This repository is designed to collect forecast data for the COVID-19 Forecast Hub run by the US CDC. The project collects forecast for two datasets:

1. weekly new hospitalizations due to COVID-19, and
2. weekly incident percentage of emergency department visits due to COVID-19 (optional, beginning June 18, 2025).

If you are interested in using these data for additional research or publications, please contact covidhub@cdc.gov for information regarding attribution of the source forecasts.

During the submission period, participating teams will be invited to submit national- and jurisdiction-specific (all 50 states, Washington DC, and Puerto Rico) probabilistic nowcasts and forecasts of the weekly number of confirmed COVID-19 hospital admissions during the preceding epidemiological week ("epiweek"), the current epiweek, and the following three epiweeks.

The weekly total COVID-19 admissions counts can be found in the totalconfc19newadm column of the National Healthcare Safety Network (NHSN) Hospital Respiratory Data (HRD) dataset.

NHSN provides a preliminary release of each week's HRD data on Wednesdays here. Official weekly data is released on Fridays here. For more details on this dataset, its release schedule, and its schema, see the NHSN Hospital Respiratory Data page.

Beginning June 18, 2025, the COVID-19 Forecast Hub will also accept probabilistic nowcasts and forecasts of the proportion of emergency department visits due to COVID-19. This new target represents COVID-19 as a proportion of emergency department (ED) visits, aggregated by epiweek (Sunday-Saturday) and jurisdiction (states, DC, United States). The numerator is the number of visits with a discharge diagnosis of COVID-19, and the denominator is total visits. This target is optional for any submitted location and forecast horizon.

The weekly percent of ED visits due to COVID-19 can be found in the percent_visits_covid column of the National Syndromic Surveillance Program (NSSP) Emergency Department Visits - COVID-19, Flu, RSV, Sub-state dataset. Although these numbers are reported in the percentage form, we will accept forecasts as decimal proportions (i.e., percent_visits_covid / 100). To obtain state-level data, we filter the dataset to include only the rows where the county column is equal to All.

We are working to make the Wednesday release of this dataset available on data.cdc.gov. Until then, we will update the dataset every Wednesday in the auxiliary-data/nssp-raw-data directory of our GitHub repository as a file named latest.parquet. These Wednesday data updates contain the same data that are published on Fridays at NSSP Emergency Department Visit trajectories and underlie the percentage ED visit reported on the PRISM Data Channel's Respiratory Activity Levels page, which is also refreshed every Friday. The data represent the information available as of Wednesday morning through the previous Saturday. For example, the most recent data available on the 2025-06-11 release will be for the week ending 2025-06-07.

The Challenge Period is rolling.

Participants will be asked to submit nowcasts and forecasts by 11PM USA Eastern Time each Wednesday (the "Forecast Due Date"). If it becomes necessary to change the Forecast Due Date or time deadline, CovidHub will notify participants at least one week in advance.

Weekly submissions (including file names) will be specified in terms of a "reference date": the Saturday following the Forecast Due Date. This is the last day of the USA/CDC epiweek (Sunday to Saturday) that contains the Forecast Due Date.

Please note the following updated deadlines during the holiday period:

• Forecasts for reference date 2025-12-27 are due on 2025-12-29 (extended deadline), with expected data release on 2025-12-29 (holiday schedule).
• Forecasts for reference date 2026-01-03 are due on 2026-01-04 (extended deadline), with expected data release on 2025-12-31 (regular schedule).

Participating teams will be able to submit national- and jurisdiction-specific (all 50 states, Washington DC, and Puerto Rico) predictions for following targets.

1. Quantile predictions for epiweekly total laboratory-confirmed COVID-19 hospital admissions.
2. Individual forecast trajectories for epiweekly total laboratory-confirmed COVID-19 hospitalizations over time (i.e sampled trajectories).
3. Quantile predictions for epiweekly percent of emergency department visits due to COVID-19.
4. Individual forecast trajectories for epiweekly percent of emergency department visits due to COVID-19 over time (i.e sampled trajectories).

Targets 2, 3 and 4 are optional for any submitted location whereas target 1 (quantile predictions for epiweekly COVID-19 hospital admissions) is mandatory for any submitted location and forecast horizon. Teams are encouraged but not required to submit forecasts for all weekly horizons or for all locations.

Teams can submit nowcasts or forecasts for these targets for the following temporal "horizons":

• horizon = -1: the epiweek preceding the reference date
• horizon = 0: the current epiweek
• horizon = 1, 2, 3: each of the three upcoming epiweeks

We use epiweeks as defined by the US CDC, which run Sunday through Saturday. The target_end_date for a prediction is the Saturday that ends the epiweek of interest. That is:

target_end_date = reference_date + (horizon * 7)

Standard software packages for R and Python can help you convert from dates to epiweeks and vice versa:

• lubridate
• MMWRweek

• epiweeks
• pymmwr

Detailed guidelines for formatting and submitting forecasts are available in the model-output directory README. Detailed guidelines for formatting and submitting model metadata can be found in the model-metadata directory README.

Pull requests (PRs) into the Hub repository to register a new model or modify an existing model's metadata must always be reviewed and merged manually.

PR that submit forecasts for an existing model can be reviewed and merged automatically if the submission content passes automated validation checks and the submitting individual has been preregistered as an authorized submitter for the model.

To authorize one or more individuals to submit forecasts for a given model, add their github usernames to the designated_github_users field in the model's metadata.

To facilitate auto-merge of valid PRs, we suggest the following workflow:

• Submit metadata first: Create a PR adding your model metadata file to the model-metadata directory by 4 PM USA Eastern Time on the Wednesday you plan to submit your first forecast.

• Include designated_github_users: In your metadata YAML file, include the GitHub usernames of all team members responsible for forecast submission in the designated_github_users field. We use this to ensure changes to model outputs are made by valid model contributors.

Once initial metadata PR is approved and merged, subsequent PRs that submit forecasts will be merged automatically, provided all checks pass.

We have made changes from previous versions of the COVID-19 Forecast Hub challenges to align COVID-19 forecasting challenges with influenza forecasting run via the Flusight Forecast Hub.

Both Hubs will require quantile-based forecasts of epiweekly incident hospital admissions reported into NHSN, with the same -1:3 week horizon span. Both will accept these forecasts via Github pull requests of files formatted according to the standard hubverse schema. The Hubs also plan to share a forecast deadline of 11pm USA/Eastern time on Wednesdays.

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.