There has been a significant amount of discussion recently about the Great Barrington Declaration1. Its authors promote an approach to COVID-19 that they call "Focused Protection". They describe this approach as follows: “…to allow those who are at minimal risk of death to live their lives normally to build up immunity to the virus through natural infection, while better protecting those who are at highest risk.”
This is a very appealing statement to those who are tired of restrictions and in a context where the economic and social impacts of the restrictions are being felt keenly by those under 60, (“retirement age” is the cut off proposed in the Barrington document) who are at lower risk of severe outcomes. Unfortunately, the claim that this approach is achievable with minimal impact is not correct for several reasons.
Evidence around long-lasting immunity is still unclear
First, the stated goal of this approach is to build up herd immunity through infection, which assumes that infection by SARS-CoV2, the virus that causes COVID-19, will automatically confer long-lasting protection against future infections.
This is not currently known to be the case. Other common coronaviruses that cause respiratory infections in humans have been shown to cause repeat infections2. With COVID-19 specifically, there have been rare individual documented cases of re-infection with SARS-CoV2.3, 4, 5, 6 At a population level, the Brazilian city of Manaus was widely cited as having reached herd immunity with approximately 66% of the population testing positive for antibodies7, 8, 9, yet there are recent reports of a resurgence of cases with up to 50 new deaths per day.10, 11
Therefore making the assumption that widespread infection will confer lasting immunity is not certain to be true.
However, if we assumed for the sake of argument that infection does confer immunity, there are still issues with the herd immunity plan. The second problem with the premise of the Great Barrington Declaration is the inaccurate assertion that if we segregate the old and the young, and let the young live 'normally', potentially getting infected along the way but not passing the virus to older people, herd immunity could be achieved with few costs in health related to COVID.
Returning to the city of Manaus in Brazil, it is important to know that although just 6% of its population is over the age of 609, the high antibody level in the city still came at a high price – a death toll estimated between 2,500 and 3,4008, 11, in a city of about 1.8 million. If we had the same overall per-capita death rate, to reach 66% antibody positivity would cost us between 6,100 and 8,300 deaths in Alberta. It is not clear what proportion of the deaths in Manaus were in those over age 60, but even if we assume that we could somehow completely protect those over 60 from infection, and that the risk of death from infection would just be in those living 'normally' (under age 60), there would still be a cost in deaths.
If we use our own Alberta data on the age-specific risk of death in those diagnosed with COVID12, and if we assumed that reaching a 50% infection rate was sufficient for herd immunity (though many estimates are that a higher percentage would be required), infecting 50% of those in the Alberta population under 60 would cost approximately 1,000 lives in that same younger population.
Assuming we were willing to pay that cost in lives for the benefit of 'normal' life in younger age groups, the other thing to remember is that death is not the only severe outcome. Hospitalization and ICU admissions are also severe outcomes that are more common than death in all age groups. Again, assuming we could somehow successfully segregate those over 60 from those under 60, and using our own Alberta data for age-specific risk of hospitalization in diagnosed cases, we would expect over 39,000 hospitalizations to achieve an infection rate of 50% in the population under the age of 60.
Using diagnosed case fatality and hospitalization rates could over-state the risks, as not all cases are diagnosed, and those cases that are more severe are more likely to be diagnosed.
However, all serology studies in Alberta have consistently shown antibody prevalence in our population at present to be less than 1%. Assuming a maximum 1% infection rate as of early August (our last serosurvey timeframe for when we have results) and calculating a non-age-adjusted ratio of diagnosed cases as of mid-July (2 weeks prior to the time of serology testing – 9673 cases) to serologically positive Albertans (1% of the Alberta population is 44,219), we could estimate that actual infections may be about 4.6 times higher than what was diagnosed.
If we reduce the estimated deaths and hospitalizations in the under 60 population by 4.6, we would still have about 240 deaths and 8,600 hospitalizations as a consequence of a 50% infection rate in Albertans under 60. If these infections were allowed to spread unchecked over a short period of time (the Barrington document does not state for how long those over “retirement age” should be restrained in their movement, but commentary on the document suggests 3 months), the hospitalization volume alone would be sufficient to impair the ability of our acute care system to manage all the other health care needs of our population.
In order to manage the demand for hospital beds and ICU care, other services would have to be paused or stopped in order to care for the acutely ill. This would worsen, not improve, the outcomes of concern in the Barrington document such as cardiac care, cancer screening and childhood immunizations.
