Speaking with Richard Bruns about indoor air quality

I recently spoke to public health economist Dr. Richard Bruns about indoor air quality — why it’s a big deal, and what we can do about it. I learned a lot and I think the topic is important, so I made a transcript which I’m sharing here.

We also talked about why it probably wouldn’t be good overall to do anything like abolish the FDA, and whether a culture’s stock of fiction (movies, stories) influences its future in a causal way.

🎧 Listen to the audio ↗

Fin

Dr. Richard Bruns is a senior scholar at the Johns Hopkins Center for Health Security and before that he was a senior economist at the US Food and Drug Administration, the FDA. So if you are thinking about pandemics, then one kind of obvious risk factor is the fact that most of us are spending 80-90% of our time indoors in buildings. buildings with unfiltered air where everyone is just breathing in a bunch of pandemic potential germs that everyone else is breathing out. But of course that also means that if indoor air were cleaner then we’d just get way fewer respiratory infections in general, things like flus and colds. And less obviously it turns out that fine particles in untreated air have a similar health burden again because they cause heart attacks and strokes.

But then we also just have… very concrete ideas for how to make indoor air cleaner, things like filtration and far UVC light. And so it was fairly easy to imagine a world where we just get our act together and save maybe hundreds of thousands of lives in the process. And Dr. Bruns has actually been doing a bunch of modeling and cost benefit analysis around indoor air quality. So he was a very good person to speak to and he was able to get fairly concrete about the kind of policy changes that could really help. We also talked about how cost benefit analysis works. works in general, why abolishing the FDA might not be a great idea, and much more. But yeah, very important problem and one we know how to fix, which is great. Okay, here is Richard Bruns.

Richard Bruns, thanks for joining me.

Richard Bruns

Glad to be here.

Fin

How do you describe what you get up to right now?

Richard Bruns

Very briefly, I am an economist that specializes in cost-benefit analysis of public health policy. For the past four years or so, I’ve worked at the Johns Hopkins Center for Health Security, which is a think tank.

Fin

And what are you currently working on, if there is one big thing?

Richard Bruns

Mainly, for the past couple of years, I’ve been doing a lot of work on indoor air quality. We’ve got a lot of projects going on at the center. And I’m working on our Model State Act for indoor air quality. And also I’ve been on the ASHRAE 241 Infection Control Standards Committee.

Fin

Awesome. And I hope to talk about both of those things. And I guess I mentioned this before we started recording, but the reason I’m especially excited to have this conversation is that, be persuaded that indoor air quality just seems like a very big deal and also seems relatively under-discussed, at least until recently. But just to begin with. What are we talking about when we use this term indoor quality?

What is Indoor Air Quality?

Richard Bruns

Okay. So, Basically what it sounds like, any condition inside that affects the air, there’s a lot of things involved with it, some more substantial than others. Historically, indoor air quality has mostly been thought of as thermal comfort or the humidity level, because that’s what people can immediately hear and perceive. So for a long time, all of the HVAC, that’s heating, ventilation, and air conditioning were really just focused on, “how do we get people comfortable inside”? And that is important. I mean, when people are very uncomfortable, you have serious productivity effects. Like there are a lot of places, non-air-conditioned workshops in the global south, etc., even some other places, have serious productivity effects. But in recent years, we’ve much more concerned with contaminants that are invisible. You don’t even smell them or notice them, but they have serious long-term health harms. One of the main ones of those is fine particulate matter or PM2.5, and also of course respiratory viruses and other pathogens.

Health Impacts of Particulate Matter vs Pathogens

Fin

Okay, got it. And I guess we should take those both in turn. So PM2.5, I guess I also hear PM5. Yeah, what does that mean? Is this like tiny bits of dust in the air?

Richard Bruns

Yeah, the number is the size. PM2.5 is smaller than 2.5 microns in diameter. A bunch of these can be strung around a hari like a beaded necklace — they’re very small, so they bypass the filters in your nose and airways designed to trap these things and often go straight into the lungs, straight into the bloodstream where they often cause heart attacks and strokes after chronic exposure (and after acute exposure if there’s enough of them).

Okay. a long time. and the chemicals and gases and they can cause problems. The reason that they’re

Fin

Is that the main health burden of particular matter in the air, or are there some cognitive effects as well?

Richard Bruns

If it’s not a respiratory virus, then PM2.5 is what kills you. There’s a bunch of other stuff people talk about, like a bunch of chemicals and gasses, and they can cause problems. The reason people talk about them is that they tend to be more industrial contaminants like in a factory; it was an OSHEA type issue. But it’s not as much of an issue any more. I’ve seen some estimates of the QALY or DALY losses that are caused by these different, all these different contaminants, and the particulate matter is an order of magnitude higher than even the second place contaminant. [So] to a first approximation, the thing that kills you with bad air quality is particulate matter and pathogens.

Fin

Okay, and just the mechanisms from the particulate matter, which just cause these like QALY losses or deaths. This is respiratory illness, heart attacks, anything else?

Richard Bruns

Actually, no. It goes straight through into the bloodstream. So most of what they cause, well, respiratory illnesses are a big chunk of it. There’s definitely asthma and other kinds of things, but there’s a giant pile of studies. Usually these are based on outdoor contaminants. Like there’s a lot of data coming out of China where they have like how the levels change for different levels in different cities. But there’s noticeable changes in all kinds of cardiovascular illnesses from this stuff when it gets too high.

Fin

Okay, right. Got it. And I noticed you didn’t mention carbon dioxide, but people like to talk about, you know, CO2 buildup having these kind of cognitive effects. I don’t know if that translates into QALY losses.

Richard Bruns

Basically, CO2 is an indicator rather than a target. Like, it’s easy to measure, and if the CO2 is high in a level, it means that the ventilation isn’t very good. So it’s a very reliable indicator. Like, if CO2 spikes above 1,000, definitely above 2,000, it means you’ve got too many people in the room exhaling too much stuff, and the HVAC system isn’t cycling it out. So it’s definitely a sign that you’re at much higher risk from especially respiratory viruses because you’re breathing in air that’s been breathed out by other people. But the CO2 itself the evidence is kind of mixed; like in submarines the levels are super high and they don’t seem to have a huge effect. So the evidence is mixed; some people will see larger effects than others. But it’s, again, even with the higher levels of reported harm, it’s nothing near the PM2.5.

