The BluePrint: #14 - Lorien Pratt: Decision Intelligence is the "glue" that allows us to use AI in new ways, such as for COVID19 safety

"Making decisions about risk and uncertainty is hard. For instance, trying to think through the probability of catching a potentially deadly virus while going to an indoor movie theater is difficult. So people tend to think in terms of binaries — 'this is safe' or 'this is unsafe' — because it’s easier.

The problem is that answering easier questions instead of trickier ones leaves you vulnerable to cognitive biases, or errors in thought that affect your decision-making.

One of the most prevalent of these biases is the availability heuristic. That’s what psychologists call the tendency to judge the likelihood of an event based on how easily it comes to mind. How much a certain event is covered in the media, or whether you’ve seen instances of it recently in your life, can sway your estimate."

"This overview of the history illustrates the pervasiveness of “belief perseverance,” the psychological tendency to maintain a belief despite clear and strong new evidence that should challenge it. In an era of amazing scientific advances, where mRNA vaccines were designed in a few days following virus sequencing also obtained in a few days, the very slow acceptance of critical new knowledge reminds us that the human aspects of science remain as pervasive as they were in past eras.

However, the intense research and debate associated with the COVID-19 pandemic has finally begun to generate a paradigm shift in the understanding of disease transmission. Not only are respiratory diseases not transmitted exclusively by droplets, but also it is likely that many or most respiratory diseases have an important airborne component of transmission. It is also clearer that for a disease to cause a fast-spreading pandemic, airborne transmission is likely to be an essential component. This does not mark a return to past miasmatic ideas, but a more informed understanding of airborne transmission as more complex and less scary than in the past, and certainly as a tractable problem."

"'We understood that washing your hands and singing ‘Happy Birthday’ wasn’t going to quite cut it,' explains Chief Operating Officer, Craig Fenton. 'Covid-19 highlighted a problem in the construction industry that buildings are designed with aesthetics and comfort in mind, not biosafety for occupants. Our goal is to help design buildings that are more biosafe and reduce the risk of disease and infections for people.'

EPCC – a world-class supercomputing centre based at the University – supported their simulation tools, while their work with UNCOVER – a network of population health researchers and information specialists providing Covid-19 evidence reviews for policymakers – helped shape the platform’s epidemiological modelling."

"During this pandemic, we’ve developed and deployed vaccines in real time. Meanwhile, sars-CoV-2 is replicating not in a dozen flasks but in tens of millions of people, some of whom have been immunized, all of whom exert selective pressure for the virus to find new, more efficient replication strategies. The virus will continue to mutate every moment of every day, for years, for decades. The fear is that it will hit upon a second citrate moment: a mutation, or set of mutations, that enables it to circumvent our vaccines, which so far have proved spectacularly effective and resilient."

As we think about restrictions lifting and potentially getting back to work, how will we know our workplaces are safe with a highly aerosolised virus going around? We already have this problem and accept roughly 20,000 deaths from flu each year - but what is the answer?

"...medical lore has it that such droplets—defined as particles expelled while breathing that are more than five microns across—cannot travel more than a couple of metres after they have been exhaled. And some of those who became infected during the lunch were farther than that from the “index” patient."

"The long-standing misunderstanding of the importance of airborne transmission of pathogens has left a large gap of information in how to best construct and manage building ventilation systems to mitigate the spread of disease—with the exception of some manufacturing, research and medical facilities. Instead, buildings have focused on temperature, odor control, energy use and perceived air quality. So while there are safety guidelines for chemicals such as carbon monoxide, there are currently no guidelines, globally or in the U.S., that regulate or provide standards for mitigating bacteria or viruses in indoor air resulting from human activities.

'Air in buildings is shared air—it's not a private good, it's a public good. And we need to start treating it like that,' said Miller.

...

Lidia Morawska, director of Queensland University of Technology’s International Laboratory for Air Quality and Health. (Credit: Queensland University of Technology) 

Lidia Morawska, lead author on the article and director of Queensland University of Technology’s International Laboratory for Air Quality and Health, said there needs to be a shift away from the perception that we cannot afford the cost of control. She notes that the global monthly cost from COVID-19 had been conservatively estimated as $1 trillion and the cost of influenza in the U.S. alone exceeded $11.2 billion annually."

Top experts in how diseases spread are calling for massive improvements to the air in buildings.

