---

Plot Options:

Download Scenarios in Graph

Use slide below graph to adjust date range.

Nowcasts

UC Irvine - Rt-estim

This Rt-estim-gamma model by Isaac Goldstein and Volodymyr Minin at UC Irvine and Jon Wakefield at University of Washington is a branching process inspired model where cases are modeled as noisy realizations of latent unobserved incidence. The number of diagnostic tests is incorporated as a covariate into the surveillance model to adjust for changes in testing volume over time.

Code Repository
Raw Data
Modeling methods

---

UC Irvine - Wastewater

This model by Isaac Goldstein and Volodymyr Minin at UC Irvine uses a modified compartmental model where viral pathogen genome concentrations are modeled as a noisy realization of a weighted sum of the number of infectious and recently recovered individuals.

Code Repository
Raw Data
Modeling methods

---

UCSF

ca-covid-r.info

The UCSF model uses raw data from three sources: new cases by county from California Health and Human Services; wastewater readings from the Cal-SuWers network (“Wastewater (C)”), BioBot Analytics (“Wastewater (B)”) and WastewaterScan (“Wastewater (W)”). The Wallinga-Teunis technique is used to estimate reproduction numbers, corrected for right censoring of the case detection data.

Raw data
Modeling methods

---

CDC Center for Forecasting and Outbreak Analytics

Rt estimates are derived from daily counts of new COVID-19 hospitalizations.To estimate Rt, CDC fit Bayesian models to the data using the R packages EpiNow2, epinowcast, or using Stan models developed by the CDC Center for Forecasting and Outbreak Analytics (CFA).

Raw data
Modeling methods

---

UC Santa Cruz

The County of Santa Cruz COVID-19 model is a probabilistic SEIR model with compartments for hospitalizations and ICU visits. Population-averaged COVID-19 data from three wastewater plants, along with hospitalization, ICU and death data, are used to inform model estimates using Bayesian techniques. These estimates include the effective reproductive number Rt and hospitalization projections, also available at the County of Santa Cruz Health Services Agency website.

County of Santa Cruz Health Services Agency website
Model code repository

---

UC Davis Wastewater

Developed in collaboration with the Sophia and HCVT teams at UC Davis, this model estimates the Rt for COVID-19 using a sequential Bayesian method that models the test positivity rate (TPR) based on the concentration of SARS-CoV-2 RNA in wastewater (WW). This method is designed to adapt to changes in virus dynamics, which provides a comprehensive understanding of TPR trends. To model the TPR, a Bayesian sequential approach employs a Beta regression model with the concentration of SARS-CoV-2 RNA in WW as a covariate. This modeling approach provides TPR estimates which are used to calculate the Rt estimates.

Modeling methods
Model code repository
Raw data

---

Santa Clara County Wastewater

The Santa Clara County Wastewater Rt model is derived from the work of Huisman et al., but includes several modifications. The model first fits an adaptive spline (bs = 'ad' in mgcv's gam() function) to the normalized wastewater data to remove noise and recover the underlying trend of interest-- log-transformed input data undergoes smoothing using a TOBIT (censored normal) error distribution; dates with concentrations of 0 are treated as non-detects with censored outcomes, and the limit of detection is initially estimated using the lowest nonzero value in recent data. Smooth fits are used to detect outliers, i.e., observations with extreme residuals and thus unlikely to be in 95% of time series generated under the smooth model. Any outliers so flagged are resampled. Then the wastewater concentrations are converted into an estimated time series of infection events by deconvolving the smoothed wastewater signal with an estimate of the shedding load distribution. For deconvolution, following Huisman’s initial approach is used (available through estimateR's deconvolve_incidence() function) with a gamma-distributed incubation period with mean 3.58 days and SD 1.77 days, combined with a fecal shedding load distribution from the systematic review of Benefield et al., gamma distributed with mean 6.7 days and SD 7 days. Once the time series of infection events has been estimated, Rt is recovered based on an estimate of the generation interval distribution using standard renewal equation methods with the default priors for Rt in EpiEstim to have an expected value of 1. Rt estimates for each sewershed are converted to the overall county estimate using a population-weighted average. Inference is then conducted by calculating bootstrap quantiles and odds of agreement of epidemic trajectory with the point estimate based on original data.
Input data: PMMoV-normalized target gene concentrations in wastewater solids provided by Wastewater SCAN.