Fin

Okay, got it. So there’s particular matter as one of the big drivers of deaths and health costs. The other one is… pathogens, are there particular pathogens that are at the top of the list?

Richard Bruns

Basically the pathogens that kill people. Every year flu kills 50,000 people in the United States. I don’t know what the stats are for the UK and Europe, but it’s a similar percentage of the population. So in G7 countries, you’re looking at hundreds of thousands of deaths per year from the flu. We’re not entirely sure what endemic COVID is going to look like, but I’m fairly certain it’s going to be at least as bad as that. If you look at the last winter of COVID. I mean, best case scenario, we get an updated booster for COVID every year, but there’s going to be new variants, new stuff happening. We’re not even sure about long COVID, so there’s a lot of uncertainty there. But the total DALY loss from persistent COVID into the future could be anywhere from the same magnitude as seasonal influenza in the past, like four or five times as bad.

Fin

Okay. And just speaking at a high level, when you’re comparing the DALY or QALY losses from the particulate matter on the one hand and pathogens on the other hand, excepting pandemics, does one stand out as a much bigger cost than the other?

Richard Bruns

No, the confidence intervals overlap. As I said, they’re both pretty big. We’re not entirely sure. There needs to be a lot more research, even on the PM2.5. We know it’s bad, but there’s a big uncertainty in the confidence intervals. So yeah, they overlap so much. We don’t know for sure which one is bad. Yeah. New strain of COVID, long COVID being worse of a long-term DALY cost than we think could make it a lot worse. But there’s, we’re also uncovering new things about how bad the PM2.5 is. So yeah, they’re both pretty bad.

Fin

And could you say a bit about how you go about coming up with the estimates for the cost of indoor air quality?

Richard Bruns

Well first of all, almost everybody who dies from the flu is going to be dying from getting the flu inside.

Over 90% of respiratory virus transmission is indoors. Quite a lot of that could be prevented by better indoor air quality. So yeah, the pathogens are almost entirely an indoor air quality problem. The PM 2.5. less so because a lot of that comes from outside. In general, indoor air actually has lower levels of smog in most areas than outside air because there’s a lot of sinks. Like they just kind of get sucked into the carpet and walls and stuff. And especially if you have any kind of good filtering system. As people would have learned in the wildfires, you close up your windows, run the air through a good filter, the particulate matter inside is actually better than the stuff outside. There’s a lot of out-gas, like chemicals, like chemicals in your house, like the formaldehyde, stuff like that, that you smell, especially anything coming off of glue and solvents and cleaning chemicals. Those are often worse inside. But again, that’s not the stuff that kills you. I still like opening the windows and getting fresh air when it’s not, you know, a smog alert.

But for how the research works, I don’t know many technical details, but it’s very rare for people to catch Covid outdoors, and to catch flu outdoors. There’s a bit of contact tracing there, a bit of statistical epidemiology of what were people doing, where do this spread events come from. If anything, COVID was a, told us a lot more about how these things spread because unlike flu, it was a new thing in the population that there was a lot of effort doing contact tracing in a new strain. So you could actually say, okay, how did this person get sick? We assume that COVID and flu spread in relatively similar ways.

Fin

Yeah.

Richard Bruns

As to how we know about the PM 2.5, again, the… A lot of the research there is very much city-level observational stuff. We see one city has higher smog levels. We do the observational things. There’s a variety of estimates. They all seem to be pretty strong, pointing in the same direction. I don’t know if you’ve heard about the natural experiment of closing toll plazas or switching to E-ZPass.

Fin

I haven’t. You want to say about that?

Richard Bruns

Okay. So back when I was in grad school, a lot of the economists were very skeptical about this epidemiological literature on the harms of air pollution. You probably are familiar with, or you’ve posted, Alex Tabarrok on his blog has talked about the increasing literature. But economists typically don’t trust the kind of observational epi studies that you’re not correcting for enough stuff. And there’s definitely a lot of value in that criticism, but a large pile of weak evidence is still pretty suggestive if you have some trust that people aren’t trying to push an agenda, which I kind of do. The example of the toll plaza is they basically switched to E-ZPass at a certain date. So when you don’t have E-ZPass, everybody is slowing down, you know, paying money into or to the attendant or throwing it in the basket. When they switch over to E-ZPass, you’re just driving straight through, so you don’t have the calls, the cars stalling and idling.

So they’ve proven that, yeah, it generates a lot more pollution. And this was basically an exogenous natural experiment that showed like… This had nothing to do with the demographics or characteristics of the neighborhood. It’s just some they randomly put easy pass at this toll station. And you can look at the rates of asthma and other illnesses in the neighborhood and they drop pretty substantially. So that tells you that the smog and pollution coming from the cars had a clear causal effect on the population nearby.

Fin

So okay, the areas near to these tolls are getting some pollution from them. We have this natural experiment where some of those tolls are being closed for random reasons. or they’re being closed at random times. Exactly. And the effects are big enough that we can observe them in asthma rates or whatever? Yes. Okay, cool. I didn’t know that.

Richard Bruns

Yeah, so I’m actually… this is something that I’m working on, the estimated cost-benefit analysis for our model state indoor air quality act. And basically, the three main numbers that I’m looking at, the monetized DALY improvement from reduced virus transmission should be about 600 to 3,000 per person per year. Okay. The reduced PM 2.5, but basically about 1,500 per person per year. I think that’s the high end and the low end is about half of that. And then you have the estimated productivity losses of people functioning better. That’s probably about 400 to 1,500 per person per year. So overall. Overall benefits from reasonable improvements in indoor air quality. This isn’t like the total cost of things affecting you. This is we bring bad buildings up to the reasonable standards. You’re looking at single digit thousands per person per year in benefits for a cost of a dozen, probably about a hundred bucks per person a year at most for the cost of the better HVAC system and filters and electricity to keep it going.

