COVID-19 Deaths Are On The Decline, But Maybe Not For The Reasons You Think

First, the good news: Consistent with my forecast back in November, I am calling the peak of the pandemic, and showing a decline in deaths for February compared to December and January. I am still projecting an effective end to the pandemic in Q3 of this year.

The bad news is the Fall wave was brutal -- and it isn't over yet. COVID has been the #1 killer in the US since December 3. For reference, each week, Heart Disease claims about 38 lives per million in the US. Last week, COVID-19 claimed 67 per million, and peaked at about 71. Some states were far worse than the US average, peaking at over 100. While COVID-19 deaths are on the decline, COVID-19 will continue to be the leading cause of death in the US through the end of this month -- but COVID-19 deaths are starting to decline.

Some new analysis I performed in January suggests herd immunity is higher than the 70% figure shared as a rule of thumb. COVID-19 Herd Immunity, based on my analysis, appears to be more likely around 85%.

Back to some good news: Vaccinations are on the rise, with over 32 million Americans receiving their first dose of the vaccine, and over 9.5 million receiving their second dose. For the past three weeks, the US have been vaccinating over 1.3 million people per day.

It might be tempting to cheer the arrival of the vaccines and credit the vaccine for the decline in deaths, however, that is not what is happening right now. The predictable decline in cases and deaths is almost entirely explained by the social calendar. Thanksgiving, Christmas, New Years -- these are social holidays where people gather with family and friends indoors, sometimes without masks. The surge in cases and deaths was entirely predictable. My forecast for End of Year 2020 was almost exactly right (99% accurate, off by less than 1,000 in predicting 350,000 deaths at year's end). My January forecast was off by only 8% (it was a little worse in terms of the death toll than I forecasted, ending with 441,296 deaths - see chart).

CHART NOTE: Briggs-3 Forecast performed better than Amherst's forecast hub ensemble, which was once again far too low -- missing the range completely. Briggs-3 also performed better than IMHE, which was once again too high -- off by 16%.

Over the next few weeks, we should see a more marked decline in deaths. Maintaining this downward trend depends on us.

Where we go from here depends on the following factors:

1. Do we continue to social distance and wear masks through the summer, to stay ahead of the COVID-19 variants, and to give time for Vaccinations to truly make a difference?
2. Do we maintain over 1.3 million vaccinations a day, and achieve 75% or better vaccinations among 65+ age cohort before the end of the Summer?
3. Do we improve vaccinations in our urban centers, and among Black and LatinX Americans, which my research suggests are getting left behind in vaccinations?

2021 Forecast For Q1 & Q2

My November 2021 forecasts (Briggs-3) represents a decelerations in deaths (shown in the red line in the chart). I projected the peak right around end of January, in terms of deaths in the US and it appear that might have been off by about a week -- it appears we are just passing the peak this week. In my November forecast (Briggs-3), vaccinations are shown with the green dashed line along with natural recoveries (blue line) are shown, and will put downward pressure on the virus, reducing deaths throughout Q1 and Q2. Vaccinations are running about 10 days behind the trend line, but catching up a little more each day. We should catch up with my forecast by the end of this month.

CHARTL Snapshot of Briggs-3 forecast back in November.

Why I Raised My Forecast of COVID19 Deaths, And What It Means To You

In the November Briggs-3 forecast I expected Herd Immunity to kick in at about 70% based on the conventional wisdom. However, as more data accumulated, I could perform my own Herd Immunity analysis and draw my own conclusions. TLDR: my analysis shows Herd Immunity is likely higher than 70% -- probably around 85%.

Here is the analysis that led me to that conclusion: I decided to examine the relationship between cumulative case growth by state and my calculation of those who have recovered and should be immune. I expected to find states that were hit hardest early on, where the infection rate was higher, to show significantly lower growth rate compared to states that had a lower infection rate based on the number of COVID-19 deaths, converted to a immunity estimate. (I divide deaths by my Infection Fatality Rate estimate of 0.4% to arrive at an infection rate with 21 days lag.)

I also considered if there might be differences in population density. For example, perhaps we need less herd immunity in less densely populated areas like Idaho than we do in densely populated Rhode Island. So, I included population density in the analysis (the size of the bubbles). Here is a snapshot of my analysis on January 12th.

How to read this chart: The dots are States, and the size of the bubble is population density. The Y-axis is cumulative case growth rate on Jan11, and the x-axis is total already immune. We can see it is near 60% in New Jersey (the big dot farthest to the right). The fact the dots on the far right, are still growing at around 4% suggested to me that Herd Immunity is higher than 70%. We should recognize we need mass vaccinations to end this pandemic. If you are interested in seeing an update of this analysis, with the labels for states, check out www.speakerrex.com/dash.html and click on Herd Immunity Analysis (the second page in that section). This dashboard is my personal workspace to look at trends. It updates automatically each day.

