Last updated: 2025-04-17.
https://fluscenariomodelinghub.org/viz.html
Previous rounds (round 1 to round 1 of 2024-2025 (round 5)) are available in the Flu Scenario Modeling Hub - Archive GitHub Repository
Even the best models of infectious disease transmission struggle to give accurate forecasts at time scales greater than 3-4 weeks due to unpredictable drivers like changing policy environments, behavior change, development of new control measures, and stochastic events. However, policy decisions around the course of infectious diseases, particularly emerging and seasonal infections, often require projections in the time frame of months. The goal of long-term projections is to compare outbreak trajectories under different scenarios, as opposed to offering a specific, unconditional estimate of what “will” happen. As such, long-term projections can guide longer-term decision-making while short-term forecasts are more useful for situational awareness and guiding immediate response.
We have specified a set of scenarios and target outcomes to allow alignment of model projections for collective insights. Scenarios have been designed in consultation with academic modeling teams and government agencies (e.g., CDC).
This repository follows the guidelines and standards outlined by the hubverse, which provides a set of data formats and open source tools for modeling hubs.
The Flu Scenario Modeling Hub is open to any team willing to provide projections at the right temporal and spatial scales, with minimal gatekeeping. We only require that participating teams share point estimates and uncertainty bounds, along with a short model description and answers to a list of key questions about design. A major output of the projection hub is ensemble estimates of epidemic outcomes (e.g., infection, hospitalizations, and deaths), for different time points, intervention scenarios, and US jurisdictions.
Those interested to participate, please read the README file and email us at [email protected].
Model projections should be submitted via pull request to the data-processed folder of this GitHub repository. Technical instructions for submission and required file formats can be found here.
Influenza Round 1 2025-26: Vaccine Coverage - Vaccine impact and pre-season projections for the 2025-26 season
The primary goal of Round 1 of 2025-26 is to evaluate the impact of flu vaccination in light of plausible trajectories and disease burden for the upcoming season. A secondary goal is to provide pre-season projections for the upcoming influenza season at a time when flu activity is low, similarly to prior influenza rounds. We will consider a single scenario axis related to vaccine coverage, while all the epidemiological uncertainty will be collapsed and left at teams’ discretion. Projections will run over a 43-week period, from Sun Aug 10, 2025 to Sat June 6, 2026. Unlike prior years, we will consider counterfactual no–vaccination scenarios, which will allow estimation of the direct and indirect benefits of vaccination in different age groups. Scenarios will follow a 3*1 structure:
This year, we leave all uncertainty regarding influenza epidemiological parameters at teams’ discretion. This includes pre-existing immunity at the start of the season, transmissibility, seeding, antigenic novelty, subtype dominance, seasonality and behavioral effects.
Teams should consider a large range of parameter uncertainty that is both consistent with the long-term behavior of influenza (eg, the last ~20 years of disease burden, timing, and effective transmissibility) and the recent past (ie, the amount of immunity accumulated in the last season, or recent seasons if relevant). Teams can calibrate their model to a range of historical influenza epidemics for multiple past seasons (eg, dashboard of past seasons’ impacts), or rely on relevant publications. For teams looking for guidance, the maximum Re for seasonal flu has ranged between 1.1 and 1.4 in recent years. Conceptually, transmissibility at the start of the season includes the effect of prior immunity accumulated in past seasons and antigenic change in viruses ultimately circulating in 2025-26. Both of these mechanisms can generate large intensity seasons. For models that consider antigenic changes, any simulated antigenic change should be consistent in both probability and magnitude with observations from past seasonal influenza epidemics (all subtypes combined).
A primary question of interest in this round is to contrast the impact of vaccination in seasons with a high or low disease burden (as more hospitalizations and deaths will theoretically be averted in a large flu season). As a result, we intend to analyze vaccine effects by splitting individual simulations by high and low total epidemic sizes (based on counterfactual scenario C). By considering the shape of individual simulations, we can also assess the differential effects of vaccination for earlier and later epidemics, and any other aspect of influenza epidemiology that may interact with vaccination impact (eg, age distribution of hospitalizations etc).