Fin

Okay, so we have three numbers there. One was the dollarified health cost from breathing in particulates in the air. Second was the health cost from pathogens. And then the third was just the general cost from productivity losses from generally bad air. Is that right?

Richard Bruns

Yes.

Fin

Okay. I think I can understand how you might figure out the productivity losses from unfiltered air. But could you say something about how you might figure out the health costs in dollar terms?

Richard Bruns

Sure. Basically, the methodology is the one used by the US Department of Health and Human Services for valuing morbidity and mortality risk reductions. It’s a standard procedure, the same one I used at FDA. It’s basically the value of a DALY is about half a million a year. So you calculate how many disability adjusted life years or quality adjusted life years. They’re basically the same for this calculation. How many of those you gain, multiply that value by half a million. EPA and HHS have slightly different values and there’s a bit of, of presumably these VSL figures or these dollar figures for a DALY or a QUALY.

Fin

They’re not just made up because they sound right. So could you say where these numbers come from?

Richard Bruns

They basically come from a revealed preference. Almost all of the numbers come from differentials in jobs. So you basically try to find jobs that appear to be observably identical in all of the characteristics. Some jobs have a higher risk of death than others. You look at how much more people have to be paid for that risk. So this $10 million, it isn’t something that an economist just decided in terms of, oh, I’m gonna put a value on a human life. It’s basically looking at the choices that people make about their own lives and their own salaries. So one way of thinking about it that’s a little almost edgelordy, but not entirely not wrong, is when we say, The value of a statistical life is $10 million. We’re not making a value judgment when we say the value of a statistical life is $10 million. We are making a prediction that if a government does something that sucks $10 million out of the economy, that will lead to people changing their behavior in ways that cause somebody to die. So, yeah. If the government collects more than $10 million in taxes to save a life, it will statistically kill more people than it saves because poverty kills people and you’ve made people poorer.

Fin

Okay, and this comes from an observation that people in general are going to have some rates at which they trade off more salary against an increased risk of dying in choosing a job where they have a choice. And then if you become wealthier, you can afford to place a higher dollar value on your safety. And so similarly, if you become poorer, then money is going to be relatively more valuable than your safety per dollar because money has diminishing value to you. And so you’re more likely to take more risk to your life in choosing a job. And from that, you can get some kind of sketchy number for the value of a statistical life by just looking at, for instance, the pay differences between safe and risky jobs. Exactly. And then figuring out what that says about how people are implicitly valuing a year of expected life for themselves. Is that roughly right?

Richard Bruns

Yeah, and they’re pretty wide, like, you know, there’s a wide array of values, but the, the average of that generally converges to 10 million in an advanced country. Of course, the, and this is one of those things that it’s easy to get misquoted and can be confusing, but. When we say the value of a DALY or a statistical life is lower in poor countries, that’s not at all a judgment on the value of their lives. It’s, again, the empirical prediction. If you take 10 million in taxes out of a poor country, you will statistically kill a lot more people than you would if you take 10 million in taxes out of a G7 country because the people in those poorer countries are… in a situation such that they have fewer resources and they’re forced to accept a higher risk of dying for their additional wages and so on.

Fin

Yeah, it’s not a good thing that these numbers are sometimes low and let’s care about making them high.

Richard Bruns

And you have to be very careful on that. Like some of the, even the official documents for doing this put up by the Bill and Melinda Gates Foundation, like they make this mistake of… If you prioritize health interventions based on how much monetized value that they’re giving, and you use a lower monetized value in a poor country, you are actually discriminating against poor people. So that’s why I think it’s very important whenever you’re doing any cross-border analysis of the costs and benefits of a health intervention, What you should do is convert the dollars into health loss, life expectancy loss, based on the exchange rate and the place you’re taking the money from.

So the conversion, the value that you place on the life should always be based on where the money’s coming from, not where it’s being spent. Because obviously, if you are proposing a government policy that takes $10 million in taxes from the people in DRC, and uses it to try to save so many lives, you’d better be saving a lot more than one life. You’d better be saving hundreds of lives for that $10 million that you’re taking away from extremely poor people, because taking… You know, taking $10 million away from people in the DRC is going to kill at least dozens of people. So that’s why you should use the lower value for statistical life if you’re taking money from them. But if you’re taking money from rich people to give to poor countries, I think it’s fundamentally wrong to use the lower VSL based on that country. Does that make sense?

Fin

I think that makes sense. Does that mean, for instance, that for the same amount of money in the same currency, You should always expect to be able to save more lives in the country with the lower value of a statistical life?

Richard Bruns

Actually no. The value of a statistical life doesn’t tell you how many lives you can save. I mean, it typically winds up being correlated, of course, but how many lives that you can save by spending money must be based on an evaluation of that particular program. The value of a statistical life is our best estimate of how many people you’re going to kill by destroying that much economic activity in that country via taxes or regulations.

Fin

Yeah okay that’s useful. And one thing that implies I guess is that the richer a country gets the smaller the cost in expected deaths, excess mortality, from raising the same amount of money; same absolute amount of taxation per person?

Richard Bruns

Of course, because by definition, people in a rich country have… Like you can think of people picking lower and higher hanging fruit. It’s like the richer you get, the more money you have to spend on doing things that reduce your personal risk. And assuming people generally spend these things in a rational manner, they’re going to start by buying more and more stuff for themselves. So you’re taking less marginally valuable purchases away from people when you tax richer people.

Fin

Yeah, that makes sense. I kind of want to ask about interventions for indoor air quality now. So I mean, I think there’s a very obvious high level question here, which is just what is on the menu for improving air quality indoors in, let’s say, rich countries where most buildings are going to have some kind of basic measures. So what are the top things? And I think that’s a really good question. I think that’s a really good question. I think that’s a really good question.

Richard Bruns

Well, your blog post has a really good summary of the tech. You even found the warnings about indoor smog and GUV. So I don’t think in terms of that technical stuff, I have that much to add. But I guess if you want me to repeat for your listeners, I can.