"To date, the transmission of COVID-19 has been classified by major institutions as Contact/Droplet, which has resulted in mitigation protocols such as handwashing, masking guidelines, physical distancing rules and disinfecting surfaces. Others have challenged this definition, and assert that in addition to contact/droplet, airborne/aerosol plays an important role in transmission. They are calling for additional mitigation protocols, such as respiratory PPE and ventilation strategies. This seminar will host an international panel composed of global leaders who will break down the science behind the two paradigms, and explore whether we should double down on current efforts or re-think our approach."

"Federal health officials on Friday updated public guidance about how the coronavirus spreads, emphasizing that transmission occurs by inhaling very fine respiratory droplets and aerosolized particles, as well as through contact with sprayed droplets or touching contaminated hands to one’s mouth, nose or eyes.

The Centers for Disease Control and Prevention now states explicitly — in large, bold lettering — that airborne virus can be inhaled even when one is more than six feet away from an infected individual. "

"Decades of painstaking research, which did not include capturing live pathogens in the air, showed that diseases once considered to be spread by droplets are airborne.[4] Ten streams of evidence collectively support the hypothesis that SARS-CoV-2 is transmitted primarily by the airborne route.[5]"

So far only 10 controlled trials of non-drug interventions have been registered, with three reported.

This makes no sense.

"...This is a symptom of a number of problems we face as humanity. These include Covid-19, climate change, and global inequality. These problems:

• Are invisible
• Involve exponential behavior dynamics
• Involve actions that play out over long chains of cause-and-effect in time and space
• Involve complex systems interactions including feedback and winner-take-all mechanisms
• Can only be solved with great science and data."

If annual averages were the only information we considered, 79 AD in Pompeii was a slightly warmer year than usual—hardly a situation worthy of serious concern. As you can see with this example, averages, when assessing outlier risks, often make very bad metrics.

"...the temperature of 95°F is such a value. Stay below it, and very few people die. Rise above it and people start dying, and the higher the temperature rises, the greater the mortality. The chart below shows why a 5.7°F rise in average temperature over 50 years can lead to such deadly results."

The agency has not fully reckoned with airborne transmission of the coronavirus in settings like hospitals, schools and meatpacking plants, experts said.

"In the last three decades, we’ve seen a massive change in marketing. It used to be that “one size fits all” television advertisements sent to the entire country at a time were very expensive. In contrast, today, we can spend much less money on a simple Google, Facebook, or Twitter ad that only appears on the screen of people who are likely to want to purchase the product or service. Once a pipe dream, this “one-to-one” advertising model is now standard practice worldwide....

However, this revolution in data and AI technology has not been, by and large, applied to the most impactful problems of our time, including Covid-19 and climate change. We see guidelines issues by federal and state authorities that have the same “one size fits all” character, reminiscent of television ads of the 1970s. Data about Covid-19 and climate is presented in summary statistics characterizing entire states or counties, and is often averaged out over long time periods, thereby losing its power."

"Evidence gathered this fall around the world and in the U.S. suggests that schools can open, even in conditions of wide community spread, and achieve low and even near zero transmission in the school building. This evidence, combined with the benefits to learners of in-person schooling and harms of remote schooling, suggests that the time has come to pursue in-person learning across most school contexts, provided that the school in question has established reasonable infection control protocols to safeguard student, educator (including paraprofessionals), and staff safety. The federal governments should include investments in school infection control in the next coronavirus relief package.

...The most important elements of infection control that matter are:

• universal masking (including while speaking)
• hand and bathroom hygiene
• achieving 4-6 air changes per hour of 'clean' air through any combination of ventilation and filtration (or outdoor classrooms)
• 3 ft social distancing for young learners at all levels of community spread
• 6ft social distancing for high schools when levels of community spread rise above 100/100,000 daily new cases; 3ft social distancing below that level
• robust quarantine policies and contact tracing practices
• and, where feasible, surveillance/screening testing, also discussed below under “testing.”

"
(boldface mine)

The project involves an environmental monitor small enough to fit on a desk that tracks air quality and ventilation; a wrist monitor that tracks sleep and physical activity; and an iPhone app that feeds researchers real-time data.

"Today there are over 55 million documented COVID-19 cases and over 1.3 million deaths, impacting countries and territories worldwide. As vaccines are developed and countries implement recovery plans, regional authorities lack the ability to test “what if” intervention plans that optimize COVID-19 mitigation strategies while reducing the economic and social impacts."

"From office workers to students, Americans facing colder weather and more time inside have a pressing question: How can they keep safe amid a pandemic that scientists say thrives in indoor settings?

The search for answers has prompted a new look at what architects and their buildings can do to help, both now and in the future."