Long-term health impacts
In addition, while hospitalizations, ICU admissions and deaths are the most obvious severe outcomes of COVID-19 illness, there is a growing body of evidence on the long term impacts that some people experience after an infection with SARS-CoV2. These include prolonged illness13, 14, sometimes called “Long COVID Syndrome”, which in some cases resembles Chronic Fatigue Syndrome, and emerging case reports of other possible long-term health impacts15, 16 that could irrevocably alter the course of people’s lives.
Limits to any "Focused Protection"
Finally, the premise that we could successfully shield continuing care facilities and hospitals from COVID-19, and that we would be able to support all those over 60 (and presumably those with high risk chronic conditions) to stay home with limited activities is not supported by evidence. In fact, those who work in continuing care facilities and hospitals can unintentionally be the source of infection in these locations.
We are working hard to ensure that every protection possible is put into place to prevent these introductions, but no measures will be perfect. In addition, we heard very clearly that the quality of life for those in continuing care was severely worsened when no visitors were allowed in, highlighting the tension between COVID protection and overall wellbeing in these high risk locations.
In addition, those over the age of 60 are often still working, contributing in many diverse fields, and the impact of having them all stay home would be significant. For example, more than 30% of Alberta physicians in 2018 were over the age of 55, and 10% were older than 6517, and removing them from the work force would be a poor choice in a time when health care is under significant pressure.
Finally, allowing the virus to spread rampantly in the age group under 60 would almost certainly result in impacts on critical services as those who are ill, even if the symptoms are mild, would need to be home for 10 days to prevent spread to those at high risk (e.g. in health care settings) and critical sector continuity would be put at risk.
Balancing COVID-19 restrictions with protecting our overall health
So, is there anything that can be taken from the Barrington document? First, the societal risks of public health measures that it outlines are real, and are exactly the reason that in Alberta we moved early on to targeting restrictions only where and when they are needed. The Barrington document implies that “lockdown” is binary – all or none, and that no restrictions should be in place for the young. This is a false dichotomy. The best way to prevent severe illness and death from COVID-19 is to prevent large spreading events, quickly identify cases, trace and isolate contacts, and keep the spread of the virus to a manageable level. This is exactly what we are doing.
Second, we already have policies that accept some risks of transmission in younger populations knowing that the benefits of activities outweigh the risks for those populations. Examples include opening schools and supporting youth sports. We can learn from what is working well in these areas and continue to judiciously expand activities in low risk populations as long as spread remains manageable.
We are not in lockdown in Alberta. We are using targeted measures to keep spread manageable and to ensure that our health system can cope with demands. We must continue to pursue this balanced approach, learning as we go along how best to minimize both the risks of public health measures and the risks of COVID-19. Herd immunity by natural infection is not a wise, or possibly even an achievable, goal to pursue.
- Great Barrington Declaration
- Lessons for COVID-19 Immunity from Other Coronavirus Infections (PDF, 1.7 MB)
- Genomic evidence for reinfection with SARS-CoV-2: a case study
- Symptomatic SARS-CoV-2 reinfection by a phylogenetically distinct strain
- COVID-19 Re-Infection by a Phylogenetically Distinct SARS-CoV-2 Variant, First Confirmed Event in South America
- Coronavirus Disease 2019 (COVID-19) Re-infection by a Phylogenetically Distinct Severe Acute Respiratory Syndrome Coronavirus 2 Strain Confirmed by Whole Genome Sequencing
- COVID-19 herd immunity in the Brazilian Amazon
- Brazil city ‘might have reached herd immunity'
- A city in Brazil where covid-19 ran amok may be a ‘sentinel’ for the rest of the world
- In Brazil's Amazon a COVID-19 resurgence dashes herd immunity hopes
- Hotspots of resurgent Covid erode faith in ‘herd immunity’
- COVID-19 Alberta statistics
- Symptom Duration and Risk Factors for Delayed Return to Usual Health Among Outpatients with COVID-19 in a Multistate Health Care Systems Network
- Long covid: How to define it and how to manage it
- A case of probable Parkinson's disease after SARS-CoV-2 infection
- New-Onset Diabetes in Covid-19
- AHS Physician Workforce Plan and Forecast: 2018-2028 (PDF, 1.4 MB)