Pejman Rohani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Pejman Rohani has a total of 188 co-authors. Here are the top co-authors by number of shared publications:
- John M. Drake
University of Georgia
17 shared publications - Tobias S. Brett
Odum School of Ecology, University of Georgia, Athens, GA, USA
10 shared publications - Matthieu Domenech de Cellès
Max Planck Institute for Infection Biology, Infectious Disease Epidemiology group, Charitéplatz 1, 10117 Berlin, Germany
8 shared publications - Aaron A. King
Department of Ecology & Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109, USA
6 shared publications - Andreas Handel
Affiliations Department of Epidemiology and Biostatistics, University of Georgia, Athens, Georgia, United States of America, Center for the Ecology of Infectious Diseases, University of Georgia, Athens, Georgia, United States of America
4 shared publications - Eamon B. O'DEa
University of Georgia
4 shared publications
Latest Publications
- Open Access
Directional antigenic drift forecasts influenza vaccine effectiveness and guides strain selection
Anticipating antigenic evolution is essential for selecting effective seasonal influenza A/H3N2 vaccine strains. To this end, we integrated data from multiple immunological assays spanning two decades into a unified Bayesian antigenic map. The map resolves twelve antigenic clusters advancing in discrete steps, with several clusters co-circulating in most seasons. In 15 of 21 seasons, the vaccine composition belonged to an earlier cluster than the dominant circulating cluster. We identified geometric features of antigenic space that accounted for three-quarters of the variation in vaccine effectiveness. The direction of each vaccine update relative to recent viral drift predicted effectiveness one season ahead out of sample. In every season, our analyses identified a virus that would have raised predicted effectiveness by 10 percentage points.
Open AccessDirectional antigenic drift forecasts influenza vaccine effectiveness and guides strain selection
Anticipating antigenic evolution is essential for selecting effective seasonal influenza A/H3N2 vaccine strains. To this end, we integrated data from multiple immunological assays spanning two decades into a unified Bayesian antigenic map. The map resolves twelve antigenic clusters advancing in discrete steps, with several clusters co-circulating in most seasons. In 15 of 21 seasons, the vaccine composition belonged to an earlier cluster than the dominant circulating cluster. We identified geometric features of antigenic space that accounted for three-quarters of the variation in vaccine effectiveness. The direction of each vaccine update relative to recent viral drift predicted effectiveness one season ahead out of sample. In every season, our analyses identified a virus that would have raised predicted effectiveness by 10 percentage points.
- Open Access
- 11 August 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (8) , e1014628
Alignment-free prediction of cross-reactivity in influenza A (H3N2) anticipates antigenic drift
Since its introduction in 1968, Influenza A (H3N2) has undergone continuous antigenic evolution, necessitating frequent vaccine updates. To predict antigenicity and characterize antigenic drift without multiple sequence alignments, we present FluEmbed , a computational framework that leverages protein language models. FluEmbed accurately quantified the antigenic impact of viral evolution from RNA sequences, achieving strong predictive performance against hemagglutination inhibition (HI) assay titers (Spearman correlation: ρ = 0.67–0.80). FluEmbed also outperformed sequence-distance baselines (e.g., Hamming and BLOSUM62) and phylogenetic tree-based models that require sequence alignment. Using this model, we conducted in-silico mutagenesis experiments to identify site/amino acid combinations that differentially impacted antigenicity. To systematically investigate how specific mutations influence immune escape, we defined two classes of mutations: ‘constrained’, where only the most likely amino acid changes at historically mutation-prone sites were considered (thereby limiting the mutation space) and ‘unconstrained’, where all possible substitutions were allowed, providing a full exploration of potential antigenic shifts. Constrained mutations often confer limited antigenic changes, whereas unconstrained mutations exhibit greater escape potential, particularly outside the dominant viral lineages. Notably, 3C.2a was the only major lineage in which constrained and unconstrained mutations showed no significant difference ( p ≈ 0.95), suggesting ongoing intra-clade competition rather than inter-lineage antigenic replacement. By enabling rapid, alignment-free antigenic prediction directly from sequence data, FluEmbed could complement traditional HI assays in real-time influenza surveillance and inform vaccine strain selection decisions.
