Ruminations on Population Immunity

A lot of people are hearing and reading about concepts they probably never heard of before.  One of them is population immunity.  If we back up for a minute and look at the big picture, the question is “How does this, or any, epidemic ever end?”  If there was not such a thing as adaptive immunity, the answer would obviously be  that it doesn’t.  Even with adaptive immunity, boosted by vaccines, the underlying pathogens for many diseases continue to circulate and cause disease and death.  Influenza is a good example.  So the goal must be to understand what slows the transmission of the infectious agent among humans to the point that we consider the burden of disease and death acceptable.  In many places, coronavirus has already reached that status, if you judge it by our approach to influenza.

Because of the adaptive immune system, once a person has been infected, they have fairly extensive protection against a significant reinfection.  And when infectious agents are part of a family, like this virus is, then being infected by any member of the family may provide an immune response against other members of the family. Not becoming significantly infected means you likely are not infectious.  It is pretty obvious that the more people in any population who are resistant to infection, the less likely ongoing transmission of the virus will be.  Epidemiologists typically assume this has to be a very high percent of the population, 70% or 80%, and many early models for this epidemic assumed that.  But one encouraging feature of the widespread attention to this epidemic is that it has drawn in statisticians and data analysts from many fields, who have been more creative in their thinking and this has pressured epidemiologists to also be less rigid.  Some of the resulting modeling is much more realistic, particularly in regard to the role of individual variation in likelihood of becoming infected or infectious.

People in any population are obviously highly variable across many dimensions.  Some variables are just basic demographics–age, sex, ethnicity.  Some are socio-economic, like income level, residence type, population density in the area of residence.  Of particular interest in this regard are the number and types of contacts with other individuals that a person has.  Some variables are health related, such as levels and types of pre-existing illness, genetic makeup and immune system strength.  If you think about transmission in any population, you have to consider all these variations in determining how quickly and how extensively an infectious agent might spread.  The nature of the agent should also be considered; some are more easily passed and more easily get a toehold in the human body.  The relationship of many variables to the course of an epidemic with a new infectious agent will not be known early on, so modeling looks like guesswork.  As an epidemic proceeds, individually determined or government mandated behavior modifications can also affect spread.

Here is an excellent blog post, thanks to Nic Lewis for passing this on to me, on population immunity as it relates to coronavirus.  (PI Post)   The author summarizes the historical development of the concept, and discusses the body of research incorporating the notion of individual variability into models of an epidemic’s course, and how those models find that this variability in susceptibility to infection and infectiousness will likely result in a significant slowing of transmission when a smaller fraction of the population is infected than traditional modeling assumes.  The author, drawing on recent research, largely attributes this to adaptive immune system characteristics.  I generally subscribe to this view, with the caveat that some antibody surveys have actually shown prevalence in specific geographic areas in excess of 40% or even 50%, but that may be consistent with my evolving understanding of what it means to be “infected” or “infectious”.

Due to fear-mongering by politicians and the media, I am concerned that the general population will think we have to somehow completely suppress this virus.  That is not going to happen and we need to accept that.  If we don’t accept this truism, how can we come to accept, as we do with influenza and many other diseases, that there is some tolerable level of disease and deaths.  Even in the best case scenario, with a very effective vaccine, there are always going to be cases, and it is likely that this virus, like flu, will be able to mutate enough to evade vaccine protection in some people.

Given that our preferred strategy for dealing with this epidemic has been mass hysteria resulting in social, educational and business shutdowns, it is critical to understand when transmission will have slowed enough to end those shutdowns and to have accurate surveys that help us understand the extent of adaptive, and innate, immune response.  If around 20% to 30% of a population having had an active infection leads to dramatic slowing of transmission, then many places are likely at or approaching that threshhold, and could feel more secure in easing restrictions.  We are very fortunate also that this virus only is really threatening to the frail elderly and a few other people with substantial pre-existing illness burden.  That makes it easier to develop policies that attempt to protect the vulnerable, while allowing normal life to go on for the general population.

And a big part of normal life is that we live in a world with infectious agents and people sometimes get sick and die from them.