Teams can use their best scientific judgment to define severity (ie, hospitalization risk given infection, or infection fatality risk) based on past seasons.
The seasonal trends, timing, weekly trajectory and age distribution of the projected 2025-26 season are at teams discretion. These features are expected to come from model assumptions, for instance from prior immunity, seeding, or contacts.
Teams who rely on subtype-specific flu models can assume any subtype dominance or co-dominance patterns in 2025-26.
We ask that teams include assumptions regarding Re, pre-existing immunity, and antigenic changes (or related parameters), seasonality, and severity in the meta-data abstract. We will also calculate Re a posteriori from the individual simulations for further analyses. To further facilitate comparison between models and individual simulations, we also request that teams tag each trajectory by S0, the all-age immunity proportion on the first day of simulations, Aug 10, 2025 (total individuals fully or partially susceptible/total population size). S0 will be submitted as a new required target at the simulation level. Additional details on initial susceptibility by age or subtype (should the model have age or subtype structure) can also be provided as optional S0 targets. See sample-level tag documentation for details on calculation and formatting of S0.
To calibrate epidemiologic patterns to past seasons, teams can use a variety of datasets (eg, ILI, % positive for influenza, FluSurv-NET hospitalizations, NHSN etc) to estimate quantities that will inform their model. Information about the epidemiology of the 2024-25 season can be found here and here for earlier seasons.
Recall that with the above published end-of-season summary estimates, the hospitalization burden is corrected for underreporting and should not be used for direct estimation of hospitalizations as would be reported to HHS Protect/NHSN. Using the data provided in the above links is fine to estimate age distribution and relative severity of different seasons.
FluSurv-NET is an influenza hospitalization surveillance network that collects data on laboratory-confirmed influenza-associated hospitalizations through a network of acute care hospitals in a subset of states (14 as of August 2023). Age-specific weekly rates per 100,000 population are reported in this system. Note that for most states participating in FluSurv-NET, only a fraction of the state population is represented in surveillance. However, for teams who prefer counts rather than rates for calibration, state population sizes can be used to approximate hospital admission counts based on the observed rates. Additional details on interpreting FluSurv-NET data are available at Influenza Hospitalization Surveillance Network (FluSurv-NET)
Assumptions regarding vaccine effectiveness: In all scenarios, we will use an all-age VE of 50% against medically attended influenza illnesses and hospitalizations, in line with the average VE reported in recent seasons. We assume that VE against hospitalizations and medical illnesses is the same. Teams who have developed age-stratified models can consider age differences in VE; in this case, these age-specific VE should apply to all scenarios A-C.
The 50% VE assumption should be considered as directly applicable to the 2025-26 season, even though the exact mix of circulating subtypes (particularly how much flu B may circulate) may differ from recent seasons. Relatedly, if teams assume a probability of antigenic change as part of their uncertainty range, the effective VE for an antigenically advanced strain should be maintained at 50% against hospitalizations and medically attended illnesses. In other words, this assumes that the vaccine would remain well matched, even if there was an antigenic change.
As previously, assumptions about VE against infection and transmission are at teams’ discretion, but we provide guidelines. In general, teams should be assuming a lower VE against infection and transmission than against hospitalization or medically attended illness. For instance, a recent household transmission study found that the overall VE for preventing secondary infections among household contacts was 21.0% (1.4%, 36.7%) and varied by influenza type. In addition, a community study from 2010-11 reports an adjusted VE of 31% (-7-55%) against community-acquired influenza infection, which includes mild and transient illnesses.
Assumptions regarding VE impact on infection and transmission should be reported in the abstract meta-data.
