PMC COVID-19 Forecasting Model, August 26, 2024
🧵1/8
We're in uncharted territory during the back-to-school period.
1.2 million people in the U.S. are getting infected per day.
1 in 41 people (2.5%) are actively infectious.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵2/8
The year-over-year graphs shows that we are experiencing much high transmission than during previous back-to-school periods.
This is a much higher and wider wave than Y1, Y2, and Y4 of the pandemic. Year 3's summer wave was very bad, but much earlier.
The 1-day isolation period, decline in most other mitigation, and a culture of denialism ("post pandemic," "Covid is over,") will fuel many infections among young people, teachers, school staff, and parents.
Expect absences at schools. Expect substantial workforce problems related to sick parents.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵3/8
Here's a zoomed in version of current transmission and the forecast.
Notice that transmission will likely remain >1 million infections per day for the next month and counting.
Note that the national peak appears just behind us. The CDC graphs may suggest that we're currently peaking, but note their data are >1 week old. The forecast provides an updated best estimate for today. The peak and shape of the observed wave could change as CDC and Biobot retroactively correct numbers (perhaps a bigger, earlier spike). Alternatively, Biobot has been slow to update their data, and their data were suggesting a later peak. However, I suspect the current picture won't change terrible much in hindsight.
We're seeking a less spiky, more plateau-like peak because there is substantial regional variation. The West and South regions have extremely high transmission and have likely peaked. The Midwest and Northeast are lower but rising.
Keep an eye on regional or local data, and overall remember that local peaks could occur weeks before or after the national peak. Also, 50-60% of transmission commonly occurs on the back end of waves.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵4/8
These statistics document the public health burden of the ongoing pandemic. Examples:
🔹Higher transmission than 86% of the pandemic
🔹34 million infections the next month
🔹>50% chance of a Covid exposure in a class of 30
PMC COVID-19 Forecasting Model, August 26, 2024
🧵5/8
This is the current heat map for the U.S. The data are from the most recent CDC update.
ALT text includes estimates of the percentage actively infectious in each of the deep red states.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵6/8
These are estimates of the percentage of each region of the U.S. that is actively infectious with Covid.
For example, while about 2.5% of the U.S. is actively infectious, the figures are higher in the South and West regions, and lower in the Midwest and Northeast (though catching up).
The estimates on the left re-weight the CDC data by combining Biobot and CDC data and carry forward to the current date using modeling, even though the public data always have a reporting lag.
The values on the right are purely derived from the most recent data reported on the CDC website (>1 week old). Anyone can compute these. Find the current CDC level, and multiply it by the PMC Multiplier. Example in next Tweet.
In either case, you'll note the estimates are similar. Feel free to report either or a range.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵7/8
>5% of people in Oregon, DC, Minnesota, and Arkansas are estimated to have infectious Covid. 12 other states are close behind.
You can take the CDC levels for a geographic entity and multiple it by the PMC Multiplier to get an estimate of the percentage of the population infectious in that region.
Here are the states with the highest transmission, acknowledging the most recent CDC data are >1 week old.
Just take the CDC level and multiply by 0.329, e.g., 23.82 x 0.329 = 7.8, meaning 7.8% are infectious. Take 100/7.8 to get the 1 in ____ infectious number.
Note, these are point estimates. Assume broader confidence intervals when focusing on smaller geographic units, especially if they have fewer sites reporting. These should serve heuristic value in local conversations, particularly in states that are experiencing very high transmission.
PMC COVID-19 Forecasting Model, August 26, 2024
🧵8/8
The full PMC Dashboard documents the ongoing toll of the COVID-19 pandemic. Please screenshot, improve, and share any of this work across platforms. #MaskUp
During this 12th COVlD wave, the CDC reports 1-in-3 states have "High" or "Very High" levels.
PMC estimates the proportion of residents actively infectious (prevalence):
◾️USA: 1 in 67
◾️IA: 1 in 27
◾️MI: 1 in 25
◾️IN & CT: 1 in 23
◾️ME: 1 in 21
◾️OK & SD: 1 in 17
🧵1/
On average, Americans have have 5.0 cumulative SARS-CoV-2 infections.
This week's infections are expected to result in 1/4 to 1 million new #LongCOVID conditions and ≈2,000 excess deaths.
🧵2/
The wave peak is now estimated >10% higher than last week at 1.2 million new daily infections, nearly double the Delta wave.
We expect sustained high transmission (≈600,000 to 750,000 new daily infections) the next few weeks as COVlD circulates through schools/families.
🧵3/
Based on today's CDC & Biobot data, we estimate the following for the week of Jan 19:
🔸1 in 52 people in the U.S. actively infectious
🔸25% chance of exposure in a room of 15 ppl
🔸Nearly 1 million new daily infections
🔸5 cumulative infections per person all-time (avg)
🧵1/5
Transmission estimates have been marginally corrected upward.
11 states have Very High COVlD levels:
🔸PA: 1 in 25 estimated actively infectious
🔸MI: 1 in 23
🔸OH & KY: 1 in 22
🔸SD: 1 in 20
🔸NE & IA: 1 in 18
🔸IL & ME: 1 in 17
🔸IN: 1 in 16
🔸WV: 1 in 11
🧵2/5
We're in the middle of a 12th COVlD wave.
The peak has likely passed, but with students headed back to school, transmission is expected to remain high for at least the next several weeks.
The size of the winter COVlD wave has been revised upward as post-holiday data come in.
We estimated 1 in 55 people in the U.S. are actively infectious.
🔥WV: 1 in 14
🔥IN: 1 in 15
🔥MI & OH: 1 in 21
🔥MO: 1 in 22
🔥CT: 1 in 24
🔥KS: 1 in 25
🔥MA & IL: 1 in 27
Quick 🧵 1/4
Nationally, we are seeing an estimated 892,000 new daily SARS-CoV-2 infections, meaning a 1 in 4 chance of exposure in a room of 15 people. Risk varies considerably by state.
We are approaching an average of 5 infections per person since pandemic onset.
🧵 2/4
We are in the 12th COVlD wave of the U.S.
Current transmission is higher than 68% of all days since the pandemic onset in 2020.
🧵 3/4
You might not have heard, but the northeastern U.S. is in a COVlD surge.
We use wastewater levels to derive estimates of the proportion of people actively infectious in each state (prevalence), e.g., 1 in 24 people in Connecticut.
We told you that 109,000-175,000 Americans would died of COVID (excess deaths) in 2025.
Today, the CDC estimates 101,000 deaths/year (flat from Oct 2022 to Sep 2024), and likely higher when considering more nebulous non-acute excess deaths (heart attack 6 months later). 1/5
The CDC estimates are actually higher than I would have guessed, given their methodology, which models estimates based on easily countable factors in healthcare and expert input on multiplier values. It lends credence to the PMC upper bound of excess deaths of 175,000/yr.
2/5
What's troubling is the CDC has annual mortality flat. My expectation based on mortality displacement and Swiss Re data is that it should be declining. If is stays flat, we're running on something like breast+prostate cancer or lung cancer deaths per year in perpetuity.
3/5