- 11 August 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (8) , e1014628
Open AccessAlignment-free prediction of cross-reactivity in influenza A (H3N2) anticipates antigenic drift
Since its introduction in 1968, Influenza A (H3N2) has undergone continuous antigenic evolution, necessitating frequent vaccine updates. To predict antigenicity and characterize antigenic drift without multiple sequence alignments, we present FluEmbed , a computational framework that leverages protein language models. FluEmbed accurately quantified the antigenic impact of viral evolution from RNA sequences, achieving strong predictive performance against hemagglutination inhibition (HI) assay titers (Spearman correlation: ρ = 0.67–0.80). FluEmbed also outperformed sequence-distance baselines (e.g., Hamming and BLOSUM62) and phylogenetic tree-based models that require sequence alignment. Using this model, we conducted in-silico mutagenesis experiments to identify site/amino acid combinations that differentially impacted antigenicity. To systematically investigate how specific mutations influence immune escape, we defined two classes of mutations: ‘constrained’, where only the most likely amino acid changes at historically mutation-prone sites were considered (thereby limiting the mutation space) and ‘unconstrained’, where all possible substitutions were allowed, providing a full exploration of potential antigenic shifts. Constrained mutations often confer limited antigenic changes, whereas unconstrained mutations exhibit greater escape potential, particularly outside the dominant viral lineages. Notably, 3C.2a was the only major lineage in which constrained and unconstrained mutations showed no significant difference ( p ≈ 0.95), suggesting ongoing intra-clade competition rather than inter-lineage antigenic replacement. By enabling rapid, alignment-free antigenic prediction directly from sequence data, FluEmbed could complement traditional HI assays in real-time influenza surveillance and inform vaccine strain selection decisions.
- Open Access
Strain coexistence and competition for pathogens with asymmetric cross-immunity and waning immunity
The coexistence of competing pathogen strains is shaped by cross-immunity, the cross-protection that infection with one strain confers against another. Although cross-immunity is often asymmetric between strains, this asymmetry is often neglected in the literature on multi-strain coexistence. The effect on coexistence-exclusion outcomes of waning immunity\textemdash which is particularly relevant for antigenically evolving pathogens\textemdash is also poorly understood. To understand how these factors affect strain coexistence, here we analyze a status-based two-strain SIRS model with asymmetric cross-immunity and strain-specific rates for transmission, recovery, and waning of immunity. We derive explicit invasion thresholds that also determine the feasibility and local stability of a unique coexistence equilibrium. Thus, these thresholds allow us to characterize the region of stable strain coexistence, as a function of the cross-immunities and rates of waning immunity. We also obtain closed-form expressions for the strain prevalences at the coexistence equilibrium, showing that the total prevalence may vary non-monotonically as the basic reproduction number of one strain increases. Finally, we show that a transient reduction in transmission can move a coexisting strain pair across an invasion boundary, driving the weaker strain extinct. Applying this result to influenza B, our analysis offers a parsimonious explanation for the disappearance of the Yamagata lineage during the COVID-19 pandemic.
Open AccessStrain coexistence and competition for pathogens with asymmetric cross-immunity and waning immunity
The coexistence of competing pathogen strains is shaped by cross-immunity, the cross-protection that infection with one strain confers against another. Although cross-immunity is often asymmetric between strains, this asymmetry is often neglected in the literature on multi-strain coexistence. The effect on coexistence-exclusion outcomes of waning immunity\textemdash which is particularly relevant for antigenically evolving pathogens\textemdash is also poorly understood. To understand how these factors affect strain coexistence, here we analyze a status-based two-strain SIRS model with asymmetric cross-immunity and strain-specific rates for transmission, recovery, and waning of immunity. We derive explicit invasion thresholds that also determine the feasibility and local stability of a unique coexistence equilibrium. Thus, these thresholds allow us to characterize the region of stable strain coexistence, as a function of the cross-immunities and rates of waning immunity. We also obtain closed-form expressions for the strain prevalences at the coexistence equilibrium, showing that the total prevalence may vary non-monotonically as the basic reproduction number of one strain increases. Finally, we show that a transient reduction in transmission can move a coexisting strain pair across an invasion boundary, driving the weaker strain extinct. Applying this result to influenza B, our analysis offers a parsimonious explanation for the disappearance of the Yamagata lineage during the COVID-19 pandemic.