Fin

That would be useful. I’m going to guess that most people haven’t read that.

Richard Bruns

Okay. So. Basically, the best interventions are filtration. You run air through a good filter, MERV 13 or better, and that cleans the crap out of it. You want, that can be done through either the central HVAC system or portable in-room filter units. What I, what you should always do, every individual who has any ability to control your own HVAC system, you basically buy a MERV 13 or better filter. I mean, I basically just went to Home Depot and got the best filter system they had. They had this weird other thing that kind of matches. But the cost difference between a crappy filter and the best filter is like $3 versus $11 if you buy them in bulk. It’s a tiny marginal difference. And with the higher-end filters, you do have to change them more often because they can get a bunch more junk sucked into them that can mess up the system. So you’re basically paying $40 a year to get the really good filters, change them out four times, and the filter gets the crap before it gets into your lungs and your bloodstream. That’s the obvious thing that everyone should do personally.

And we should find ways of incentivizing everyone to do that in larger rooms that they’re responsible for, which we’ll get to later. And then after you’ve done that, then you can start to move into in-room portable. They call them portable HEPA filter units because that’s how they’re branded. But really, the quality of the filter doesn’t matter as much as how much air you’re passing through the filter. Passing a lot more air through a MERV 13 filter will do a lot more good for you than passing a little bit of air through a HEPA filter. That’s why this Corsi Rosenthal boxes are a really good DIY project. You’re probably familiar with them.

Fin

Yeah, I do have to describe what they are.

Richard Bruns

Yeah, so basically you get a box fan, and you make a box with them, so the fan is pointing upwards. So the fan is blowing the air upwards, and it’s taking the air through four different filters, MERV 13 or better, that you put around it in a box. So you’ll have it on the table or on the floor. Basically, think of a cube. The bottom of the cube is on the floor, the top of the cube is the box fan blowing upwards, and the sides of the cube are four different filters. You can get really fancy if you’re a computer hobbyist and you have access to a bunch of computer fans, the kind of thing that would cool off a desktop computer. Those are actually a little better than box fans in terms of quietness and efficiency because they’re more optimized and better engineered, but of course there’s a higher capital cost there. And with those, you can have like a grid of four or six computer fans blowing upward and going through maybe just two filters if you want a smaller profile.

So those DIY filter units typically perform a lot better even than the commercial units. A lot of these commercial units, they’re very quiet, but that’s because they’re hardly doing anything at all. I go, but they advertise, oh, they’ve got this nice HEPA filter, but they’re hardly moving any air through it. What you look for is a high clean air delivery rate. And that’s what you can get yourself by just using a good fan. Obviously, noise starts to be an issue in some settings, but… It’s that that’s what you do for safety. Germicidal UV really isn’t quite there yet for individuals to start to buy. We’re hoping that in the next few years it will be, but there’s more research that’s needed on safety. And you don’t want to use GUV in an area where there’s not good filtration because the ultraviolet light, basically, when it hits any kind of surface, it starts to generate indoor smog. There’s a lot of other weird interactions, like terpenes are especially nasty. And so that, GUV isn’t quite there yet. Basically, personally, do it as much filters as you can and try to get that in other areas.

Fin

I should try saying that back. So we’re talking about two methods which are filtration on one hand and GUV on the other. On the filtration, I’m imagining that you’re passing air with a fan through some big filter, where the filter is some foam material with lots of holes which stop particles getting through. Is that right?

Richard Bruns

Mostly, although with new filters there’s a bit of an electrostatic effect where the fiber is a bit charged electrostatically and so the particles in the air are attracted to them. And that stops working after a couple months, so in addition to the filter getting clogged up the electrostatic effect fades away.

But anything other than filtration is just bad news. Don’t do it. Like, there’s gonna be a bunch of crap out there. Some of it… like, there’s a bunch of different things. But anything that freshens the air by, like, putting any odors or chemicals or trying to blast it with ozone, like, no, that’s all bad news. Like, you’re basically polluting the air more. It won’t do anything about the particulate matter. There might be some effect on infections, but it’s usually outweighed by the smog. So it’s all about filtration. Okay.

Fin

Air ionizers feel like a thing I’ve come across?

Richard Bruns

No. There’s a few of [these things] that cause more good than harm, but the amount they’re doing is a lot less value for dollar than just running air through a filter.

Fin

And with filtration how quickly do you run up on diminishing returns with more filters?

Richard Bruns

So you don’t run up against diminishing returns in terms of air changes per hour. That will increase exponentially. The model that we use for predicting infection rate, the Wells-Riley model, which has some problems, but it’s mostly okay. Basically, every time you double the air changes per hour, you’re cutting your infection rate in half. So in an absolute term of your infection risk, you basically got a decaying exponential kind of curve. So as you’re moving along that, you have less absolute infection risk per each one that you spend. but the time at which it starts to not be economical for most settings. It’s like 10 to 15 air changes per hour are still sensible to invest more in. So really the limit is gonna be based on the noise of the units. Basically add as many units as you can, quiet units until they make so much noise that the use of the space is compromised.

Fin

Got it, and I’m just imagining, let’s say a bedroom or a living room, and maybe it has… a little portable HEPA filter in the corner. Does a ballpark number come to mind for what would be the, what was it, air changes per hour?

Richard Bruns

Air changes per hour, ACH. Yeah, those are, it’s not the best one, but it’s one that people are often familiar with. What you actually want to aim for is more of a liters per second per person kind of calculation. But, like 20 or 30 liters per second per person is, again, that should be like a bare minimum kind of situation. But in general, given, again, I recommend the do-it-yourself things rather than the commercial ones. But especially during respiratory virus season. Two or three in a bedroom would be reasonable, but it’s more important, of course, to put them in places where lots of people are gathering. So if it’s just your private room and nobody else is in there, yeah, you might have a little bit of particulate matter, but that’s not the real risk compared to getting a disease from someone else, either COVID or flu, which is going to be with us for a while. So it would be much better to have five or six air cleaners scattered around whenever you’re hosting people for dinner in your living room or dining room, et cetera, to really clean the places where people are densely gathered. So don’t think about it on a per room basis, think about it on a per person basis. And having one of those air filters for every, okay, this is just kind of, This is not official engineering guideline, but I don’t think this number is actually there. Engineers always make things really complicated, but as a rule of thumb, if you’re thinking about a basic thing that you buy off the shelf or a Corsi box, you basically want one of those for every two to four people that you’ve got in the gathering.