Modeling methods
Raw data

---

CDPH Wallinga Teunis

The Wallinga and Teunis method is based on the probabilistic reconstruction of disease transmission chains to estimate the number of secondary cases per infected individual. It is robust and only requires only case incidence data and the distribution of the serial interval (the time between the onset of symptoms in a primary case and the onset of symptoms of secondary cases) to estimate the % of cases from each day which were infected by cases on previous days. This data used for the estimate is derived from the CDPH line list of cases indexed by the date of symptom onset. The serial interval is given by a Weibull distribution with mean interval of 5.5 days. The methodology is implemented in the R programming language by the R0 package.

Modeling methods

---

CDPH Cislaghi

This method calculates R-effective as the number of newly diagnosed cases on day s over the number of newly diagnosed cases on day (s - T), where T is the incubation time. CDPH uses officially reported California case data indexed by episode date, and an incubation period (T) of 3 days. To smooth the curve and to avoid strong daily variations due to noise, R-effective was calculated as a seven-day moving average.

Modeling methods

---

CDPH Wastewater

estimateR Package

R forecast Package

Wastewater samples are collected and processed by WastewaterSCAN, an organization that partners with the Cal-SuWers Network to monitor wastewater at wastewater treatment plants in California. To reduce the effect of noise, the wastewater data is processed with spline smoothing followed by a square root transformation. To convert from viral shedding events in wastewater to a time series of infection events, the estimateR method uses a gamma delay distribution with a mean of 5 days and a standard deviation of 0.5 days (Huisman et al. 2022). Using the inferred infection time series, estimateR calculates the effective reproduction number by implementing a wrapper around the EpiEstim package (Scire et al. 2022). The generation time distribution input to EpiEstim is gamma distributed with a mean of 6.84 days, and a standard deviation of 4.48 days (Manica et al. 2022).

To account for lags introduced by deconvolution and wastewater data reporting, wastewater for each treatment plant is forecasted until the present day using a TBATS time series model (T: Trigonometric seasonality, B: Box-Cox transformation, A: ARIMA errors, T: Trend, S: Seasonal components) optimized by the forecast package. TBATS is an exponential smoothing model designed to forecast data with complex seasonal patterns (De Livera et al. 2011). This step provides a nowcasted, current-day estimate of the effective reproduction number. The effective reproduction number for each wastewater treatment plant is subsequently aggregated to the county level via population weighting.

CDC overview of wastewater surveillance

---

CDPH Hospitalizations

estimateR Package

R forecast Package

Statewide hospitalization data from the National Healthcare Safety Network (NHSN) is processed with locally estimated scatterplot smoothing (LOESS). Two delay distributions are then used to deconvolve from observed hospital admissions to inferred infections: (1) an estimate of the incubation period, based on Aguila-Mejia et al. (2022) and (2) the delay from symptom onset to hospital admission, based on the California COVID-19 case registry. Based on this inferred infection time series and a generation time distribution, estimateR calculates the effective reproduction number. The generation time distribution is given by a gamma distribution with a mean of 6.84 days and a standard deviation of 4.48 days (based on Manica, De Bellis & Guzzetta et al. 2022).

To account for lags introduced by deconvolution and NHSN data reporting, hospitalizations are forecasted until the present day prior to estimation of the effective reproduction number. Forecasting is executed using a TBATS time series model (T: Trigonometric seasonality, B: Box-Cox transformation, A: ARIMA errors, T: Trend, S: Seasonal components) optimized by the forecast package (De Livera, Hyndman & Snyder et al. 2011).