To recap my herd immunity analysis, what we see is states with higher infection rates do grow at a slower rate compared to states with a lower infection rate -- but, the regression line isn't crossing zero growth near 70%. It is pointing to a higher herd immunity threshold.

Granted, this growth is against the backdrop of Thanksgiving, Christmas and New Years, which are uniquely dense indoor gatherings, ideal for the spread of the virus. Still, if we had Herd Immunity near 70%, areas like New Jersey, New York, etc. should have much lower cumulative case growth rate than what I was observing in this data.

Recently, the entire rate of growth has declined, but this is not related to herd immunity, it is related to the difference in our social interactions. So, what happens if too many people assume we've reached herd immunity and begin to interact in-person, without masks, with their friends and family similar to the levels of activity during Thanksgiving, Christmas and New Years? The simple answer is more infections and more deaths.

At the same time, we were behind my forecast for vaccinations. We are catching up, but I am concerned about the 2nd dose vaccination rates, as well as vaccinations in our urban centers and among Black and LatinX Americans. Add to this continued polarization -- polls showing Republican Women in particular with vaccine hesitancy. These factors led me to raise the forecast.

Here is a table to compare Briggs-3 and Briggs-3.1 Forecast, along with other leading forecast models:

Basically, I am forecasting an additional 86,000 lives lost to COVID-19 in the next 90 days. I changed the assumption of cumulative case growth from 2% in February and March to 4%, April and May from 1% to 2%. Both predict a peak in deaths around now, with a steady decline through Q1 and Q2. But, the cumulative case growth rate makes a difference.

A forecast is not destiny. We can be vigilant about social distancing and mask wearing for a few more months and stay on track the Briggs-3.0 track, with a slower rate of growth in cases and deaths. We can support the vaccination efforts, especially among Black and LatinX Americans in our urban centers. I'll share my analysis on why I am concerned about vaccination rates in our urban centers, and among Black and LatinX in my next post.

What are the implications to you, your family and your business?

First, take care of you and your family during these next few months and help navigate the vaccine scheduling complexities. My research suggests meaningful risk, especially among those 65 and older. Their risk of dying this year of COVID is about 1 in 138 (their risk of dying of COVID between April 1, 2020 and Dec 31, 2021 is about 1 in 90!). We have already lost 1 out of every 149 people 65 and over. In states like New Jersey, it is worse, with 1 out of 81 people 65 and older are dead of COVID-19. By the end of this year, my forecast is it will be 1 out of ever 50 people 65 and older in New Jersey will have died of COVID-19.

If a person, 65 and older, get both doses of the vaccine, that risk drops to less than 1 in 23,000 -- less than the risk of dying in a auto accident.

Share these stats with your loved ones. Right now, it is tough to get the vaccine. Help them stay positive and persistent if they are struggling to get the vaccine. Consider getting to know your local pharmacies that is providing vaccines, and what time they release any unused doses. There are no shows, and some pharmacies have a policy using these doses for qualifying people that are standing by/on their call list.

Second, the new variants are no joke.

In the immediate term, consider the quality of your mask. Consider if double-masking if you don't have access to N94 or N95. Consider the risk/benefit of different interactions with non-household members. Continue to be selective. If you decide to get together with family members not in your household, consider testing prior to the visit.

If you get the vaccine, keep in mind you need both doses and at least a week after the second dose to get to the full protection. Even after vaccination, masking up will help reinforce positive behaviors of those not yet vaccinated.

In the longer term, I expect an effective end to the pandemic by the end of Q3. That doesn't mean COVID19 completely goes away. Rather, it means the death rate will drop to such a low point that it no longer meets the definition of pandemic (10% elevated death rate in a broad geography). You can see my forecast chart below and see the end of the pandemic is within sight. Perhaps social businesses like Theatres, Travel, etc see a significant re-bound as pent up demand converts back to business.

On the positive side, vaccinations are running at more than 1.3 million per day. We've crossed over to where there are now MORE vaccinations than confirmed cases of COVID19 in the US. That is great news!

We are catching up to the vaccination forecast I made back in November. My forecast of vaccinations, and their efficacy, is what let me calculate the end of the pandemic in Q3. If we can continue to accelerate vaccinations, especially among 65+, we can see an effective end to the pandemic by the fall, before we enter the risky fall and winter months. Keep in mind, we will win the battle against the pandemic with the 65+ population, which account for 80% of the deaths. This is the group that needs 75% or better vaccination rate to effectively end the destruction of SARS-CoV-2.

At the same time, the challenging part of vaccination is not getting the first 40% or so vaccinated. It is getting the next 30% or so that brings us above 70% vaccinated. The Ad Council is doing important leadership work. We are all going to need to pull together to accomplish this monumentally meaningful effort.