Fin

Okay, nice. Yeah, I guess a heuristic here is that you can just keep halving the infection rate by doubling the amount of filtration in a room. And in that sense, it’s not like, let’s say a fire alarm where once you have one fire alarm, you’ve checked off the fire alarm box, right? Yeah,

Richard Bruns

Exactly. It’s definitely a, you want to gears level, more power kind of situation. More filtration is better. Most places have extremely low amounts. You should be thinking in terms of five to 10 air changes per hour, 15 to 30 liters per second per person.

Germicidal (UVC) light

Fin

Okay, great. And then you also mentioned germicidal light. I guess many people have heard of this. This is light which sterilizes the air. Am I right in thinking? This helps with pathogens, so it kills germs, but doesn’t necessarily help with particulate matter.

Richard Bruns

Exactly. They actually make the particulate matter situation worse. So it’s the kind of thing that you only want to pull out in an extreme pandemic scenario. Like basic respiratory virus season, as long as you don’t have terpenes or other chemical sources in the room, they’re going to be doing more good than harm, especially for a mass gathering kind of event. But it’s. It has not gotten to the point where there’s actually a reasonable consumer product that is being sold that we can recommend. There’s a lot of things that you might find on Amazon, like the blue lights and other stuff, but people that we trust say, like, eh, don’t really go there yet. Okay. So it’s really not, there’s not a consumer product. Like there’s some, there’s things that you can rent if you’re doing a thing and if you have the right connections, but there’s not a lot of things that you can rent. Unfortunately, there’s not yet any consumer product that we can actually recommend for GUV. We hope that’ll change in a few years.

Fin

Ok, and just to be clear, I take it we are implicitly talking about Far-UVC, right? Which is a particular narrow bit of the UV spectrum? Yeah, UV,

Richard Bruns

222 is the number that’s usually thrown around. There’s Upper Room UV, which is, you can’t let that shine on someone individually, but it’s quite useful. So, if you are doing an HVAC renovation, There is existing products that blast the air going through your ducts with UV light. That is a safe, established technology. So it’s… Yeah, that is probably something that I would recommend, especially if you’re gonna be having a lot of gatherings in your house. You can have the in-duct UV, which is, you know, just in the box, it sterilizes the air going through. That’s, yeah, that has a good ratio of benefits to costs. There’s stands and stuff that you can have on the walls that are just mounted upwards. Most of our target audience would probably be able to follow the advice of, you know, Don’t let it shine in anybody’s eyes. Don’t have a reflective surface on the ceiling that can bounce into someone’s eyes. So the upper room UV, that’s the 240 or something or 50 something, I forget the number. That’s also a relatively established technology. And if you’re, that’s something that you can do if you have a mass gathering, shine that up. But the super safe, like,

Fin

but appears not to have these damaging properties as well. So you could, once it’s cheap and doesn’t produce this smog quite as much, you could just shine it on people, put them in rooms. Currently, however, very expensive, not widely available. And it has this smog issue, which you mentioned. Do you want to say more about that?

Richard Bruns

Yeah. When you, When you shine the UV onto anything, there’s what’s called secondary chemistry in the room of what does the, whenever UV light hits something, I think of yellowing or other effects or sunburn, any substance will change when hit by UV light that will often release other chemicals. That’s relatively understudied. I know, like, I’m not even sure what products would have terpenes in them. I just know, like, that’s one of the things that causes a lot of smog and problems. So I couldn’t tell you, like, what’s a particular brand name or plastic or thing because the research is still very young. But we would need to know basically what’s the… What things you should not have in the room when you’re shining this Far-UV light, and we don’t have that list yet.

Fin

Okay, so on Far-UVC, it sounds like the priority there is just getting it cheaper, which sounds like doing more R&D, is that right?

Richard Bruns

Yeah, they’re basically, within the next few years, they should basically have LEDs that produce them in the right wavelength. Right now they don’t, or they’re very expensive, but that’s ongoing.

Fin

Okay, and then it sounds like… We also need to just know more. So that sounds like doing studies to rule out potential harms to humans, and also figuring out the effectiveness. So doing trials at increasing scales. Is that right?

Richard Bruns

There’ve been a lot of rooms like EPA and NIST have big test chambers where they’re set up like a room and they can do all kinds of different things. So it’s been proven pretty conclusively that UV lights will actually sterilize a lot of area. So we know, like to within a reasonable amount of confidence interval, how much good you’re doing. And in those test chambers, we have to a much larger confidence interval, how much smog is being generated. And it’s the kind of thing where if you have decent ventilation and you have either vulnerable populations or a lot of virus risk, then it definitely is going to do more good than harm once the technology gets rolled out. but we need to have more practical advice on what are the interactions. Because you’re looking at one or two orders of magnitude difference in smog generation based on what other chemicals are in the room when you’re shining this far UV light. So we need to have a better study of what are the things that you don’t want to mix with it and have a simple list of make sure that you don’t have these materials or these fabrics or these chemicals or other things in the room when you’re shining the light. And we don’t have that advice ready to go, which means the smog generation changes by a couple orders of magnitude, and at the high end cancels out the health benefits.

Fin

Got it. Yeah, maybe one question here is if you just had a bunch of money to spend on trials and studies, what would you prioritize?

Richard Bruns

The interaction effects. The most important thing is to know the second, the technical term is the secondary indoor chemistry, the chemistry of what happens when different materials are blasted by the thing and how that affects smog and health generation.