We have provided weekly state-level coverage to use in scenarios A-B here for age groups: 0-4 yr, 5-12 yr, 13-17 yr, 0-17 yr, 18-49 yr, 50-64 yr, 65+ yr. Projected coverage for scenarios A-B are based on vaccination rates reported in 2023-24, which is the most recent complete year of vaccination data. We provide cumulative coverage curves for business as usual scenarios (A: same coverage as in 2023-24), with 35% correction down in individuals under 65 yrs to match scenario B assumptions. The 35% reductions have been applied to each jurisdiction and age group under 65 yrs for scenario B. I.e., if Alabama reported 50% coverage in age group 18-64 in 2023-24, the assumed coverage in this age group is 50% in scenario A and 50*0.65=32.5% in scenario B. The timing of vaccination and relative differences between age groups and states will proceed similarly to past years, which has been taken into account in the vaccination file that we provide for the 2025-26 season. In summary, coverage data in this file can be used as is, without further adjustment.
Unlike prior years, we are now considering a hypothetical counterfactual scenario, scenario C, where vaccination coverage should be set to 0% in all age groups and jurisdictions.
Simulations should be paired across all 3 scenarios, i.e. simulations from scenarios A-B-C should be paired using all parameters affecting disease dynamics (i.e. same initial immunity conditions, Re, IFR, IHR, behavior, and any other relevant disease parameters), with the only difference being vaccine conditions.
ROUND 1 FLU 2025-26
- Scenarios set (no changes after): Friday, Aug 1, 2025
- Projections due: Friday, Sep 5, 2025
- Report finalized: No later than Friday, Sep 12, 2025
- Prior immunity is at teams’ discretion.
- Prior influenza immunity is assumed to be a combination of residual immunity from previous infections and previous seasonal vaccinations. The exact specifications of prior immunity are left at the discretion of each team, and will depend on model specification, but we provide suggestions below.
- At the onset of a typical influenza season (all subtypes combined), modeling has estimated that around 30-35% of the population has prior immunity (65-70% susceptible), the effective reproduction number ranges from 1.2-1.4, and the attack rate (final size) is between 8-25%, Yang et al, 2015. The 2009 pandemic, which was marked by the emergence of a new strain to which individuals under the age of 50 yrs were susceptible, was associated with greater transmission (cumulative attack rates 32% over 2009) and decreased prior immunity compared to a regular season (prior immunity in 2009 is ~25% instead of ~33%).
- Teams are allowed to vary prior immunity by virus subtype, age or other demographic characteristics, and state. Teams should plan to include these assumptions in their abstract metadata.
- No major interactions with future COVID-19 surges (immunological, social, behavioral) should be considered in this round.
- Hospitalization and death targets include the impact of all influenza subtypes combined. Weekly surveillance updates can be found here.
- Subtype-specific models are allowed, and we do not make any assumptions about the particular subtype(s) circulating in 2025-26.
- Vaccine-induced immunity has been found to decrease rapidly over the course of an influenza season, Ferdinands et al, 2017
- Age-stratification is recommended.
- Age-strata:
- Required overall population
0-130 - Recommended age-strata:
0-4,5-17,18-49,50-64, and 65+ (65-130) (or some aggregation of this, like 18-64, etc.). Most of the burden on hospitalization and deaths come from the 0-4 and 65+ age groups.
- Required overall population
- Sun Aug 10, 2025 to Sat June 7, 2026 (43 weeks)
In this round, we will require submission of 300-600 individual trajectories for each target while submission of quantiles is optional. As a result, the only required targets for trajectories will be weekly incident hospitalizations and associate trajectory-level tag to index inital susceptibility conditions (S0). Additional indident death at national level, and emergency department visit are optionals. Estimates of cumulative counts can be obtained from weekly trajectories and hence we do not require trajectories for cumulative counts. Similarly peak targets (peak hospitalization magnitude and peak timing) can be reconstructed from weekly trajectories. Teams who wish to submit quantiles along with trajectories should provide quantiles for weekly and cumulative counts, as well as for hospitalization peak size and peak timing.
