Putting the Herd (Immunity Debate) to Rest

Understanding infectious systems (and debating them) requires a person to be able to reason about highly complex biological structures, of which, contain hundreds of demographic and epidemiological variables.[*]

It seems commonplace in debates/discussions concerning herd immunity that extrapolating demands on others are made – particularly to maintain specific elimination thresholds (one example easily comes to mind: non-medical vaccine exemptions should be void to maintain herd immunity). 


Elimination Thresholds and the Dynamics of Immunity

Two authentic hazards arise when debates concerning herd immunity/vaccination rates place consistent emphasis on maintaining elimination thresholds (ie 90-95% vaccination rates)

(1) the debate begins to distract from the fundamental extreme dynamics of epidemic theory, particularly to herd immunity[* pg297]  

(2) the debate portrays a human life as a instrument in medicine that can/should be utilized in a societal defense against viruses/bacteria/disease. (the dismissal pertaining to an individual’s choice regarding a medical procedure)

For this post, I will only be addressing the first point (the complexities relating to the influences on herd immunity). To learn more about the second issue you can read this post.

I feel the need to address this particular point because in common debates regarding herd immunity (pertaining to national vaccination programs) a fundamental cornerstone of epidemiology is continually dismissed: the extreme dynamic consequences of the intrinsic nonlinearity of host-agent systems.[*] 

In laymens terms – this shit is complex.


Does Herd Immunity Exist?

Absolutely.

There is no denying that herd immunity is an extremely potent natural phenomenon which is altered by a multitude of influences (which I am going to attempt to list briefly a little later). In the presence of a national vaccine program, it is purposefully manipulated with the intention that the recognized gains will out weigh any known disadvantage(s) that might occur (financially, safety, policy, etc). [*]

In the United States, we rely on the ACIP to examine new (and current) vaccination campaigns of which must be weighed very cautiously. To intentionally modify the current existing herd immunity against a non-fatal, acute childhood disease can carry unfortunate characteristics such as protecting one population while actually placing another group at higher risk than before (ex varicella campaign).[*][*][*][*]


On the other hand, there are campaigns that historically have provided success. So far, the only infection to be eradicated worldwide is smallpox (variola major -WHO 1977). This accomplishment generated much optimism in the philosophy of eradication thresholds that other infectious diseases (such as measles and pertussis) were targeted.  Unfortunately, this thinking may have been misguided for many reasons: smallpox was unique given its low communicability, the high average age of infection, the ease of diagnosis and the stability of vaccine storage conditions.[*]


It is also important to reveal, that there are complete absences of herd immunity seen in several diseases (some of which we currently vaccinate against such as rubella, diphetheria, pertussis).[*]

The blanket justification of herd immunity in support of elimination thresholds for vaccination on the current US vaccine schedule has the very real potential in becoming a crude and one-dimensional debate.


The Fundamentals

When speaking of herd immunity in relationship to a national vaccination campaign, one must be specific to which disease they are discussing and they must show an understanding to the intricate nature of influences of immunity within the individual and within a community.

To begin, a foundation must be understood. Epidemic theory considers three variables: agent, host and environment (each of which has many components/interactions/influences in-and-of themselves).

Agent
An agent is any infectious pathogen. These vary in biological makeup, size, transmission, and habitat. In the construction of mathematical models of epidemics and herd immunity, all possible variations in all aspects of the agent's behavior (particularly in relation the host and the environment) must be taken into account.[*]

Host
The classification of a host is relevant when an agent invades a foreign entity (aka the host) resulting in a defensive immune response with the purpose of protection. Immunity attained can range from temporary to permanent. [*] 


What is of particular study is the host’s response with antibodies specific to the infectious antigen. These seropositive individuals are those who have current infections or who have experienced an infection in the past – moving them from the category of ‘infected’ to ‘recovered’.[*]    

Of course, when referring to the context of herd immunity, one must consider both the individual hosts and the population as a whole.

Environment
Consideration and concern is given to the environment and vicinity in which both the host and agent dwell. This can range from geographical heterogeneity to seasonal variations (again, for both host and agent).