Ventilation vs filtration

Fin

Great. So we have talked so far about. Two ways to improve indoor air quality, they are filtration and UV light, especially Far-UVC light. Filtration we can do right now, Far-UVC has a way to go on the getting it cheaper side, but it could be close. Is there anything else? One thing that occurs to me is ventilation. Like if it’s nice out, I could just crack open a window or use some ventilation system might improve the air indoors as well. Is that comparably effective to those other methods?

Richard Bruns

Well, ventilation without filtration doesn’t do a whole lot. Okay. You basically need to be passing the air over [filters]. Indoor and outdoor have different sources and different sinks of different kinds of pollution. Unless you are doing a lot of burning of candles or indoor flame sources, or you’re cooking something that burns. Indoor is not really a source of the fine particulate matter. Like, yeah, if you burn something on the stove, your PM’s gonna spike up. But most of the time, the smog sources are actually higher outdoors than they are indoors. And that’s especially true if we got wildfires like we saw recently, or if you live in a city or near a big highway. The stuff outside is generating the particulate matter in the smog, and your house is actually a sink. It sucks into the carpet, it sucks into the wallpaper so that it goes down. So if you open a window without filtration, and the outside levels of smog are high, you’re actually letting in more bad stuff. And you’re letting out some bad stuff. There’s some stuff in your house that is a source of some foul-smelling things, some like chemicals. But those are the stuff that doesn’t kill you as much. So yeah, I like opening my window if it’s a clean air day. Like if there’s no kind of ozone or smog alert, sure. You’re clearing out more air, it feels nicer. The stuffiness inside your house doesn’t go away. You get to hear the birds singing, et cetera. But if there’s any kind of ozone or particulate warning outside, you need to be closing it up and running your HVAC through filters.

Policy changes (state and national level)

Fin

Uh-huh. All right, so. There are things that individuals can do right now to improve their air. So, you know, I could go out and buy some more of these Corsi-Rosenthal boxes, for instance. But what about on the national level? So, if we took a country like the US, for instance, what are the measures that seem most promising here? Is it new regulations, new standards or something? And yeah, what are people working on?

Richard Bruns

We’ve actually been doing this for about a year, and I can tell you what we’ve been doing. So, at a very high level, when you’re thinking about how do I change a country, you think about the legal and cultural and policy scenario. So, I’m going to give you a little bit of a background. In the United States, we have a federal system, and building codes are very much a local issue. Especially given the current political environment, there’s not much that’s going to happen at the federal level that will actually have teeth. There’s a couple of things that might happen, like the Airborne Act that Representative Beyer has introduced will give some tax credits for doing stuff. So that’s always possible and we’re hopeful. But the real action has to be at the state and local level in terms of building codes. So for that reason, at the Center for Health Security, we basically have produced a model state law that we got a committee of like some of the 20 top names in the indoor air community. We hired some lawyers who are skilled at drafting model state legislation, a guy who worked in the Model State Emergency Health Powers Act. And we’re basically saying, What is a law that states could pass that would actually be a long-term, serious solution to this problem that would… exist well in the current legal and regulatory environment in terms of being constitutional, being legal, et cetera.

So we basically put a lot of work into producing this model law that’s like, okay, this is what we think would produce good outcomes. And it’s important to think about the difference between a law and a regulation. The regulation is the thing that says, here’s how many liters per second per person you have to have, here’s what the filters have to look like, here’s the technical details of implementation. Our model state law doesn’t give those numbers. It’s basically the enabling statute that says the state is going to set up a committee of people who will then write the regulations. And then there’s this legal structure of enforcement such that if anybody complains, that any resident of a building complains that the air is making them sick, then that triggers an inspection. And if the building is found to not meet the standards, then you have this legal structure of, okay, you’re now actually ordered to make improvements so people aren’t getting sick anymore. Okay, got it. And then the work begins of, all right, we actually have to find a state to lobby them to pass this thing or some version of it and make it pass through the political process.

Fin

Okay, that sounds exciting. Yeah, I mean, I’m pretty naive about just how building regulations work in general, let alone in the US specifically. So presumably there’s a bunch of existing regulations and laws state by state. And presumably they say things about “you can’t use certain toxic chemicals” or whatever. How are those standards maintained?

Richard Bruns

As one of my colleagues said, to call it a patchwork would be an insult to patchworks. There’s a lot of laws about particular contaminants. Like a lot of state laws will talk about radon in the building code, like you have to have these particular radon mitigation measures. And some will have laws about schools or government buildings, but there’s never any kind of comprehensive law that would protect indoor air quality for all public buildings. And it’s all based on local laws and politics. And a constant issue that you have, just about any political system ever, is a law is set up, but unless there’s some constant enforcement mechanism, it just kind of gets fading away. So we have this situation where there’s a lot of laws on new buildings, and it gets more and more strict all the time, sometimes much more strict than I think a cost-benefit analysis would say. Like in D.C. now new buildings have to be net zero, just meet extremely high standards. But trying to regulate via approval of new buildings can backfire because you’re making new housing construction so expensive that people are getting stuck in the old buildings. I And aside from major renovations, there’s basically nothing in terms of how are we regulating the air quality or the contents of older buildings.

Fin

Okay. Um, but okay, maybe we could just take water quality as an example. Let’s say I’m a tenant and I notice the water quality is falls short of one of these regulations, like What do I do? What’s the process from that point to the water actually gets better and these regulations do something, you know?

Richard Bruns

I don’t actually know the details of that. I mean, most of my knowledge of this is like, OK, what happened in Flint? And like, basically nothing happened until it became a national issue. I probably should know those details, but sorry, I don’t.

Fin

Presumably there is an understanding that when you are building a new building or installing a new filtration system or some new plumbing. that such and such laws exist and you have some reason to stick to them.