- Open Access
- 4 August 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (8) , e1014538
Quantifying the impact of vaccination on pertussis dynamics in Sweden
The last few decades have witnessed a resurgence in pertussis notifications in a number of countries with high vaccine coverage, including Sweden. The underlying causes of the resurgence have been the subject of much scientific debate. To arbitrate among the putative drivers of the resurgence in Sweden, we formulated a mechanistic transmission model which we fit to age-structured time-series notifications data via likelihood maximisation. Given our model, we find the data are best explained by the combined effects of a low basic reproductive number, incomplete (leaky) DTaP-derived immunity, a much lower reporting probability of infections in older individuals and waning of vaccine-derived immunity. In addition, our modelling explains the post-2014 resurgence as a combination of two factors. First, a dynamical transient known as the honeymoon effect, in which a rebound in transmission follows after a rapid decrease in the average population susceptibility. Second, an increase in the infection reporting probability from 2014 onwards, likely due to the use of new laboratory testing methods. Additionally, we used our fitted transmission model to reconstruct indirect protective effects of vaccination. Our results suggest immunization prevents about 50% of potential infections in individuals too young to receive vaccination. However, our statistical inference demonstrates that pertussis elimination is not possible with routine immunization using acellular vaccines due to the combined effects of vaccine leakiness and waning immunity.
- 4 August 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (8) , e1014538
Open AccessQuantifying the impact of vaccination on pertussis dynamics in Sweden
The last few decades have witnessed a resurgence in pertussis notifications in a number of countries with high vaccine coverage, including Sweden. The underlying causes of the resurgence have been the subject of much scientific debate. To arbitrate among the putative drivers of the resurgence in Sweden, we formulated a mechanistic transmission model which we fit to age-structured time-series notifications data via likelihood maximisation. Given our model, we find the data are best explained by the combined effects of a low basic reproductive number, incomplete (leaky) DTaP-derived immunity, a much lower reporting probability of infections in older individuals and waning of vaccine-derived immunity. In addition, our modelling explains the post-2014 resurgence as a combination of two factors. First, a dynamical transient known as the honeymoon effect, in which a rebound in transmission follows after a rapid decrease in the average population susceptibility. Second, an increase in the infection reporting probability from 2014 onwards, likely due to the use of new laboratory testing methods. Additionally, we used our fitted transmission model to reconstruct indirect protective effects of vaccination. Our results suggest immunization prevents about 50% of potential infections in individuals too young to receive vaccination. However, our statistical inference demonstrates that pertussis elimination is not possible with routine immunization using acellular vaccines due to the combined effects of vaccine leakiness and waning immunity.
- Open Access
Identifying the viral and epidemiological factors behind the apparent global extinction of influenza B/Yamagata
Until 2020, two lineages of the influenza B virus had co-circulated globally. The COVID-19 pandemic led to a near-absence of influenza infections. While B/Victoria reemerged in late 2021, there have been no reports of B/Yamagata since the pandemic. To investigate which epidemiological and immunological factors were primarily responsible for the extinction of B/Yamagata, we developed a global model for the two influenza B lineages. To mimic the transmission impacts of the pandemic, we implemented a transient reduction in contacts and identified parameter values that recapitulated viral coexistence dynamic before the pandemic and the qualitative post-pandemic outcomes of B/Victoria (reemergence in late 2021) and B/Yamagata (extinction). Our results suggest that, rather than immunological or evolutionary mechanisms, the extinction of B/Yamagata was mainly driven by its lower basic reproduction number making the virus particularly vulnerable during the early phase of the pandemic. Stochastic simulations of our best-fitting model suggest that B/Victoria was also close to extinction during this period. We investigate the model to assess the feasibility of B/Victoria eradication through vaccination and the potential for a sustained re-emergence of B/Yamagata in the 2026-27 flu season, thus highlighting important considerations for biosafety. Significance: For decades, the Yamagata and Victoria lineages of the influenza B virus co-circulated globally, causing millions of infections in humans. Only Victoria reemerged after the COVID-19 pandemic. By combining mechanistic transmission modeling with worldwide influenza surveillance data, we show that Yamagata’s extinction is most consistent with a lower intrinsic transmissibility than that of the surviving Victoria lineage. Contrary to other proposed explanations, differences in cross-immunity and immune escape appear insufficient to explain the observed outcome. Our analyses further indicate that Victoria narrowly avoided extinction during the COVID-19 disruption, but now transmission of Yamagata would be sustained if the lineage were reintroduced. These results inform both influenza eradication efforts and biosafety policy.
