A new study, which was published in October 2020 on the preprint server medRxiv *, shows that recognizing the differences in the population composition, the connectedness and distribution as well as the inter-individual differences in terms of immunity, susceptibility and infectivity for the estimation of the achieved Critical herd immunity is a result of the natural severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) more precisely. This knowledge is critical to shaping sensible containment policy for the future.
Classic epidemiological models
In the traditional infectious disease model, it is assumed that contact between people takes place homogeneously between age groups and different parts of the country. Individual differences in terms of infectivity and susceptibility to the virus are also not taken into account.
Newer models therefore attempt to take into account heterogeneous mixing, recognizing that social interactions for most people take place within a relatively small circle, which is limited by similarities in age and location. The number of people moving across the boundaries of these groups is small.
Threshold for artificially high herd immunity?
The traditional calculation for detects the threshold above which the infected percentage of the population (the fraction that has been removed from the susceptible fraction) is large enough to prevent enough new infections from occurring to allow the epidemic to spread. Thus, the herd immunity threshold (HIT) is a measure of the level of immunity in the population that will prevent further outbreaks from occurring. The classic herd immunity threshold is expressed as 1 – 1 / R0 (R0 is the reproductive number).
The 1918 pandemic flu is believed to have a R0 ~ 2.9, which corresponds to a herd immunity threshold (HIT) of about 66%. Again, most scientists believe that herd immunity to COVID-19 requires 60% or more of the population to be infected for an R0 of 2.5. In reality, however, the 1918 HIT was likely closer to 33% as it is believed that only a third of the world’s population is actually infected.
The 2009 H1N1 influenza pandemic was estimated to have an R0 between 1.2 and 1.6, which corresponds to a HIT of 28.5% according to the classic model. This is in contrast to the CDC’s estimates that only a fifth of the US population was infected before the pandemic died out.
These overestimations are based on the assumption implied in traditional models that the susceptible population and mixing patterns are homogeneous.
The network on which the model is based.
Understand pre-existing immunity
The current model examines how differences in the spatial distribution and in the immune status of susceptible individuals affect herd immunity. This has been brought to the fore in the actual HIT debate by previous researchers.
Many uninfected people have T cells that cross-react with SARS-CoV-2 due to previous annual coronavirus exposures. In fact, 9 out of 10 people will have antibodies to coronavirus at some point in their life, although they can last anywhere from months to 34 years.
Currently, some researchers estimate a 35% helper T cell prevalence due to previous exposure to these common coronaviruses. It is therefore necessary to study the role of these cells in relation to clinical severity and outcome.
The presence of innate immunity to the disease is another factor to evaluate. Such individual variation contributes to the final estimate of the HIT.
An earlier two-compartment model, in which one group is immune to the virus due to innate or cross-reactive immunity, shows the size of that group in relation to the other along with the R0 value and the degree to which both groups are mixing leads to significant deviations in the HIT. The authors suggest that some sites may have already achieved herd immunity. A modified form of this model is also used in the current study to illustrate the role of partial immunity.
Space models to include heterogeneity
In one of the earlier models, the researchers presented six age-stratified groups in a community with heterogeneous contact between the groups. This showed that the HIT could be reduced significantly to, for example, 43% with heterogeneous contact.
The current study first visualizes the use of unified non-pharmaceutical interventions (NPIs) like home protection and other measures throughout the course of the epidemic. They are trying to find a value for the minimum level of NPIs necessary to ensure that the first wave ends up achieving herd immunity, which will prevent a second wave.
The infected proportion of the population is then calculated as the disease-related HIT. According to this model, the HIT is ~ 47.5% compared to 77% for the classic model. This is reflected in the results of large-scale serological tests in New York City, which were hard hit by the pandemic. The antibody tests showed a prevalence of 27% that ranged from ~ 13% to ~ 52% in various locations.
A second estimate takes into account the possibility of cross-reactive T cells in about 35% of the population, accelerating recovery by one day and reducing mortality by 10% over the non-immune population. This reduces the disease-related HIT estimate to ~ 34%.
The researchers point out the importance of such calculations: “Even a modest heterogeneity in terms of infectivity and duration of disease progression affects the extent of disease-related herd immunity.”
Targeted inoculation of super-spreaders
The simplified homogeneous mixture model, which also gives the same infectivity for all persons in a population, has led to many public health strategies that address all persons in a population equally. On the contrary, the authors say, better models are needed that take into account differences in a number of areas, including susceptibility, commingling, and infectivity.
According to the authors, the key to preventing a second wave is to immunize highly connected or vulnerable or infectious people in any location, as they are central to the further transmission of disease.
Many seroprevalence studies continue to predict that the achievement of herd immunity is not yet in sight due to the low prevalence and the very different range depending on the location. The current model suggests that this is actually what is expected of a population. Of course, the antibody test itself is limited by a lack of sensitivity in the early convalescence phase, variable test quality, a rapid decrease in the antibody titer and increased antibody titer with the severity of the disease.
In Manaus, Brazil, where the pandemic has taken a heavy toll, both hospital stays and deaths from COVID-19 have fallen sharply, although seroprevalence has never risen above 20%. This could mean that low levels of antibody in some areas are compatible with herd immunity.
Other researchers support the view that heterogeneity in contact networks and in degree of transmission enables a much lower HIT than previously assumed. Shielding strongly connected nodes after loosening such restrictions can lead to an increase in infections, with the second wave being more significant than the first.
The investigators summarized: “We show that the proportion of the population infected to achieve herd immunity may be lower than normally assumed, which would have a significant impact on public health.”
Following the study, the level of herd immunity needs to be quantified and this knowledge translated into public health strategies targeting the most central foci of transmission in each region.
* Important NOTE
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be considered conclusive, guide clinical practice / health-related behavior, or treated as established information.