the extrinsic incubation period (EIP), is a key determinant of mosquito-borne disease transmission and varies with temperature. For Zika virus (ZIKV), estimates of EIP come from laboratory studies that differ in experimental design, mosquito species, and viral dose, making them difficult to combine. We analysed data from eight studies using a Bayesian statistical approach that accounts for uncertainty in the time mosquitoes become infectious, because mosquitoes are typically tested only at discrete time points after infection. Our analysis showed that ZIKV EIP decreases non-linearly with increasing temperature, ranging from approximately 60 days at 20°C to approximately 8 days at 32°C in Aedes aegypti . Across temperatures, Aedes albopictus had consistently longer EIPs than Aedes aegypti . Compared with Bayesian EIP models developed for other arboviruses, including dengue, yellow fever, and West Nile virus, ZIKV showed a stronger non-linear temperature response. These estimates and their uncertainty provide improved parameters for models predicting when and where Zika transmission is most likely.

  • Open Access

    Suitable seasons: Global monthly habitat suitability for the arbovirus vectors Aedes aegypti and Aedes albopictus in 1975-2024

    by Tarique Siddiqui, Nadezhda Malysheva, Anna-Maria Hartner, Semyon Butyrin, Diogo Parreira, Jakob-Wendelin Genger, Christopher Irrgang

    The mosquito species Aedes aegypti and Aedes albopictus are the primary vectors of the arboviruses dengue, Zika, and chikungunya. Expansion of these vectors into previously non-endemic regions due to climate and environmental changes has accelerated global burden from arboviral diseases. To combat this, predictive models accurately mapping Aedes habitats are essential for epidemiological modelling, effective vector control, and disease prevention. We introduce the Climademic Suitability Model, a machine learning model that delivers monthly global predictions of Aedes habitat suitability at 0.25° spatial resolution between 1975-2024. The model leverages integrated climate, land use, human population, and mosquito surveillance data to provide an explainable view of the factors governing habitat dynamics. SHAP-based explainability analysis identified temperature and dew point temperature as dominant features driving habitat suitability. Long-term analysis reveals a complex global redistribution of expanding and contracting vector habitats. Suitable areas for both species now encompass regions home to over 5 billion people, coinciding with the world's most pronounced population growth and surpassing projections previously placing this threshold at 2050. The Climademic Suitability Model serves as a framework for near-real-time vector surveillance, climate scenario projection, and integration into transmission models to advance epidemic preparedness in an era of accelerating environmental change.

  • Open Access

    Information Pathways and Voids in Critical German Online Communities During the COVID-19 Vaccination Discourse: Cross-Platform and Mixed Methods Analysis

    by Silvan Wehrli, Anna-Maria Hartner, T Sonia Boender, Bert Arnrich, Christopher Irrgang

    Background: In Germany, the messaging app Telegram (Telegram FZ-LLC) served as a tool to organize protests against public health measures during the COVID-19 pandemic. A community of diverse groups formed around these protests, which used Telegram to discuss and share views outside of the general public discourse and mainstream information ecosystem. This increasingly included conspiracy theories and extremist content, propagated by sources that opposed the mainstream positions of the government and traditional media. While the use of such sources has been thoroughly investigated, the role of mainstream information in these communities remains largely unclear. Objective: We aimed to better understand the use of mainstream information, that is, from government actors and established media outlets, within critical Telegram communities in the context of the COVID-19 pandemic in Germany. We focused on the discourse about the COVID-19 vaccination, a key public health measure. As a central element of this study, we compared the Telegram discourse with the discourse on X (formerly Twitter, X Corp) and in online news—this cross-platform analysis aimed to put the results into a broader societal context. Methods: We analyzed Telegram, X, and news data between 2019 and 2023 for popular topics related to the COVID-19 vaccination discourse. We used a mixed methods approach, including text clustering for the exploration of popular topics, a 2-stage keyword filtering scheme for multitopic classification, link sharing analysis for assessing the prevalence of mainstream information, correlation-based time series analysis for measuring the similarity of discourse dynamics, and thematic analysis to examine the reasons for sharing information. Results: We identified 5 popular vaccination-related topics that were discussed on both Telegram and X, namely death, long COVID, measures in schools, mandatory vaccination, and virus variants. On average per topic, 58% (SD 5.2%) of Telegram posts and 21% (SD 4.9%) of X posts contained an external link. Of these posts containing external links, 11%‐35% of Telegram posts and 44%‐60% of X posts contained a mainstream link per topic. The correlations for week-to-week changes in discourse intensity between Telegram, X, and online mainstream news ranged from no positive association (coefficient 0.6) per topic. Finally, the thematic analysis resulted in 5 themes describing the usage patterns of mainstream information on Telegram and X. The identified themes are observing news, news comments, directed accusations, participation, and reference in discussion (only X). Conclusions: Mainstream information sources were part of the information mix within the analyzed critical Telegram communities. However, the role and prevalence of these sources varied. We argue that differences between platforms may indicate the existence of information voids, which pose a particular challenge in managing infodemics. These insights emphasize the importance of contextualized cross-platform analyses for understanding complex information pathways and their potential for targeted crisis communication.


