How policy may influence population susceptibility to SARS-CoV-2 infection: an epidemiological perspective
The susceptible population, or those at risk of SARS-CoV-2 infection and subsequently person-to-person transmission, is a function of time. As the pandemic unfolds, this population evolves dynamically. Community immunity is dependent on susceptibility and its inverse, non-susceptibility, so it too changes over time (the herd threshold, on the other hand, is a calculated based on the infectivity of virus and is less likely to change over time). The quest of public health is to drop the reproductive rate below one, so the pathogen can no longer secondarily infect additional hosts. This is achieved through various policies.
There are several ways to alter the balance of susceptible to non-susceptible individuals that would result from various policy interventions. First, we can sequester the susceptible individuals from the population - self quarantine - and/or sequester the infected individuals from the population - self-isolation. This is quite effective in fact, as long as people adhere to it. A major problem is identifying those infected versus those not. This would require a two-stage testing approach (serological testing for past or present infection followed by nucleic acid testing for current infection) on nearly everyone in the population in order to sort them into these two bins. Of course, with limited testing resources this is not feasible. Second, we could potentially immunize individuals or have some other kind of prophylaxis. This has a very strong assumption that a prophylaxis is possible and effective. Third, we could allow individuals to acquire immunity through natural infection, as is the case when someone gets sick and recovers with immunity. The big issue here is whether an antibody-mediated SARS-CoV-2 immune response exists (it does), is protective (it likely is), and how if so, how long will it persist (unknown). To answer questions about what the most effective policy strategy would be, we would have to assign some measure of cost associated with each infection and the policy itself. This could be a financial cost, healthcare utilization cost, quality of life cost, and so on.
From a public health and health care perspective, the cost of quarantine is assumed to be less than cost of natural immunity. This is due to resource limitations and potential for infectious disease morbidity and mortality. This does not consider non-infectious outcomes, such as economic impact, lifestyle changes (changes in physical activity or dietary habits), mental health, nor chronic disease sequalae, which are in fact much more difficult to quantify then the infectious outcomes. When a general population-level quarantine (i.e. shelter in place/stay at home order) ends susceptible individuals have the potential to become exposed and infected, and induce secondary outbreaks. To some extent this can be minimized by a gradual phase-out of these orders and intensive public health surveillance thereafter. Importantly, the number of susceptible people will be smaller in this post-quarantine time because some number of people initially susceptible at the start of the outbreak have since become infected and recovered or died (assuming the birth/migration rates stay constant). This policy position is predicated on the assumption we will be better prepared to deal with infection at a later point in time than we are today, and the burden of disease will be lower. From a medical standpoint and a public health standpoint this allows time to replenish and restock the healthcare system and personal protective supplies, further research on the pathogen and contagion, and development of prophylaxis and treatments. The potential harm impacts the economy as well as individual. Aside from the obvious economic impact, people's health will be impacted by mental health-related exacerbations and lack of resources to seek health care, which may manifest for years to come. By abstaining from a generalized quarantine, such as Sweden and some other places have experimented with, more people may develop natural immunity quicker, but the infectious disease morbidity and mortality will be greater since the pool of susceptible people is larger at the outset. In their minds the cost of quarantine exceeds cost of natural immunity. Thus the pandemic could exact a more severe toll on the population but over a shorter period of time, kind of similar to what we are seeing happen to front line works right now in the U.S.
Up to this point susceptibility has been described as a dichotomous yes/no, and in fact is frequently modeled as such. This doesn't match reality: there is a spectrum of susceptibility. Some individuals who have not been infected and are considered susceptible and then exposed may never become infected. Some individuals who are infected may not be infectious to others. The are a bevy of reasons for this, and may include age, immune function or compromise, initial exposure level, the propensity of the virus to replicate in the host and viral shedding (and replication in the new host), social networks, and so on. A lot of this is not well understood at the moment. The important point is that models that estimate herd immunity based on susceptibility often make an assumption that susceptible individuals are infectable and that once infected can transmit and spread infection.
In short, when considering how policy impacts the pool of susceptible individuals in the population, we must also consider 1) availability of a prophylaxis such as a vaccine or effective treatment, 2) whether the infectivity of the virus remains constant, 3) other external forces that influence the susceptible population, 4) the stability of the target population, and 5) whether susceptible=infectable=transmissible. I am reminded of the way the healthcare system in the U.S. frequently operates; we have no qualms about spending inordinate amount of money at the point of care for an individual, yet shy away from spending on upstream social determinants of health.
DISCLAIMER: This post should not be misconstrued as advocating for any particular policy position. I present this in a neutral light, supported only by the fundamentals of infectious disease epidemiology. Any reference to this post must be include this disclaimer.
Cite: Goldstein ND. How policy may influence population susceptibility to SARS-CoV-2 infection: an epidemiological perspective. Apr 15, 2020. DOI: 10.17918/goldsteinepi.