Open AccessIdentifying the viral and epidemiological factors behind the apparent global extinction of influenza B/Yamagata
Until 2020, two lineages of the influenza B virus had co-circulated globally. The COVID-19 pandemic led to a near-absence of influenza infections. While B/Victoria reemerged in late 2021, there have been no reports of B/Yamagata since the pandemic. To investigate which epidemiological and immunological factors were primarily responsible for the extinction of B/Yamagata, we developed a global model for the two influenza B lineages. To mimic the transmission impacts of the pandemic, we implemented a transient reduction in contacts and identified parameter values that recapitulated viral coexistence dynamic before the pandemic and the qualitative post-pandemic outcomes of B/Victoria (reemergence in late 2021) and B/Yamagata (extinction). Our results suggest that, rather than immunological or evolutionary mechanisms, the extinction of B/Yamagata was mainly driven by its lower basic reproduction number making the virus particularly vulnerable during the early phase of the pandemic. Stochastic simulations of our best-fitting model suggest that B/Victoria was also close to extinction during this period. We investigate the model to assess the feasibility of B/Victoria eradication through vaccination and the potential for a sustained re-emergence of B/Yamagata in the 2026-27 flu season, thus highlighting important considerations for biosafety. Significance: For decades, the Yamagata and Victoria lineages of the influenza B virus co-circulated globally, causing millions of infections in humans. Only Victoria reemerged after the COVID-19 pandemic. By combining mechanistic transmission modeling with worldwide influenza surveillance data, we show that Yamagata’s extinction is most consistent with a lower intrinsic transmissibility than that of the surviving Victoria lineage. Contrary to other proposed explanations, differences in cross-immunity and immune escape appear insufficient to explain the observed outcome. Our analyses further indicate that Victoria narrowly avoided extinction during the COVID-19 disruption, but now transmission of Yamagata would be sustained if the lineage were reintroduced. These results inform both influenza eradication efforts and biosafety policy.
- Open Access
- 1 July 2026
- Published by The Royal Society in Royal Society Open Science
- Vol. 13 (7)
Epidemic dynamics under variable compliance with interventions
Behavioural non-pharmaceutical interventions, such as mask-wearing and social distancing, can reduce the spread of infectious diseases although not all members of a population may comply with public policies. We developed a compartmental model to understand the spread of an infectious disease within a population comprising compliant and non-compliant groups. The model takes into account various behavioural factors influencing flows between compliant and non-compliant groups, such as peer pressure, risk perception, fear of infection and severity of the policy. The coupled social-disease system under various social mechanisms exhibits distinct disease dynamics, revealing the important role of social mechanisms in controlling disease spread in a mixed population. Further, our results show that social mechanisms and the initial state of the population affect the time delay before outbreaks occur and peak prevalence. Our findings emphasize that dynamic social factors may significantly affect outbreaks and we suggest that models incorporating coupled social mechanisms warrant further investigation.
- 1 July 2026
- Published by The Royal Society in Royal Society Open Science
- Vol. 13 (7)
Open AccessEpidemic dynamics under variable compliance with interventions
Behavioural non-pharmaceutical interventions, such as mask-wearing and social distancing, can reduce the spread of infectious diseases although not all members of a population may comply with public policies. We developed a compartmental model to understand the spread of an infectious disease within a population comprising compliant and non-compliant groups. The model takes into account various behavioural factors influencing flows between compliant and non-compliant groups, such as peer pressure, risk perception, fear of infection and severity of the policy. The coupled social-disease system under various social mechanisms exhibits distinct disease dynamics, revealing the important role of social mechanisms in controlling disease spread in a mixed population. Further, our results show that social mechanisms and the initial state of the population affect the time delay before outbreaks occur and peak prevalence. Our findings emphasize that dynamic social factors may significantly affect outbreaks and we suggest that models incorporating coupled social mechanisms warrant further investigation.
