When it comes to the highly debated topic of COVID-19 herd immunity, many questions still remain. Where are we today? How can we find out what things look like in our communities? How much further do we have to go to reach a threshold where the virus does not easily find its next susceptible person?
It is thought that most people who become infected will be immune from becoming infected again, at least for some period. Therefore, tracking estimated immunity within communities and getting to a better understanding of when herd immunity is reached will remain important even if/when a vaccine becomes available. Of note, getting infected a second time has been well documented in a handful of people, though it's thought to be very rare.
The threshold needed to reach herd immunity for COVID-19 is not known at this point. Herd immunity for COVID-19 (hC) begins with a relatively simple and classic definition, where hC = 1 – 1/R0, using the basic reproduction number (R0) to identify the threshold at which point the epidemic stops growing – note, this is not the point where the pandemic ends as Carl Bergstrom's Twitter lesson reminds us. Once we reach the threshold of "herd immunity," we will still continue to see new infections as part of the "overshoot," reflecting the additional number of individuals infected as the transmission begins to slow and the pandemic declines. Even so, this classic definition, based on an R0 of 2.5, yields a herd immunity level of 60%. However, in the August issue of Science, researchers suggest an age-adjusted threshold for herd immunity of 43%, while others indicate it could be as low as 15% and as high as 80%. The challenge remains that this is an overall threshold, and it does not offer the real-time insights of how far along the path to herd immunity our communities are right now.
RTI’s COVID-19 Data Insights tool introduces a measure of estimated acquired immunity, providing basic insights into the level of immunity resulting from reported COVID-19 infections to date. Estimated acquired immunity considers the proportion of the population with acquired immunity based on the total cumulative number of positive cases reported. As a real-time measure, this also assumes that the total number of positive cases reduces the population of susceptible individuals. This simplified approach offers a rough estimate for localities. A challenge of this approach is that it depends on testing being available to all who need it, when they need it. Nevertheless, in the absence of a perfect testing strategy, this can provide some insights.
For example, Cook County (IL) has a population of approximately 5.2 million based on the latest US Census Bureau’s American Community Survey population estimates, reporting a total of 175,586 confirmed COVID-19 cases to date (as of October 26). The COVID-19 Data Insights tool shows the level of estimated acquired immunity to be around 3.5%. Given the availability of zip code prevalence data across the Chicago metropolitan area, we can see the local variation based on the disproportionate rates of infection across the city, with estimates ranging from 0.4% up to 6.2%.
However, we have found that immunity estimates begin to break down when attempting to calculate for age-specific populations as the pandemic's true size and breadth remain woefully underestimated and poorly captured. Where de-identified case line data is available with age characteristics, comparison to local demographics raises concern, particularly among school-age children (ages 10 and under). For example, the total number of positive cases in one county is 712 among children ages 10 and under, including elementary schools, preschools, and daycares. Compared to the total population of 159,646 children in the county, this suggests estimated acquired immunity is less than 1% for this age group. In total, young children represent only 2% of the county's positive cases, raising questions and concern that we are not finding the instances when they occur.
Early seroprevalence study results from the CDC suggest that the range of underestimated cases varies from 6 to 24 times the numbers reported, varying by age, race, and gender. Even so, at 24 times the observed rate, estimate immunity among young children would only be 12%, far below any of the suggested thresholds of herd immunity. Data on positivity rates for age groups is extremely limited, further amplifying the uncertainty.
While we continue to advance our knowledge of the virus and where the threshold of herd immunity may be, these measures do serve as cautionary signals as we head into the holiday season. Increased testing to find people infected with the virus and rapid reporting of cases will improve our ability to estimate what percent of the population has immunity, at least in theory.
Fill out the form below to visit the tool and see estimated acquired immunity at the county level.
Contact our team if you would like to learn more about our methods or about how we can partner to help you gain valuable insights from your data at a hyper-local level.
The authors would like to thank Chloe Stephenson, Anne Marie Miller, and Helen Jang for contributing to this piece.
Sources and methods:
In order to assess estimated acquired immunity, we combine prevalence data on COVID-19 infections from the cumulative total positive cases provided by the Johns Hopkins University COVID-19 Data Repository and the City of Chicago on HealthData.gov, and population estimate data from RTI SynthPop™. We assume that the available number of total positive cases is a true reflection of population prevalence. Our estimation assumes that individuals cannot be re-infected (very limited evidence), individual who test positive are no longer susceptible (limited evidence), and population estimates do not account for births, deaths, or mobility.