Richard Bruns

Yeah, there’s definitely regulations on what pipe material you use in your building. But most of the water quality issues don’t come from the last little bit of the building pipes, I don’t think. I’m not an expert on this. I think most water quality issues come from the actual water distribution, the river, the pipes, etc. And if If something is happening, if pipes in your building are causing the problem, first of all, you have to test, like test the water coming into the building, test the water coming out of the tap, to actually prove that the pipes in the building are causing the contamination. And then it would be a building inspector kind of issue. But the rule generally applies to when the building is doing construction or major renovation, they say, are you using the proper material in your pipes? Are you following the building codes? And if so, If something is built according to a modern building code using modern materials, it’s probably going to be very good for a very long time, at least in the water system. The problem with the HVAC system is it’s a lot easier for those to get messed up and their performance doesn’t match the building code, how it was designed.

Fin

I guess in some sense, air quality is much less centralized, like you’re not hooked into some central air supply. Each building needs to meet all the standards individually.

Richard Bruns

Exactly. Each building has to do its own filtration. You’re not getting filtered air from a big pipe.

Fin

On tap. Yeah. If only. Can you say more about the model state law? What you’re doing? Yeah, maybe. I realize it’s not released yet, but if you could say some things about what it says, that’d be interesting.

Richard Bruns

Yeah, yeah. I could definitely give an overview of what it does. So, first of all, It’s basically telling the state to set up an advisory panel and assign some state agency to be in charge of this thing. Say, all right, specifically delegating legal responsibility so the correct state agency has regulatory authority and they set up an advisory council of building owners, industrial hygienists, researchers, et cetera, who will agree on the standards. And we don’t say this in the law, but we definitely would encourage them to basically follow existing standards like ASHRAE 62.1 or 241. And a lot of people have concerns about the Model State Act because they’re saying, you’re not instituting a national standard. Every state could do their own thing. Like, how do you know they’re going to do something reasonable? And the answer is, we don’t. That’s just what it’s like living in a federal system. There’s nothing stopping a state or locality from having a completely bizarre building code, but we find that they’re generally mostly well-aligned. There’s all these little quirks in different areas, and that’s just kind of inescapable. But we’re pretty sure that… Like we’re actually going to issue recommendations for what kind of regulation we would put out.

But it definitely has to be tailored to local state needs. Like each state is gonna have it be in a different climate zone. Some of them will have like very different installed bases of how often you have gas stoves versus electric kind of stuff. So we do have to have built into it, yeah, the state’s gonna do their own thing. So the state has this process of setting up their regulations, doing public advocacy and outreach, and that generates relatively small costs. I mean, it’s not that expensive to hire a few FTEs to be in charge of things. The two main things that the state does that are the main requirements on building owners is, number one, build a safe environment. You have to monitor at certain frequencies. It doesn’t have to be constant real-time monitoring, but that tech is getting pretty cheap, so we would hope they would do it. You basically have to monitor the levels of pollutants and contaminants in your building. You have to post them on site and also upload them to a state database. So the first thing we’re doing is you have to measure and record what’s going on. And we don’t tell them exactly what to measure in the law, but the regulation would say, okay, you’re measuring particulate matter, you’re probably gonna measure CO2, et cetera. Just let people know what’s going on, because the problem, aside from thermal comfort or humidity, air quality at most levels is completely invisible to people. You don’t know what you’re breathing unless it gets super bad. And just letting people know what they’re getting, a consumer choice, is the first thing.

And the second one, which you could do either a couple years later, or it could be optional, although we really encourage people to do it, is to basically set up an OSHA-like complaint system. We’re not requiring annual inspections. We’re not requiring… Some state laws, like Maryland, did something that would have required an annual building envelope inspection, and that’s just, that’s very expensive for not as much value. So we’re not requiring annual inspections. It’s optional. If you get annual inspections and you pass them, the state will look more favorably if a complaint happens. But that’s purely voluntary. The mandatory thing is if someone, if a resident or occupant of your building, which is generally going to be people who work there. If someone who works in the building issues a complaint, goes to the state agency and issues a complaint about the air quality, and that’s the thing that the state, the act requires states to set up this complaint process. So it has to be, let people know that they can complain about this thing. Set up a website where people can enter a form and say, hey, the air where I work is crap and I think it made me sick. Here’s my doctor’s note, et cetera. The state will then follow up on these complaints and will then do an investigation based on those complaints. And if the investigation finds that, yeah, the air in here is crap and it’s plausible that it caused this person to get sick, then you order remedial action.

Fin

So, okay, this model law calls for monitoring in some way. I’m imagining that means that buildings would have monitoring equipment, and then the data from them would go to a database. Would that be public?

Richard Bruns

Yes, it should be. It’s publicly… that is part of it. The state must make the data publicly available.

Fin

Okay, cool. I know that you also looked at an existing Senate bill, I think in Connecticut, which was for school air quality. Curious what that said and also what happened to it, did it go through?

Richard Bruns

Honestly, I don’t know. I haven’t heard anything. Like a lot of these things don’t make it because of there’s some kind of opposition or it’s hard to get things done. Connecticut is in that situation where in the past, they hadn’t thought they needed air conditioners in their schools. So I come from North Carolina. I was quite old before I realized that anybody ever went to a school that didn’t have an air conditioner. I’m like, how does that even work? Our entire school calendar in North Carolina is shifted forward. We go to school starting in early August, and the kids don’t mind it because everything outside is so miserable and so wretched, you can’t do anything. So we are perfectly happy to go into an air-conditioned school system for the month of August and just… Okay, that’s fine. It works great. And then we get out of school a lot earlier.

So, like the fact that a school wouldn’t have air conditioning just boggles me. It’s like if poor countries in rural North Carolina and Alabama can have decent air conditioning systems in their schools, like why can’t richer states? But somehow, like they just never had it in the past. People are always so attached to their existing things. Like, we never had air conditioning in the past, and it seems like a horrible imposition to tell us to order it, and that just makes no sense to me, whatever. But definitely with global warming, Connecticut is now in a situation where if you could get away without school AC in the past, you definitely can’t now. There’s heat waves, it’s screwing with kids, it’s messing up their test scores. So they’re saying, all right, let’s just get away with it. You got to have decent HVAC systems in here, and you have to actually shut down school if it’s too hot, if you don’t have the HVAC system. So the effect on test scores, the effect on health, you know, there’s huge benefits. to putting proper HVAC systems in schools, but the problem is the cost is all up front and the cost is a money cost. And the benefits are, you know, people just aren’t getting sick 20 years in the future. So it’s often hard politically to convince people to pay upfront money costs for future life and health gains.