- Open Access
- 15 April 2026
- Published by The Royal Society in Journal of The Royal Society Interface
- Vol. 23 (237)
A single autoregressive integrated moving average with exogenous variable outperforms ensembles of autoregressive models for forecasting influenza hospitalizations in the contiguous United States
Infectious disease forecasting is important to public health decision-making, particularly for mitigating the burden of seasonal influenza. We propose and evaluate comprehensive autoregressive modelling approaches (autoregressive integrated moving average (ARIMA) and autoregressive integrated moving average with exogenous variables (ARIMAX)) for short-term forecasts of influenza-related hospitalizations across the contiguous United States (US). We used data from the National Healthcare Safety Network to forecast the influenza seasons of the years 2022 to 2024 in 48 states. We compare automatically tuned models (AUTO) and ensembles of different sizes (ES27 or ES64). Base models are ARIMA models or ARIMAX models using forecasting covariates with a one-week lag of the forecast day, including: mean temperature, mean hospitalizations by epidemiological weeks, mean hospitalizations in adjacent states and the average of hospitalizations across the contiguous US. We also investigated the effect of forecasting based on log-transformed hospitalizations data and back-transforming the results (LB). Overall, 29 approaches were compared to the AUTO ARIMA model without logarithmic transformation and the FluSight baseline using weighted interval scores. Our results indicate that using the LB approach, which forecasts based on epidemic growth rate and back-transform the results, most increased model performance. The average of hospitalizations across the contiguous US (AVG) forecasting feature was the most useful covariate. Ensemble approaches typically did not improve performance. The best model was the AUTO ARIMAX using the AVG covariate with log-back transformation.
- 15 April 2026
- Published by The Royal Society in Journal of The Royal Society Interface
- Vol. 23 (237)
Open AccessA single autoregressive integrated moving average with exogenous variable outperforms ensembles of autoregressive models for forecasting influenza hospitalizations in the contiguous United States
Infectious disease forecasting is important to public health decision-making, particularly for mitigating the burden of seasonal influenza. We propose and evaluate comprehensive autoregressive modelling approaches (autoregressive integrated moving average (ARIMA) and autoregressive integrated moving average with exogenous variables (ARIMAX)) for short-term forecasts of influenza-related hospitalizations across the contiguous United States (US). We used data from the National Healthcare Safety Network to forecast the influenza seasons of the years 2022 to 2024 in 48 states. We compare automatically tuned models (AUTO) and ensembles of different sizes (ES27 or ES64). Base models are ARIMA models or ARIMAX models using forecasting covariates with a one-week lag of the forecast day, including: mean temperature, mean hospitalizations by epidemiological weeks, mean hospitalizations in adjacent states and the average of hospitalizations across the contiguous US. We also investigated the effect of forecasting based on log-transformed hospitalizations data and back-transforming the results (LB). Overall, 29 approaches were compared to the AUTO ARIMA model without logarithmic transformation and the FluSight baseline using weighted interval scores. Our results indicate that using the LB approach, which forecasts based on epidemic growth rate and back-transform the results, most increased model performance. The average of hospitalizations across the contiguous US (AVG) forecasting feature was the most useful covariate. Ensemble approaches typically did not improve performance. The best model was the AUTO ARIMAX using the AVG covariate with log-back transformation.
- Open Access
Global comparison of influenza A and B epidemiology identifies consistent geographic and socio-demographic predictors
Seasonal influenza outbreaks are caused by types A and B that together account for an estimated 3-5 million severe cases each year. Most attention has focused on influenza A viruses (IAVs) due to their rapid evolutionary dynamics and high disease burden, and has been concentrated in well-observed high-income regions. Here, we use a macroecological approach to compare and contrast the global epidemiology of IAVs and influenza B viruses (IBVs) across 111 countries and 15 influenza seasons (2010-2024). We first show how temporal correlations between countries depends on both distance and geographic region. For both IAV and IBV, we find high overall synchrony among northern temperate countries, whereas tropical countries display marked heterogeneity. At the longer time scale of influenza seasons, we next quantify sampling intensity, positivity, seasonality, fade-out dynamics and the timing and variability of epidemic peaks. We then describe how these long-term epidemiological outcomes change in association with a suite of 17 geographic, climatic, and socio-economic variables. In addition, we document persistent surveillance gaps, particularly in Africa, and highlight ongoing but spatially variable impacts of the SARS-CoV-2 pandemic-era on sampling. Overall, we find strong correspondence between the macroscopic features of IAV and IBV epidemiology, with critical roles played by geography and climate (especially latitude and temperature), economics ( per capita GDP) and demographics (population size and per capita birth rate). Significance Statement: The global circulation of seasonal influenza A and B viruses (IAV and IBV) imposes major human health impacts each year that very widely across space and time. An improved understanding of these dynamics could improve public health preparedness, response, and intervention efforts. Here we offer a comprehensive comparison of IAV and IBV dynamics across 15 seasons, 111 countries, and six continents. We demonstrate the impact of distance and region on temporal correlation, quantify how measures of influenza seasonality change with geographic and socioeconomic factors, and predict how frequently influenza cases are absent from countries. Our study finds widespread similarities between IAV and IBV (along with key differences), documents notable geographic clusters of countries with shared dynamics, and highlights persistent gaps in global influenza surveillance.