We require that simulations are paired across time, age group, targets (deaths, hospitalizations, etc.) and vaccine coverage levels. Any trajectory from scenario A should have a matched trajectory in scenarios B and C. Information on pairing structure will be gathered at the submission stage.
Each trajectory should be tagged by the proportion of susceptibles, S0, on the first day of simulations, Aug 10, 2025. This information should be provided at submission as an additional target.
-
Weekly reported state-level incident hospitalizations, will be based on the HHS/NHSN COVID and flu NHSN Weekly Hospital Respiratory Dataset. This dataset has previously been updated daily and covers 2020-2025. Weekly hospitalizations should be based on the “previous_day_admission_influenza_confirmed” variable, without any adjustment for reporting (=raw data from NHSN dataset to be projected). A current version of the weekly aggregated data has been posted here. Note that the dataset was paused May-Nov 2024 so this period has notoriously low reporting.
-
No case targets
-
No infection target
-
All targets should be numbers of individuals, rather than rates.
In this round of flu projections, we do not have an epidemiological scenario axis so that all epidemiological uncertainty is at the discretion of modeling teams. To allow for post-hoc comparison of uncertainty within and between models and trajectories, we will assign a tag to each set of paired simulations based on a key epidemic driver (where a set of paired simulations is a replicate of a particular model simulation across the 3 vaccine coverage levels in scenarios A-C). The tag will consist of initial susceptibility to infection proportion, which is a relatively well defined parameter for mechanistic approaches and an important driver of epidemic dynamics.
We define the trajectory-level tag, overall S0, as the all-age proportion of individuals who are susceptible to infection on the first day of simulations on Aug 10, 2025. Note that this is before vaccination takes place, so that S0 will be the same across all vaccine scenario levels even if a model assumes that vaccination reduces susceptibility.
Submission of an overall S0 will be required, in the form of a new required target (formatting guidelines in model-output README). This overall S0 encapsulates susceptibility across different possible types of model structures, including age groups, different levels of partial susceptibility, subtype or strains, etc. Below we have provided guidance on how to estimate overall S0 for models with different structures. Disaggregated estimates of S0 (eg by age or subtype) can be also submitted as optional targets along with overall S0, but only overall S0 is a required target.
-
Weekly national incident deaths, from the CDC multiplier model (i.e., true mortality burden from the pyramid). These are real-time model estimates updated weekly during the influenza season. The model relies on influenza deaths reported in the hospitals via the FluSurv-NET system, adjusted for under testing of flu in the hospital and the proportion of deaths occurring outside of the hospital. There is no state detail. Preliminary estimates from the CDC burden model suggest that between October 1, 2023 - June 15, 2024, between 25,000 – 74,000 deaths were due to flu illness or flu-related complications, while these numbers are 27,000-130,000 for the 2024-2025 season. We have provided several historical seasons of weekly death estimates for calibration from 2018-19 to 2024-25. Further, see here for summary estimates for past seasons.
-
Additional incidence target based on ED visits: Weekly reported state-level incident ED visits for influenza (NSSP dataset). This dataset covers 2022-present. Weekly hospitalizations should be based on the “percent_visits_influenza” variable. Note that this is a relative indicator of incidence, ie total ED visits for flu divided by total ED visits. Data broken down by age groups are graphed here.
-
Additional targets if submitting quantiles in addition to trajectories:
- Cumulative hospitalizations. Cumulative outcomes start at 0 at the start of projections, on Aug 10, 2025. This is similar to prior influenza rounds.
- Cumulative deaths. Cumulative outcomes start at 0 at the start of projections, on Aug 10, 2025. This is similar to prior influenza rounds.
- State-level peak hospitalizations.
- State-level timing of peak hospitalizations.
- Vaccination coverage, age population. Variability in the age distribution of hospitalizations between states is allowed as long as it aggregates to the scenario definition for the US overall (population weighted average).
- Variability in severity between states is possible.
