COVID-19 and the Swiss Cheese Effectby edward tenner
edward tenner, a research affiliate of the Smithsonian Institution and Rutgers University, is currently a Visitor in the Program in Interdisciplinary Studies at the Institute for Advanced Study at Princeton.
Published August 18, 2020
After culture-altering catastrophes, the search for underlying errors that make sense of the improbable seems both natural and necessary. But there’s more to gaining insight than identifying who flubbed what. We need to understand the mechanisms by which systemic problems become crises — and these often depend on bewilderingly complex sequences of events.
A Pandemic Like No Other
The Covid-19 pandemic, many would say, was widely foretold. The journalist Laurie Garrett was already warning the world about the risk of a global infection on the scale of the 1918 influenza outbreak in her 1994 book, The Coming Plague: Newly Emerging Diseases in a World Out of Balance, and sounded an alarm about the fragile state of global public health in Betrayal of Trust. In September 2019, when Covid-19 may have already leapt from beast to man in China, she wrote an online column for the journal Foreign Policy on the approach of “an apocalyptic pandemic.”
Between 2011 and 2018, the World Health Organization had fought against some 1,500 separate epidemics. Yet the report of the independent Global Preparedness Monitoring Board cited by Garrett missed an essential feature of Covid-19. It foresaw that a new pandemic would devastate developing countries’ economies, but it predicted that advanced industrial nations like the United States and Germany would lose only half a percentage point of GDP. In fact, while the toll in human life has been mercifully far smaller than the benchmark 1918 pandemic (caused by the H1N1 flu virus), the purely material price has been orders of magnitude greater. The Congressional Budget Office has conservatively estimated an inflation-adjusted toll of $7.9 trillion for the U.S. over the next decade, a 3 percent loss in GDP.
Why did the experts miss the economic threat of the new pandemic? One reason was a confluence of circumstances that synergistically magnified the impact. Like SARS and swine flu, Covid-19 almost certainly originated in China. But Covid-19 had one unexpected, and by now notorious, characteristic: a high degree of contagion during a long asymptomatic period, which made traditional contact tracing inadequate for quarantining carriers. (China was incredibly fortunate that the virus was detected before the mass travel of the Lunar New Year and acted decisively to shut it down.)
It took months for U.S. public health authorities, who had been monitoring travelers from China, to realize the extent to which the virus had been spreading in Europe. Some of the “super-spreaders” never experienced symptoms of their own yet managed to infect dozens. As late as March 8, Anthony Fauci was discouraging citizen use of face masks — a position shared at the time by the U.S. Surgeon General and the WHO. Even as late as May 21, a distinguished group of frontline physicians wrote in the New England Journal of Medicine that “we know that wearing a mask outside health care facilities offers little, if any, protection from infection.”
Once the value of wearing masks to protect against asymptomatic transmission was understood, a bizarre new obstacle to containment emerged: the transformation of reluctance to wear masks into a sign of political solidarity with Donald Trump. With hindsight, then, it has not been the virulence of the infection — the 1918 flu and both SARS and MERS were more lethal — but a confluence of obstacles to decisive action that has made the pandemic so grave.
The pandemic illustrates how chains of improbable independent factors can turn unfortunate events into catastrophic ones. Risk analysts call this the Swiss cheese effect (or Swiss cheese model)
A Slice to Remember
The pandemic illustrates how chains of improbable independent factors can turn unfortunate events into catastrophic ones. Risk analysts call this the Swiss cheese effect (or Swiss cheese model). Imagine slicing a block of Emmenthal that contains randomly placed hollow spaces — the latest thinking, by the way, is that they are caused by tiny particles of hay in the milk, not gas bubbles — and then randomly rearranging and stacking the slices. Every so often the holes will align so light is visible through the stack. If we think of each slice as a potential barrier to some unwanted event, it may be only a matter of time before a new way to stack creates a fateful tunnel.
The destiny of the Titanic offers the best-known example. Count the “slices” that had to align:
• the limited maneuverability of the new great liners
• the odd but universal custom (as a parade of captains testified) of maintaining speed through ice fields
• the fairly unusual dark color of the fateful iceberg
• the rare atmospheric conditions that concealed the iceberg’s presence
• the smoldering coal fire (the indirect result of a mining strike) that may have weakened the steel used in the hull
• the angle of the collision
• miscommunication with the nearest potential rescue ship, the Californian
• the absence of lifeboat drills
Likewise, the severity of the 1918 pandemic was related to eccentricities of timing and vulnerabilities. As the epidemic specialist Dr. Jonathan D. Quick has pointed out, President Woodrow Wilson was so concerned about maintaining wartime morale that he — like leaders of the other belligerent nations — suppressed news of the outbreak until it was too late, resulting in the loss of at least 675,000 lives in the U.S. alone. (Fun fact: the pandemic was called the Spanish flu because only in neutral Spain were newspapers allowed to publicize it in its early days.)
Over a quarter of the U.S. Army contracted the flu — in contrast to Covid-19 and most flus, H1N1 proved especially lethal to the otherwise young and fit — and nearly 30,000 died before reaching France.
The most vivid illustration of the Swiss cheese effect deserves to be remembered alongside the Titanic and the pandemic. On March 27, 1977, two Boeing 747s (flown by KLM and Pan Am, respectively) collided on the runway of Los Rodeos airport in the Canary Islands, resulting in a loss of 583 lives.
The pilot of the KLM plane, Jacob Van Zanten, was one of the most respected in Europe — as, indeed, Captain Edward Smith of the Titanic had been in transatlantic shipping. Van Zanten trained junior pilots and had been featured in KLM advertising. If the KLM 747 had not required additional fuel, it would have taken off while the weather was clear. But a thick fog rolled in shortly before the plane was due to leave. The terrain was also relatively unfamiliar to both pilots: Because of a terrorist bomb attack on the islands’ main airport, the two charters had been diverted to this secondary airport. And because of ground congestion, the two giant planes had to make an unusual maneuver, putting both on the runway at the same time — at an airport with no ground radar and with the planes invisible (thanks to the fog) to the control tower.
Meanwhile, the pilots were communicating with the tower over two-way VHF radios like walkie-talkies, which could render simultaneous speech unintelligible. Van Zanten, impatient to take off, failed to understand that the other plane was in his path. When he did discover it, he desperately accelerated to gain altitude and nearly cleared the Pan Am 747 — but instead sheared off its top, killing all passengers in his own plane and sparing only a small number of Pan Am crew and passengers in the nose.
Think of all the cheese holes that had to line up:
• the diversion of the planes due to a terrorist bombing
• the fog that obscured vision
• the added weight of a fresh load of fuel that prevented the KLM plane from clearing the Pan Am plane
• Van Zanten’s hurry, ironically caused by his need to return to Amsterdam before he exceeded his maximum allowed flying hours
• Van Zenten’s vast experience, which made other crew members hesitant to interfere
• • •
Tenerife catalyzed a very successful new approach to flying called Crew Resource Management, training that prioritized communication and early identification of potential problems. Crew Resource Management effectively demonstrated that messing with the cheese slices by analyzing social processes can interrupt the accumulation of small failures that lead to very, very big ones.
Could similar approaches prevent epidemics or help reduce their impact? The Swiss cheese metaphor certainly makes it clearer that we really do need to understand the social processes that allow those holes to align catastrophically.