Mathematical Modeling – SIR Model

Epidemiology gives birth to herd immunity theory when the first mathematical model examined how infectious agents affected large populations over time.[*][*][*]

Obviously, for ethical reasons and financial reasons (hopefully the former out weighs the later), experimentation or field trials are prohibitive – making mathematical modeling critical in making theoretical predictions of how a disease will spread and can be useful for evaluating control strategies (particularly in the case of a bio-weapon attacks).[*][*]

The Mass Action Principle (SIR) has been widely applied and accepted in epidemic theory since 1927 – when Kermack and McKendrick published 3 papers outlining and describing a mathematical model in which they considered a fixed population with 3 compartments: susceptible; infected; recovered (SIR).[*][*][*]

Susceptibles (S) – Individuals that are susceptible have, in the case of the basic SIR model, never been infected, and they are able to catch the disease. Once they have it, they move into the Infected compartment.[*]

Infected (I) – Infected individuals can spread the disease to susceptible individuals. The time they spend in the infected compartment is the infectious period, after which they enter the recovered compartment. [*]

Recovered (R) – immune to the disease or otherwise removed from the population. Individuals in the recovered compartment are assumed to be immune for life. [*]

In the SIR model, vaccination is equivalent to complete removal (aka transfer to the Recovered compartment). It is assumed that vaccinated individuals can not infect or be infected. [*] 

The above described SIR model is helpful although it is written using an equation that implies a deterministic model (no randomness with a continuous time). [*]

To account for this, the SIR model is the basis for other similar models (SEIS, MSIR, MSEIRS) that make small adjustments in its parameters that attempt to alleviate certain ‘real-world’ problems, for example:[*][*] 

SEIS  - considers the exposed or latent period of the disease (a person is not immediately infected).

MSEIRS – considers an infection that does not leave a lasting immunity in which individuals that have recovered will return to being susceptible again, moving back into the S compartment.

MSIR – considers a disease where an individual is born with a passive immunity from the mother.

Although mathematical equations are very useful in understanding basic principles and the interplay between variables, their assumptions can lead to oversimplification.[*][*]

My concern centers on the simplicity of many mathematical models, particularly in the face of such biological complexity. Especially useful tools in modern complex theories incorporate multiple algorithms and concepts of TCS (theoretical computer science), however these tools are not yet utilized and relatively unknown in epidemiology.[*][*]



Complexities of the Herd

There are several assumptions made in the formulation of the above mentioned equations. These assumptions have benefits and disadvantages.

Benefits may include being utilized for a general guide to risk assessment or a supportive piece to compare alternative policies/intervention – a wide-ranging compass to help make epidemiological decisions.

However, it is clear that to make the forecast more realistic, it is necessary to introduce more details in the disease dynamics. Models that incorporate even the most elaborate derivations omit important features.[* pg 296]

Here are a handful of features to begin to take into consideration that influence the dynamics of disease, immunity, transmission, and recovery.[*]


Geographical heterogeneity
This refers to gender (i.e. male-female ratios), age, and factors correlated with residence. Because of existing heterogeneity, estimated elimination thresholds vary between local communities – significant local differences in population dynamics arise which, consequently adjust estimates.[*][*]

Areas of similar dynamics and variation pertaining to heterogeneity:

• Age-Structured populations

• Variable infectivity

• High contact probabilities

• Persistence of pathogens within hosts

• Variations in infection risk by age group

• Limitation of application to a closed population (no immigration or emigration)[*]

• Demographic turnover (birth or death).[*]

Host genetic factors
A particularly new dynamic gaining more comprehension is the role of host genetic factors – this dynamic is critical because mathematical models rely on and individual within the population as having an equal probability as every other individual of contracting and transmitting a disease.[*][*][*]
Research is continuing to learn strategies to identify host genes responsible for resistance/susceptibility to particular agents and the relationship between vaccine efficacy and genetics.

Areas of similar dynamics and variation pertaining to host genetic/immuno factors:

• intra-host dynamics

• Maternal immunity

Spatial epidemiology
The most fundamental of these is the problem of defining the spatial location of the entities being studied. For example,

In regards to the study on human health, spatial position of humans can pertain to the area or point with where an individual/group live, or with a point located where they work, or by using a line to describe their weekly trips. Each variation has dramatic effects analysis and on the conclusions obtained.[*]

Other issue arise in the application of spatial analysis which includes the limitation of mathematical knowledge problems in computer based calculations.[*] 


Antigenic shift
Antigenic shift is contrasted with antigenic drift (a natural mutation over time). The issue with antigenic shift concerns our lack ability to forecast ability of viruses to alter their genetic makeup (quickly creating mutant antigens) and bypassing the antibody barrier a host/community.[*]


Phylodynamics
Viral phylodynamics examines how epidemiological, immunological, and evolutionary processes impact viral genetic variation.