---

CDPH EpiNow2 - Cases

EpiNow2 uses a renewal equation-based model to estimate the time-varying reproduction number (R-effective) based on timeseries data by date of infection and an estimate of the generation interval. As most timeseries data are not based on date of infection, EpiNow2 uses the incubation period and data reporting delay distribution to convert the input timeseries to an infection date-based timeseries.

Code repository
Modeling methods
Model vignette

---

Ensemble Nowcast

The ensemble nowcast takes the median of all the nowcasts available on a given date and smooths it with a three-day moving average. The methodology aims to be robust to outliers and to avoid kinks in the median nowcast when an input source for the ensemble estimate is unavailable on a certain date. Ensemble confidence interval calculation includes only models with corresponding confidence interval estimates available.

---

Forecasts

COVID-19 Forecast Hub

covid19forecasthub.org

The COVID-19 Forecast Hub is run by the US CDC. The project collects forecast for the weekly new hospitalizations due to COVID-19. Forecasts are weekly, out to four weeks, at the state and national level.

Raw data
Code repository

---

UC Irvine - SIRS

vnminin.github.io

This model by Vladimir Minin and Damon Bayer at UC Irvine is a SEIR type model stratified by Omicron vs non-Omicron infection with hospitalization compartments. The model is calibrated to California hospital data from the California open data portal using data on COVID variant prevalence from GISAID."

Raw data
Code repository

---

ARIMA-based Model

R forecast Package

Model forecasts are the result of utilizing the forecast package's automatic ARIMA forecasting model. Note that the form of the model may vary between counties and over subsequent published forecasts.

Modeling methods

---

Holt’s Model

R forecast Package

The Holt's linear trend method extends simple exponential smoothing to allow the forecasting of data with a trend (Holt 2004).

---

Damped Holt’s Model

R forecast Package

The Damped Holt's model is an exponential smoothing model designed to damp erratic trends (Gardner & Mckenzie 1985).

---

NNETAR Model

R forecast Package

The neural network autoregression (NNETAR) model is a system of feed-forward neural networks with a single hidden layer and lagged inputs.

---

Harmonic Regression

R forecast Package

A dynamic harmonic regression model captures a seasonal pattern using Fourier terms with short-term time series dynamics handled by an ARIMA error (Young et al. 1999).

---

TBATS

R forecast Package

The acronym “TBATS” stands for: trigonometric seasonality; Box-Cox transformation; ARIMA errors; trend; and seasonal components. TBATS is an exponential smoothing model designed to forecast data with complex seasonal patterns (De Livera et al. 2011).

---

CDPH Human Judgement Forecast

This forecast is compiled from individual judgment forecasts from members of the CDPH modeling team. A probability distribution spline is fitted for each submission (in form of quantile predictions), and the median ensemble is calculated from all the forecasts in each epiweek for the final forecast.

---

Ensemble Forecast

The ensemble forecast takes the median of all the forecasts available on a given date and fits a smoothed spline to the trend. The methodology aims to be robust to outliers, and to avoid artifacts (i.e. abrupt kinks) when the median forecast switches from one source to another.

---

Scenarios

COVID-19 Scenario Modeling Hub

Scenario Modeling Hub

Scenario Hub Viz

Round 18

No booster scenario assumes no future recommendation to get additional booster doses or receive additional vaccination.

High-risk booster scenario assumes that the uptake of annual reformulated booster in high-risk groups (65+ and other individuals with underlying risk factors for severe COVID-19 outcomes) follows uptake observed for the booster dose during the 2023-24 season and negligible vaccination among others outside of high-risk groups.

All booster scenario that the uptake of annual reformulated booster in all groups follows uptake observed for the booster dose during the 2023-24 season.

Low immune escape scenario assumes immune escape occurs at a constant rate of 20% per year.

High immune escape scenario assumes immune escape occurs at a constant rate of 50% per year.

Scenario Hub Contributors

• Northeastern University MOBS Lab — GLEAM COVID
• University of Virginia — adaptive
• University of Notre Dame - FRED

Scenario Hub Ensemble LOP Untrimmed

The LOP untrimmed ensemble projection is calculated by averaging cumulative probabilities of a given value across submissions. All values are included in the average. From the resulting distribution, medians and uncertainty intervals are derived. Ensemble projection includes only those submissions that reported quantiles for their targets.

