closed sign
Dr. Ari Joffe is a specialist in pediatric infectious diseases at the Stollery Children's Hospital in Edmonton and a Clinical Professor in the Department of Pediatrics at University of Alberta. He has written a paper titled COVID-19: Rethinking the Lockdown Groupthink that finds the harms of lockdowns are 10 times greater than their benefits.
The below Q&A is an exchange between Joffe and Anthony Furey.

You were a strong proponent of lockdowns initially but have since changed your mind. Why is that?

There are a few reasons why I supported lockdowns at first.

First, initial data falsely suggested that the infection fatality rate was up to 2-3%, that over 80% of the population would be infected, and modelling suggested repeated lockdowns would be necessary. But emerging data showed that the median infection fatality rate is 0.23%, that the median infection fatality rate in people under 70 years old is 0.05%, and that the high-risk group is older people especially those with severe co-morbidities. In addition, it is likely that in most situations only 20-40% of the population would be infected before ongoing transmission is limited (i.e., herd-immunity).

Second, I am an infectious diseases and critical care physician, and am not trained to make public policy decisions. I was only considering the direct effects of COVID-19 and my knowledge of how to prevent these direct effects. I was not considering the immense effects of the response to COVID-19 (that is, lockdowns) on public health and wellbeing.

Emerging data has shown a staggering amount of so-called 'collateral damage' due to the lockdowns. This can be predicted to adversely affect many millions of people globally with food insecurity [82-132 million more people], severe poverty [70 million more people], maternal and under age-5 mortality from interrupted healthcare [1.7 million more people], infectious diseases deaths from interrupted services [millions of people with Tuberculosis, Malaria, and HIV], school closures for children [affecting children's future earning potential and lifespan], interrupted vaccination campaigns for millions of children, and intimate partner violence for millions of women. In high-income countries adverse effects also occur from delayed and interrupted healthcare, unemployment, loneliness, deteriorating mental health, increased opioid crisis deaths, and more.

Third, a formal cost-benefit analysis of different responses to the pandemic was not done by government or public health experts. Initially, I simply assumed that lockdowns to suppress the pandemic were the best approach. But policy decisions on public health should require a cost-benefit analysis. Since lockdowns are a public health intervention, aiming to improve the population wellbeing, we must consider both benefits of lockdowns, and costs of lockdowns on the population wellbeing. Once I became more informed, I realized that lockdowns cause far more harm than they prevent.

There has never been a full cost-benefit analysis of lockdowns done in Canada. What did you find when you did yours?

First, some background into the cost-benefit analysis. I discovered information I was not aware of before. First, framing decisions as between saving lives versus saving the economy is a false dichotomy. There is a strong long-run relationship between economic recession and public health. This makes sense, as government spending on things like healthcare, education, roads, sanitation, housing, nutrition, vaccines, safety, social security nets, clean energy, and other services determines the population well-being and life-expectancy. If the government is forced to spend less on these social determinants of health, there will be 'statistical lives' lost, that is, people will die in the years to come. Second, I had underestimated the effects of loneliness and unemployment on public health. It turns out that loneliness and unemployment are known to be among the strongest risk factors for early mortality, reduced lifespan, and chronic diseases. Third, in making policy decisions there are trade-offs to consider, costs and benefits, and we have to choose between options that each have tragic outcomes in order to advocate for the least people to die as possible.

In the cost-benefit analysis I consider the benefits of lockdowns in preventing deaths from COVID-19, and the costs of lockdowns in terms of the effects of the recession, loneliness, and unemployment on population wellbeing and mortality. I did not consider all of the other so-called 'collateral damage' of lockdowns mentioned above. It turned out that the costs of lockdowns are at least 10 times higher than the benefits. That is, lockdowns cause far more harm to population wellbeing than COVID-19 can. It is important to note that I support a focused protection approach, where we aim to protect those truly at high-risk of COVID-19 mortality, including older people, especially those with severe co-morbidities and those in nursing homes and hospitals.

You studied the role modelling played in shaping public opinion. Can you break that down for us?

I think that the initial modelling and forecasting were inaccurate. This led to a contagion of fear and policies across the world. Popular media focused on absolute numbers of COVID-19 cases and deaths independent of context. There has been a sheer one-sided focus on preventing infection numbers. The economist Paul Frijters wrote that it was "all about seeming to reduce risks of infection and deaths from this one particular disease, to the exclusion of all other health risks or other life concerns." Fear and anxiety spread, and we elevated COVID-19 above everything else that could possibly matter. Our cognitive biases prevented us from making optimal policy: we ignored hidden 'statistical deaths' reported at the population level, we preferred immediate benefits to even larger benefits in the future, we disregarded evidence that disproved our favorite theory, and escalated our commitment in the set course of action.

I found out that in Canada in 2018 there were over 23,000 deaths per month and over 775 deaths per day. In the world in 2019 there were over 58 million deaths and about 160,000 deaths per day. This means that on November 21 this year, COVID-19 accounted for 5.23% of deaths in Canada (2.42% in Alberta), and 3.06% of global deaths. Each day in non-pandemic years over 21,000 people die from tobacco use, 3,600 from pneumonia and diarrhea in children under 5-years-old, and 4,110 from Tuberculosis. We need to consider the tragic COVID-19 numbers in context.


Comment: Considering the 'absence' of flu this year, the faulty PCR tests that deliver false positives, the hysteria and incentives to label every possible death as caused by coronavirus, we can assume the coronavirus mortality statistics are significantly warped.


I believe that we need to take an "effortful pause" and reconsider the information available to us. We need to calibrate our response to the true risk, make rational cost-benefit analyses of the trade-offs, and end the lockdown groupthink.

Canada has already been going down the lockdown path for many months. What should be done now? How do we change course?

As above, I believe that we need to take an "effortful pause" and reconsider the information available to us. We need to calibrate our response to the true risk, make rational cost-benefit analyses of the trade-offs, and end the lockdown groupthink. Some considerations I have suggested elsewhere include the following:
  • We need to better educate ourselves on the risks and trade-offs involved, and alleviate unreasonable fear with accurate information. We need to focus on cost-benefit analysis - repeated or prolonged lockdowns cannot be based on COVID-19 numbers alone.
  • We should focus on protecting people at high risk: people hospitalized or in nursing homes (e.g., universal masking in hospitals reduced transmission markedly), in crowded conditions (e.g., homeless shelters, prisons, large gatherings), and 70 years and older (especially with severe comorbidities) - don't lock down everyone, regardless of their individual risk.
  • We need to keep schools open because children have very low morbidity and mortality from COVID-19, and (especially those 10 years and younger) are less likely to be infected by, and have a low likelihood to be the source of transmission of, SARS-CoV-2.
  • We should increase healthcare surge capacity if forecasting, accurately calibrated repeatedly to real-time data (up to now, forecasting, even short-term, has repeatedly failed), suggests it is needed. With universal masking in hospitals, asymptomatic health care workers should be allowed to continue to work, even if infected, thus preserving the healthcare workforce.