- Prior immunity (due to a combination of vaccination and natural infection) can be the same or vary between states.
- At the discretion of the teams.
- Teams should include their best estimates of influenza seasonality in their model but we do not prescribe a specific level of seasonal forcing.
- No reactive NPIs to COVID-19 or influenza in this round; low level masking allowed at groups’ discretion.
- We leave seeding intensity, timing and geographic distribution at the discretion of the teams. In addition to the HHS hospital dataset, the flu portal dashboard is a good resource for state-specific information on strain circulation and epidemic intensity (e.g., weekly % positive, or weekly ILI*%positive), and can be used to adjust seeding.
- The mix of circulating strains at the start of the projection period is at the discretion of the teams based on their interpretation of the scenarios and best scientific judgement. Variation in initial prevalence between states is left at teams’ discretion.
All of the teams’ specific assumptions should be documented in meta-data and abstract.
| Scenario | Scenario name | Scenario ID for submission file ('scenario_id') |
|---|---|---|
| Scenario A. Business as usual vaccine coverage | usualVax | A-2025-07-29 |
| Scenario B. Low vaccine coverage | lowVax | B-2025-07-29 |
| Scenario C. Counterfactual no-vaccine | noVax | C-2025-07-29 |
- Due date: Friday Sep 5, 2025
- End date for fitting data: Between Saturday July 5, 2025 and Saturday Aug 9, 2025
- Start date for scenarios: Sunday Aug 10, 2025 (first date of simulated transmission/outcomes)
- Simulation end date: Saturday June 6, 2026 (43-week horizon)
- Desire to release results by early September 2025
- Simulation trajectories: We ask that teams submit a sample of at least 300 simulation replicates, paired across vaccination levels. Simulations should be sampled in such a way that they will be most likely to produce the same summary statistics as that quantile submitted. For some models, this may mean a random sample of simulations, for others with larger numbers of simulations, it may require weighted sampling. We also require that each trajectory should be tagged by the proportion of susceptible S0 on August 10, 2025.
- Geographic scope: state-level and national projections
- All states not required, US overall recommended.
- Results:
- Summary: Results must consist of a subset of weekly targets listed below; all are not required. Weeks follow epi-weeks (Sun-Sat) dated by the last day of the week.
- Weekly Targets (subset of: hospitalizations, deaths)
- Weekly incident
- Metadata: We will require a brief meta-data form, from all teams.
- Uncertainty:
- For trajectories (required submission): we require at least 300 trajectories.
- For quantiles (optional submission) We ask for 0.01, 0.025, 0.05, every 5% to 0.95, 0.975, and 0.99.
Groups interested in participating can submit model projections for each scenario in a PARQUET file formatted according to our specifications, and a metadata file with a description of model information. See here for technical submission requirements.
The target-data/ folder contains the target data in a hubverse compliant time-series format.
The data are automatically updated on Monday morning. The code to generate the
data is available in the src folder.
The past version of the time-series files are stored in the
auxiliary-data/target-data_archive
folder, with the date the data was archived append to the filename.
Weekly Hospital Respiratory Data (HRD) Metrics by Jurisdiction from the NHSN Weekly Hospital Respiratory Dataset will be used for incidence hospitalization target. The data are weekly.
The target to be projected is confirmed influenza hospital admissions, reported
as Total Influenza Admissions or totalconfflunewadm. This is a total for
the week running Sunday to Saturday. The same indicator is available for 6
age groups (0-4, 5-17, 18-49, 50-64, 65-74 and 75+ (numconfflunewadmped0to4,
numconfflunewadmped5to17, numconfrsvnewadmadult18to49, etc).
Influenza admission reporting became mandatory in this dataset on Feb-02-2021, and was until May 11, 2024. During May 11-Nov 2, 2024, reporting was paused and less than 75% of hospitals reported. Mandatory reporting resumed again on Nov 11, 2024. Accordingly, the data before February 2, 2021, and between May 1-Nov 11, 2024, should be treated with caution.