---

LEMMA

localepi.github.io

LEMMA provides scenarios to depict how Omicron might affect hospitalizations in California in the coming weeks and months.

Optimistic assumes the Omicron hospitalization rate in immunonaive is much lower than for Delta and length of stay is shorter than for Delta.

Central assumes the Omicron hospitalization rate in immunonaive is lower than for Delta.

Pessimistic assumes the Omicron hospitalization rate in immunonaive is slightly lower than for Delta and booster vaccine effectiveness is lower,.

LEMMA scenarios either include 'for COVID' only (no incidental COVID) or 'with COVID' (includes incidental COVID). The current model assumption is that 40% of COVID positive patients are incidental admissions (not primarily due to COVID).

LEMMA is an open-source SEIR model with compartments for hospitalization and symptom severity. The model is calibrated to hospitalization, ICU and death data using Bayesian methods. LEMMA provides credibility intervals and scenarios for future cases, hospitalization, ICU, deaths and Rt. Users can download the LEMMA package and input their own data and priors for parameters using R or a simple Excel interface.
Raw data

---

Columbia

Columbia University, Shaman Group

The Shaman group from Columbia University projects nationwide, county-level estimates of R-effective, daily new confirmed case, daily new infection (both reported and unreported), cumulative demand of hospital beds, ICU, and ventilators, as well as daily mortality (2.5, 25, 50, 75 and 97.5 percentiles).

No change assumes that current contact rates will remain unchanged in the future.

1x increased transmission (5_1xbeta_Lmin0p5) assumes a one-time 5% increase in contact rates at the start of the projection. The following week, the effective reproduction number R is set to 1. As the simulation progresses, R_eff decreases as the fraction of susceptible individuals in the population decreases.

Raw data
Modeling methods

Variants

CDPH Projection Models

Models based on previously observed variant proportions are used to generate real-time estimates of variant proportions. Each model is applied separately to both the integrated genomic epidemiology dataset (IGED) and Terra.

The logistic proportion model is based on methods presented in Althaus et al 2021. It uses the last 60 days of data to estimate a trend in variant growth assuming it is following a logistic pattern.

The multinomial spline model is based on methods presented in Davies et al 2021. It uses a multinomial logistic regression with cubic spline terms to estimate temporal trends in the frequency each variant is observed.

The variant simple growth model is currently only estimated statewide and uses variants specific effective reproduction numbers (R_eff) and simple growth models to estimate variant specific case counts. Variant proportions are then estimated from these case counts as the proportion of projected new cases caused by each variant. The model to estimate variant specific reproduction numbers is based on methods presented in Figgins & Bedford 2021.

The ensemble mean model is the average of the available projection models listed above. The range displayed in the plot is the lowest to highest uncertainty estimates from the input models.

Note: Particularly in regions with smaller populations, uncertainty in the actual proportions and projections should be heeded. Uncertainty can be viewed by hovering over the relevant data in the plots.

---

Plot Options:

Download Scenarios in Graph

---

Download Interactive Plot

Use slide below graph to adjust date range.

Forecasts

CDC FluSight

CDC FluSight GitHub

CU Ensemble

A weighted ensemble of five component models: 1) SIR compartmental model; 2) historical-drift model where the mean and variance of the drift for each week and location are derived by sampling first-order lags in incidence from historical data; 3) Exponential-Smoothing model that decomposes a time series to a baseline, trend and seasonal components; 4) Light Gradient Boosting Machine (LightGBM), a ML ensemble decision tree method designed for classification and regression tasks that has been effectively adapted for time series forecasting; and 5)Temporal-Fusion-Transformer model- a transformer-based neural-network architecture tailored for time series forecasting.

Predictive Science (PSI)

Deterministic fit of a compartmental SIHR model to weekly, CA region level, hospitalization incidence profiles, followed by stochastic simulations using the inferred parameter distributions. The ‘H’ compartment is divided into two compartments to ensure the correct generation time for the Infectious compartment and proper fitting of the portion of infectious individuals that are hospitalized. For each of the five CA regions, the model fit is inferred by an MCMC procedure and run forward at a weekly cadence.