Dynamics of transmission is considered at the level of cells within an infected host, individual hosts within a population, or entire populations of hosts.[*]

Currently, it is understood that viruses within similar hosts, such as hosts that reside in the same geographic region, are expected to be more closely related genetically if transmission occurs more commonly between them.[*]

Areas of similar dynamics and variation pertaining to phylodynamics:

• pathogen population genetics

• evolution and spread of resistance to immunity/medication

• Strain (biology) structure and interactions

Seasonal variations
There are several biologically distinct mechanisms in which seasonality and climate change impacts host-pathogen interactions. Strong pressures on population dynamics are exerted by temperature, rainfall, seasonality and climate change – responses can range from simple annual cycles to more complex multiyear fluctuations.[*][*]

Although scientists are only beginning to understand how seasonal external drivers influence the majority of host–parasite systems, empirical evidence strongly supports the strength and mechanisms of which seasonality alters the spread and persistence of infectious diseases. [*][*]

To present two examples for further understanding:

Agent: rotavirus.
Seasonal affects: winter peaks; timing shifts with latitude[*]

Agent: meningococcal meningitis
Seasonal affect: wind speed and low humidity affect respiratory/aerosol transmission[*]

Areas of similar dynamics and variation pertaining to seasonal variation:

• alterations in immune system defenses (weakened during winter and during harsh weather)

• periparturient rise (pregnant women lowering their own immunity to prevent harming the fetus)

• diseases that are cyclical in nature

• diseases that are seasonal in nature

Ending the Debate on Herd Immunity 

Agent-Host-Environment: This relationship is complex and depends on such factors as a precise course of infection (not only with an individual but within the demography of the host population). Other factors include duration of immunity (natural or artificially acquired), maternally derived protection, age-related changes in the intimacy of contacts – not to mention a prevailing level of genetic and spatial heterogeneity in both susceptibility and resistance to infection.[*]

There is no denying that mathematical models aid in defining details associated with infectious systems. Adjustments made to incorporate dynamic influences have the ability to make useful generalities and estimates – particularly to elimination thresholds and the course of infection within a population.[*] 

However, the science and intuition built on decades of practical epidemiological experience still often fail to predict outcomes/implications of vaccination programs.[*]

Each vaccination campaign entails a massive disruption of the previous balance which results in a destabilization of epidemiologic patterns for many years.[* p297]

...are vaccinated individuals the ones stepping outside the herd?

Herd immunity existed prior to vaccination.

Preceding national vaccination programs, epidemiological patterns of immunity/disease existed for the greater good. With each national vaccine campaign, epidemiology is modified (benefiting some and disadvantaging others).

While it is perceived beneficial that some vaccinated diseases have diminished over time (such as measles, chicken pox, rubella), it is very likely that these infections existed in a precise biological niche that was very much intentional – benefits in which our understanding can not yet comprehend.[*]

How exactly can the herd immunity debate end? 

(1) Accepting that the encompassing complexity of infectious disease (and nature in general) is not something human understanding will be able comprehend in entirety. Yes, artifical modification (via vaccine) can impart benefits – but it also exposes populations/individuals to disadvantages (many of which cannot be forecasted in advance) – implementation of vaccine programs must not be taken lightly nor forced upon a population without individual consent. 

(2) Above all else, voluntary consent married with adequate comprehension of each vaccine is fundamental as national vaccine campaigns develop and progress. This is whatshould be the underlying principle in discussions on herd immunity – NOT mandatory vaccination.

This post is dedicated to those families that have been deliberately and maliciously accused of ‘free riding’ off of vaccine-induced/artificial herd immunity – especially those who have been told this by a medical professional in hopes of altering a consent of vaccination.

I like to believe all parents are doing the best they can. Please choose what works best for your family while honoring the rights of others to make that same choice.