The COVID-19 modeling figures are scary. Have we mortgaged our future on an inaccurate science?


In the slide on page 13, Dr. Peter Donnelly warned those who were virtually connected to Ontario’s COVID-19 disaster technical briefing at the end of last week, that would be the most disturbing of all. Indeed, the numbers were staggering: without physical measures to distance themselves, Ontario would experience 100,000 deaths during the pandemic.

This was not the only shocking figure: the impact of COVID-19 would have been estimated at 300,000 cases and 6,000 deaths at the end of this month without intervention. According to the models, approximately 220,000 cases and 4,400 deaths were saved by closing schools, banning large gatherings, closing non-essential workplaces, closing outdoor recreation facilities and now paying people $ 880 to walk their dogs in closed parks. Donnelly, President and CEO of Public Health Ontario, said Friday that another 1,350 deaths (1,600 to 250) could be prevented in the next two weeks.

The range of scenarios seems dizzying, the forecasts seriously extreme. Already, critics insist on the high bias of the models, which leads many to wonder to what extent we must trust the models on which so many upsetting decisions are made.

It is not possible to know exactly where we will end up … it is difficult to know exactly where you are … projections and modeling of a whole new viral disease are very inaccurate … it is not an exact science.

Donnelly used the exact word, or variations of it, six times throughout the briefing. Yet imperfect science sheds light on Ontario’s strategy. And it highlights how the assumptions used to model the pandemic can rest on “very fragile foundations,” as Robert Dingwall, sociology professor at Nottingham Trent University in the UK said this week in response to a study questioning the benefits. school closings in terms of scientific data. proof.

Ontario models have projected mortality based on global experience to date with COVID-19, as well as data collected in Canada. Tables released Friday set the case fatality rate (the percentage of confirmed infections that end in death) at 2.1% overall for Ontario, 0% for those under 40, at a truly frightening rate of 15.9% for people 80 years of age. and more. However, the case fatality rate is based on known infections, and two-way biases can swell or underestimate it – most notably, mild cases can produce a false high.

A study published last week in Lancet Infectious Diseases, based on laboratory confirmed and clinically diagnosed cases in mainland China as well as international cases, the best estimate of the case fatality rate was 1.38% overall. The estimate of the overall infection death rate for China (the percentage of people infected and dying, including those with no symptoms) was 0.66%.

The virus is extremely effective at spreading between us, and it is very likely that almost everyone will understand it.

Donnelly said the Ontario projections were based on “other things”, although he did not specify. Officials have not released the actual models, just the projections. It is not known exactly which variables were connected. They were also vague when a reporter from the Globe and Mail about the limits of death projections and their confidence in those limits.

Most projections are made by comparing curves from different countries and applying the same trajectories here, said Murat Kristal, of York University’s Schulich School of Business, where scientists have developed an analytical dashboard to predict the global spread of COVID over five days. 19.

Italy is one of the most affected countries in the world. On March 1, there were 1,701 confirmed cases in Italy. As of April 6, there were 132,547, nearly 78 times. On April 1, the number of confirmed cases in Ontario was 2,292. “If we apply the same trajectory from Italy, the number of cases would be around 180,000 in Ontario by April 30,” said Kristal.

The trajectory of confirmed cases in Ontario between March 1 and April 1 was 159 times. “If the same trajectory is applied from April 1 to 30, in the worst case, we can reach 300,000 confirmed cases in total.”

Social distancing does not change the likelihood of contracting the virus, just the moment. Never seen before in humans, the virus is extremely effective at spreading between us, and it is very likely that almost everyone will understand it.

“The question is when, not if. And that’s a crucial question, “said Robert Smith, professor of disease modeling at the University of Ottawa.

The idea behind social distance is to avoid drowning intensive care units. Last week, Ontario released guidelines on triage when there is a significant increase in the number of cases. In the worst case, anyone with a 30% chance of dying would be denied an intensive care bed, not just COVID-19 patients, but those admitted for cardiac arrest, people with advanced cancer or those with a “frailty score” greater than or equal to 3, which means that only “very fit” or “well” (people who do not suffer from any active underlying disease, who often exercise or are “very active occasionally”) would receive an intensive care bed. Those denied admission to the intensive care unit would receive palliative care, including pain and “comfort drugs,” the document said.

The idea behind “flattening the curve” is to give each seriously infected person the opportunity to receive intensive care and ventilation. But intubation does not guarantee survival. Early reports suggest that even with the support of a ventilator, half of patients die between day 1 and day 18 of intensive care.

Those most at risk for intubation are also most at risk of dying from competing causes, such as heart disease or chronic obstructive pulmonary disease. The 27 deaths in an Ontario nursing home are tragic. But the median survival for residents of retirement homes at the end of their life is five months, according to a 2010 study. Although very discouraging, “it is very likely that most, if not all, of these deaths would have occurred independently of the COVID-19 epidemic, “said one doctor, whom we agreed not to identify” because dissidents in the medical community are likely to be targeted, “said the doctor.

“We must consider that our health system is mortgaging the future prosperity of our young people for an impossible promise of immortality.”

The World Health Organization warned last week, however, that young people fall seriously ill and die from COVID-19. In Italy, 10 to 15 percent of all members of intensive care units are under the age of 50. In Canada, as of April 1, the highest proportion of reported cases was among those aged 40 to 59 (36 percent), followed by 20 to 39 (29%) and 60 to 79 (25%), although the proportion the highest number of hospitalizations and ICU admissions is among those aged 60 and over (59%).

“There has been a trend over the past few months, almost a disdainful attitude, to say,” Well, this disease is serious in the elderly, and it works well in the young, “said Dr. Mike Ryan, director of the WHO emergency program. “We collectively live in a world where we have tried to convince ourselves that this disease is mild in young people and more serious in the elderly, and therein lies the problem.”

On April 7, 2020, doctors cared for intensive care patients in the COVID-19 ward of Maria Pia Hospital in Turin, Italy.

Marco Bertorello / AFP via Getty Images

Ontario models do not make assumptions about unknown cases. “Let’s be honest and frank,” said Donnelly. “There are clearly many cases in the province that we do not know about … But again, these worst-case models and the direction in which we are now thinking do not depend on this type of data.”

At the start of an epidemic, he said, it is about providing policymakers with an important “early guide” on what to do. As soon as the government’s “command table” saw the figure suggesting that there could be an overall mortality of between 90,000 and 100,000 deaths, they quickly decided to close the schools.

Modeling is good if the data is perfect. But we never have perfect data, and critical information is missing: what is the transmission rate? What proportion will live or die? Can infected people be reinfected? To what extent do asymptomatic people drive this? The data changes almost daily. When you look at a timescale of several months, it’s like driving through a blizzard.

York’s analysis suggests that the rate of growth in the number of confirmed cases in Canada appears to be slowing. Epidemiologists are already preparing to point the finger when attack rates are not the worst, as governments have instituted social distancing.

But an almost medieval “just shut the door and stay at home” public health approach is not feasible in the long term, given the impact on the economy – hundreds of billions of dollars to support millions of people. who have lost their jobs and their businesses – and society as a whole, said Dr. John Ioannidis, epidemiologist at Stanford University, in a recent Monk Debates podcast.

COVID-19 is a serious threat, he said, but “there are many things we can go wrong if we don’t do it right”.

• Email: [email protected] | Twitter:


Please enter your comment!
Please enter your name here