  • Augmentation of wastewater-based epidemiology with machine learning to support global health surveillance

    by Eva Aßmann, Timo Greiner, Hugues Richard, Matthew Wade, Shelesh Agrawal, Fabian Amman, Sindy Böttcher, Susanne Lackner, Markus Landthaler, Serghei Mangul, et al.


  • Open Access

    Peer Relationships Are a Direct Cause of the Adolescent Mental Health Crisis: Interpretable Machine Learning Analysis of 2 Large Cohort Studies

    by Heiner Stuke, Robert Schlack, Michael Erhart, Anne Kaman, Ulrike Ravens-Sieberer, Christopher Irrgang

    Background: Converging evidence indicates an adolescent mental health crisis in Western societies that has developed and exacerbated over the past decade. The proposed driving factors of this trend include more screen time, physical inactivity, and social isolation, but their causal influence on mental health is insufficiently understood. Objective: The objective of this study is to test whether and based on which predictor variables the development of mental health in adolescents in the last decade can be predicted and to better understand the causal chain of factors at work. Methods: We implemented an interpretable machine learning pipeline based on gradient boosting regression with repeated cross-validation to assess the development of mental health throughout adolescence in members of 2 longitudinal cohort studies, the British Millenium cohort (MC; n=8599) and the German Health Interview and Examination Survey for Children and Adolescents (KiGGS) cohort (n=1212). In total, 144 (MC) and 102 (KiGGS) predictors assessed at the age of around 13.8 years (MC) and 11.6 years (KiGGS) were used to assess mental health at the ages of around 16.7 years (MC) and 16.4 years (KiGGS). Based on these predictive models, we used permutation-based feature importance analyses to identify relevant predictors and predictor domains. Moreover, we performed partial dependence analyses in a causal inference framework to determine the direct effects of physical inactivity, screen time, and peer problems on the development of mental health. Results: The average cross-validated Pearson correlation coefficient (r) between predicted and true mental health in late adolescence was 0.614 (MC) and 0.466 (KiGGS). Feature importance analyses indicated a strong impact of preexisting mental health and weaker impacts of sex (female as a risk factor), physical health (chronic disease as a risk factor), lifestyle, and socioeconomic and family factors (eg, low parental education, income, and mental health as risk factors). Causal inference analyses suggested a strong direct effect of peer relationships, but only a small direct effect of physical inactivity and a very small direct effect of screen time. Conclusions: Mental health development during adolescence can be assessed by a combination of variables from early adolescence. Peer problems represent an important direct cause of mental health development, and their deterioration may contribute to the current mental health crisis.


  • Open Access

    Holistic forecasting for future pandemics: a review of pathogens, models, and data

    by Luis Roger Esquivel Gomez, Nadezdha Malysheva, Juliane Pfeil, Zewen Yang, Sandra M. Bütow, Christopher Irrgang, Nils Körber, Georges Hattab, Denise Kühnert, Katharina Ladewig


  • Top Cited Papers

    The Way Forward to Embrace Artificial Intelligence in Public Health

    by Georges Hattab, Christopher Irrgang, Nils Körber, Denise Kühnert, Katharina Ladewig