UGA Flucast

A spatial time-series model that uses the R-INLA package for estimating a jointly fitted time-series model that includes seasonality terms, state-specific connectivity, and a random walk component.

UMass Flusion

Ensemble of statistical and machine learning time series models.

CDC Ensemble

The CDC Ensemble from the CDC FluSight Challenge includes all available flu forecasts for the state of California.

CDPH Models

These forecasts consist of four time series forecast approaches:

• Autoregressive integrated moving average (ARIMA): uses a weighted linear sum of recent past observations or lags.
• Holt’s linear trend method: extended simple exponential smoothing to allow the forecasting of data with a trend (Holt 2004).
• Damped Holt’s: an exponential smoothing model designed to damp erratic trends (Gardner & Mckenzie 1985).
• Neural network autoregression (NNETAR): feed-forward neural networks with a single hidden layer and lagged inputs
• Harmonic Regression: a dynamic harmonic regression that models a seasonal pattern using Fourier terms with short-term time series dynamics handled by an ARIMA error (Young et al. 1999).
• TBATS: the acronym “TBATS” stands for trigonometric seasonality; Box-Cox transformation; ARIMA errors; trend; and seasonal components. TBATS is an exponential smoothing model designed to forecast data with complex seasonal patterns (De Livera et al. 2011).
• CDPH Human Judgement Forecasts: This forecast is compiled from individual judgment forecasts from members of the CDPH modeling team. A probability distribution spline is fitted for each submission (in form of quantile predictions), and the median ensemble is calculated from all the forecasts in each epiweek for the final forecast.

CDPH Ensemble

The ensemble model is the median value of all available CDPH models. The lower and upper 95% prediction intervals for the ensemble are calculated by interpolating quantiles of all available models for a given geographic region.

Scenarios

Flu Scenario Modeling Hub
Scenario Hub Viz

Round 1 2024-2025 Scenarios

The subtype H3N2 dominant scenarios assume that the next season is dominated by influenza A/H3N2, indexed on the age dynamics and severity based on the 2017-2018 season.

The subtype H1N1 dominant scenarios assume that the next season is dominated by influenza A/H1N1, indexed on the age dynamics and severity based on the 2019-2020 season.

The high vaccine coverage scenarios assume that the vaccine coverage is 20% higher than in the 2022-2023 flu season in all age groups and jurisdictions.

The medium vaccine coverage scenarios assume that the vaccine coverage is same as in the 2022-2023 flu season in all age groups and jurisdictions. Overall US coverage is about 49% in this scenario.

The low vaccine coverage scenarios assume that the vaccine coverage is 20% lower than in the 2022-2023 flu season in all age groups and jurisdictions.

Scenario Hub Contributors

California Department of Public Health – FluCAT

The model is a stochastic, mechanistic model implemented via the tau leap method for the state of California.

Northeastern University MOBS Lab — GLEAM Flu

GLEAM is an age-structured metapopulation model that includes high resolution demographic data, short-range commuting flows, domestic air traffic, and age-specific contact patterns.

National Institutes of Health – Flu_TS

Flu_TS is a dynamic harmonic regression model with ARMA errors and exogenous covariates for cumulative vaccination coverage, seasonal vaccine effectiveness, weekly A/H3N2 circulation, and weekly influenza infections.

Predictive Science Inc. – PSI-M2

For each state/territory, scenario projections are generated using a mechanistic model with S[Sv]I[Iv]HR compartments, where 'v' subscripts indicate vaccinated and each compartment is age-stratified.

University of Southern California (USC) — SIkJα

Discrete heterogeneous rate model where rates are learned using regression weighing the recently seen data higher. Past seasons' rates are used as a proxy for seasonality.

University of Virginia Biocomplexity Institute – PSI-M2

Metapopulation simulation over age- and spatially-stratified synthetic contact network.