Open AccessGlobal comparison of influenza A and B epidemiology identifies consistent geographic and socio-demographic predictors
Seasonal influenza outbreaks are caused by types A and B that together account for an estimated 3-5 million severe cases each year. Most attention has focused on influenza A viruses (IAVs) due to their rapid evolutionary dynamics and high disease burden, and has been concentrated in well-observed high-income regions. Here, we use a macroecological approach to compare and contrast the global epidemiology of IAVs and influenza B viruses (IBVs) across 111 countries and 15 influenza seasons (2010-2024). We first show how temporal correlations between countries depends on both distance and geographic region. For both IAV and IBV, we find high overall synchrony among northern temperate countries, whereas tropical countries display marked heterogeneity. At the longer time scale of influenza seasons, we next quantify sampling intensity, positivity, seasonality, fade-out dynamics and the timing and variability of epidemic peaks. We then describe how these long-term epidemiological outcomes change in association with a suite of 17 geographic, climatic, and socio-economic variables. In addition, we document persistent surveillance gaps, particularly in Africa, and highlight ongoing but spatially variable impacts of the SARS-CoV-2 pandemic-era on sampling. Overall, we find strong correspondence between the macroscopic features of IAV and IBV epidemiology, with critical roles played by geography and climate (especially latitude and temperature), economics ( per capita GDP) and demographics (population size and per capita birth rate). Significance Statement: The global circulation of seasonal influenza A and B viruses (IAV and IBV) imposes major human health impacts each year that very widely across space and time. An improved understanding of these dynamics could improve public health preparedness, response, and intervention efforts. Here we offer a comprehensive comparison of IAV and IBV dynamics across 15 seasons, 111 countries, and six continents. We demonstrate the impact of distance and region on temporal correlation, quantify how measures of influenza seasonality change with geographic and socioeconomic factors, and predict how frequently influenza cases are absent from countries. Our study finds widespread similarities between IAV and IBV (along with key differences), documents notable geographic clusters of countries with shared dynamics, and highlights persistent gaps in global influenza surveillance.
- 11 February 2026
- Published by Taylor & Francis in Future Microbiology
- Vol. 21 (3) , 321-327
Can vaccine-preventable disease resurgence be anticipated? Leading indicators and tipping points
Vaccination programs have averted millions of childhood deaths, yet vaccine-preventable diseases (VPDs) continue to resurge as coverage declines and pathogen evolution undermines previously successful vaccines. Anticipating resurgence is a public health priority. We review theoretical and empirical advances in the study of early warning signals (EWS) of epidemic transitions, with a focus on critical slowing down (CSD) - a phenomenon in which recovery from perturbations becomes slower near the epidemic threshold. We summarize the mechanisms that generate CSD, indicators that can be extracted from surveillance data, and the conditions under which signals may be detectable. We then examine case studies to illustrate the opportunities and challenges of applying EWS to VPD resurgence. Theory and computer simulations show that CSD can precede both elimination and resurgence, with increases in variance and autocorrelation calculated from disease surveillance reports emerging as consistent indicators. Empirical evidence supports this potential, though performance depends on noise structure, seasonality, spatial clustering, and outbreak responses. Case studies highlight both successful applications and contexts where signals were weak or absent. EWS offer a promising framework for anticipating VPD resurgence, but further research is required to refine methods, integrate mechanistic and social-behavioral drivers, and evaluate applicability across pathogens and settings.