Unlike flu admissions, flu deaths are no longer mandatory to report in this system and hence a different source of data will be used for deaths.
If teams need more refined age-specific hospitalization data for calibration, or a longer dataset, they can apply the age distribution of flu hospitalizations available in Flusurv-NET to the all-age HHS rate. Flusurv-NET is CDC’s parallel and long-running hospitalization surveillance system, which is based on a smaller set of US hospitals (9%). We have provided state-specific time series here (2003-present). Because Flusurv-NET data is not available for all states, we recommend that teams use the national age distribution of hospitalizations estimated in Flusurv-NET and apply it to national and state-specific HHS data. Adjustment for demographic differences between states can be considered, but are not required.
Earlier analyses have shown that influenza hospitalization estimates are consistent between the NHSN and FluSurvNET surveillance systems in areas where both systems overlap. Further the ED data is also well correlated with the other two hospitalizations indicators. Magnitude can differ however. For instance, the Flusurv-NET data estimates ~15% fewer hospitalizations nationally than the NHSN system based on 2021-22 data; however these differences may vary during the 2025-26 season.
Weekly Emergency Department Visit from the National Syndromic Surveillance Program:
- NSSP Emergency Department Visits by demographic category since Oct 2022.
- NSSP Emergency Department Visits by state since Oct 2023.
The data are filter to keep only information about "Influenza".
- Delphi CMU group: See https://delphi.cmu.edu/flu/
- API documentation
- HHS flu hospitalizations
- Outpatient ILI (influenza and other similar illnesses) computed from medical insurance claims
- API documentation
The repository stores and updates additional data relevant to the Flu modeling efforts in the auxiliary-data/ folder:
-
Reports: Reports from Flu Scenario Modeling Hub rounds results. Each report contains an executive summary with key messages and results, and analyses of ensemble and individual projections.
-
Population and census data: National and State level name and fips code as used in the Hub and associated population size.
-
Rounds: Information on ongoing round and previous round available in the repository
For more information, please consult the associated README file.
We aim to combine model projections into an ensemble.
We are grateful to the teams who have generated these scenarios. The groups have made their public data available under different terms and licenses. You will find the licenses (when provided) within the model-specific metadata files in the model-metadata directory. Please consult these licenses before using these data to ensure that you follow the terms under which these data were released.
All source code that is specific to the overall project is available under an open-source MIT license. We note that this license does NOT cover model code from the various teams or model scenario data (available under specified licenses as described above).
Those teams interested in accessing additional computational power should contact Katriona Shea at [email protected].
Additional resources might be available from the MIDAS Coordination Center, please contact [email protected] for information.
Scenario modeling groups are supported through grants to the contributing investigators.
The Scenario Modeling Hub site is supported by the MIDAS Coordination Center, NIGMS Grant U24GM132013 (2019-2024) and R24GM153920 (2024-2029) to the University of Pittsburgh.
- Shaun Truelove, Johns Hopkins University
- Cécile Viboud, NIH Fogarty
- Justin Lessler, University of North Carolina
- Sara Loo, Johns Hopkins University
- Lucie Contamin, University of Pittsburgh
- Emily Howerton, Penn State University
- Claire Smith, Johns Hopkins University
- Harry Hochheiser, University of Pittsburgh
- Katriona Shea, Penn State University
- Michael Runge, USGS
- Erica Carcelen, John Hopkins University
- Sung-mok Jung, University of North Carolina
- Jessi Espino, University of Pittsburgh
- John Levander, University of Pittsburgh
- Samantha Bents, NIH Fogarty
- Katie Yan, Penn State University
- Wilbert Van Panhuis, University of Pittsburgh
- Jessica Kerr, University of Pittsburgh
- Luke Mullany, Johns Hopkins University
- Kaitlin Lovett, John Hopkins University
- Michelle Qin, Harvard University
- Tiffany Bogich, Penn State University
- Rebecca Borchering, Penn State University