- 11 February 2026
- Published by Taylor & Francis in Future Microbiology
- Vol. 21 (3) , 321-327
Can vaccine-preventable disease resurgence be anticipated? Leading indicators and tipping points
Vaccination programs have averted millions of childhood deaths, yet vaccine-preventable diseases (VPDs) continue to resurge as coverage declines and pathogen evolution undermines previously successful vaccines. Anticipating resurgence is a public health priority. We review theoretical and empirical advances in the study of early warning signals (EWS) of epidemic transitions, with a focus on critical slowing down (CSD) - a phenomenon in which recovery from perturbations becomes slower near the epidemic threshold. We summarize the mechanisms that generate CSD, indicators that can be extracted from surveillance data, and the conditions under which signals may be detectable. We then examine case studies to illustrate the opportunities and challenges of applying EWS to VPD resurgence. Theory and computer simulations show that CSD can precede both elimination and resurgence, with increases in variance and autocorrelation calculated from disease surveillance reports emerging as consistent indicators. Empirical evidence supports this potential, though performance depends on noise structure, seasonality, spatial clustering, and outbreak responses. Case studies highlight both successful applications and contexts where signals were weak or absent. EWS offer a promising framework for anticipating VPD resurgence, but further research is required to refine methods, integrate mechanistic and social-behavioral drivers, and evaluate applicability across pathogens and settings.
- Open Access
- 22 January 2026
- Published by Taylor & Francis in Future Microbiology
- Vol. 21 (2) , 135-140
No evidence for critical slowing down before measles outbreaks in the US, 2006–2025
Measles, once eliminated in the United States, has resurged with major outbreaks in 2014, 2018/2019, and 2025. Anticipating such resurgences is a public health priority. Early warning signals (EWS) based on critical slowing down predict increases in variance and autocorrelation near epidemic thresholds. We assessed the reliability of these statistical EWS for anticipating US measles outbreaks. We analyzed weekly US measles case data from 2006-2025, harmonizing US Centers for Disease Control and Prevention (CDC) surveillance reports with a curated domestic case dataset. Outbreaks were defined using baseline thresholds, identifying three major events. Using the spaero R package, we estimated variance and lag-1 autocorrelation in rolling windows and tested for trends using autocorrelation-adjusted Spearman correlations. Sensitivity analyses evaluated robustness to detrending and bandwidth choices. No consistent increases in variance or autocorrelation were detected in the full time series or in pre-outbreak windows. Outbreak duration, size, and fade-out frequency showed no significant trends across the study period. Statistical EWS did not reliably anticipate US measles outbreaks. More research is needed to evaluate their applicability to vaccine-preventable disease resurgence, particularly in settings shaped by heterogeneous susceptibility and outbreak responses.
- 22 January 2026
- Published by Taylor & Francis in Future Microbiology
- Vol. 21 (2) , 135-140
Open AccessNo evidence for critical slowing down before measles outbreaks in the US, 2006–2025
Measles, once eliminated in the United States, has resurged with major outbreaks in 2014, 2018/2019, and 2025. Anticipating such resurgences is a public health priority. Early warning signals (EWS) based on critical slowing down predict increases in variance and autocorrelation near epidemic thresholds. We assessed the reliability of these statistical EWS for anticipating US measles outbreaks. We analyzed weekly US measles case data from 2006-2025, harmonizing US Centers for Disease Control and Prevention (CDC) surveillance reports with a curated domestic case dataset. Outbreaks were defined using baseline thresholds, identifying three major events. Using the spaero R package, we estimated variance and lag-1 autocorrelation in rolling windows and tested for trends using autocorrelation-adjusted Spearman correlations. Sensitivity analyses evaluated robustness to detrending and bandwidth choices. No consistent increases in variance or autocorrelation were detected in the full time series or in pre-outbreak windows. Outbreak duration, size, and fade-out frequency showed no significant trends across the study period. Statistical EWS did not reliably anticipate US measles outbreaks. More research is needed to evaluate their applicability to vaccine-preventable disease resurgence, particularly in settings shaped by heterogeneous susceptibility and outbreak responses.