Fin

Yeah, okay. That sounds like a fairly sensible piece of legislation. For what it’s worth, I think the state of Simmer in the UK, certainly. most slash nearly all homes just have no AC. So the country will just like plunge into a state of overheating for like a week or two every year. But there’s this status quo bias right where people just don’t yeah don’t read much into it.

Standards bodies

[Anyway, ] you’re talking about the State Model Act for Indoor Equality, and then whilst you were talking about it, you mentioned a standard, ASHRAE standard, which the state could follow. So ASHRAE, this is like a kind of trade association of like air people?

Richard Bruns

Yes, it stands for, well, it used to stand for the American Society of Heating, Refrigeration and Air Conditioning Engineers. Now it just means ASHRAE because they’re going international and they’re trying to expand beyond America. But yeah, it’s the Trade Association for HVAC Engineers. They are the standards body along with ANSI that produces like, there’s hundreds of them of very technical things.

And the new standard, like, so 62.1 is already incorporated by reference into a lot of state building codes. So in most places, any new building has to satisfy or comply with 62.1. But of course, existing building stock doesn’t get, like there’s no requirement in state laws that say older buildings have to comply with it.

Fin

Yep, okay.

Richard Bruns

Right, so ASHRAE recently passed a new standard. Standard 241 for control of infectious aerosols. They were basically requested to do that by the Biden administration. They got it done very quickly in like only six months from start to finish, which is super fast for a standard. And it was me, I was the only economist, the only public health, one of the few public health people in a room with a bunch of engineers. So there was a lot of figuring out how to work with each other, a bit of culture stuff, but we’re very, we got things working well. We started collaborating on this giant spreadsheet that was using the Wells-Riley model. to do a proper Monte Carlo probabilistic simulation of the risk of various spaces based on increases in airflow rates. So I was on the risk assessment team, which is basically deciding what the airflow rates are going to be in the infection risk mitigation mode. So what ASHRAE Standard 241 does is saying, at certain times, you must have the capacity to run your building in infection risk mitigation mode. And we’ve left it kind of vague as to when exactly it gets turned on. I, and I think other people on the committee, want the infection risk mitigation mode to turn on every time there’s the winter respiratory virus season, what we used to call flu season, from December to March in the northern hemisphere. So whenever there’s an enhanced risk of respiratory virus transmission, you are running at these higher levels, and it doesn’t have to be outside air. It just has to be filtered air delivered at a higher rate to reduce the risk of infection. And we set the standards for that. Nobody has adopted it yet, but we’re hoping that people will, and we would definitely hope that if a state adopts the Model State Act for Indoor Air Quality, it will be a model state. they would start to use ASHRAE Standard 241 as part of that and say, here’s the procedure for having higher filtration rates when people are at higher risk of viruses to protect them.

Fin

Okay, great. And I’ll just try saying some of that back to make sure I’m getting clear on how these things fit together. So, state legislation sits at a higher level than standards in the sense that it’s presumably not mentioning a bunch of numbers about airflow rates and air changes or whatever. It’s just saying let’s set up a body and here’s how all these different agencies and bodies are going to fit together. Standards do specify Concrete requirements. Like, they’re really, you can just plug them in and they’ll tell you what to do.

Richard Bruns

Okay, a bit more detail. So, you’re right, you have the legislation, which is the thing that’s passed by the state legislature. And then you have regulation. The regulation has the force of law, and that’s done by the regulatory body. A standard is privately produced. So a standard is only voluntary. It only becomes an industry standard. However, regulations often choose to incorporate standards by reference. So the legislation is the law, the regulation is done by bureaucrats with force of law, and often the regulation adopts a standard which is often done by a private entity. Okay,

Fin

Great, thanks. And in this case, so ASHRAE is the body that’s coming up with this standard, 2-4-1. ASHRAE is like a trade association, it’s not a government body. Yes. But the standards it produces are being referenced in regulations backed by law. Yes. That’s how it has teeth. Yes. Or at least one way it has teeth. Okay, great. And then the hope is that that happens with 241, so now you have this kind of detailed standard which has been produced by people thinking hard about what actually needs to be in the standard. that’s a valuable off-the-shelf product, for want of a better word, that you can now be referenced in a bunch of different legislations. That’s the idea.

Richard Bruns

Exactly. The entire… thrust of my work for the past year or more at the Center for Health and Security has been to help produce these off-the-shelf solutions for political change. So we can now, we can find our allies and go to a state legislature and say, look, here’s the model law. If you want to do something about indoor air quality and you have the political coalition for it, this is the best thing to do. Slight tweaks for your state, use this model law, plug that in, pass that, and then That will tell regulators what to do. And we’ve got this off-the-shelf solution from ASHRAE that says, this is a good standard that you can adopt, and this is what you should encourage people to do.

Fin

Okay, great. And then you said something about this, but it sounds like getting state legislation is more promising than… like federal stuff or getting like countrywide regulatory bodies to implement standards? Like maybe the EPA could do this. I don’t know. How does that work?

Richard Bruns

Well, currently the EPA has no authority over indoor air. I mean, and I think that’s baked into the statute. It would require a literal act of Congress for EPA to have authority over indoor air. And in the US federal system, building codes have always been a state and local kind of matter. Basically by, in theory, anything that doesn’t… involve interstate commerce should only be done at the state and local level. And of course, the federal government has various reasons to claim jurisdiction on things that are kind of in this gray fuzzy area of maybe it’s interstate commerce or not, but it would be a stretch to claim that the building code for a particular building actually is under interstate commerce or is actually constitutionally allowed for the federal government to do anyway. How many… You could find ways of doing it if the entire country was behind you, but in the current political environment, it may not survive a court challenge.

Fin

Makes sense.

Richard Bruns

So, it basically has to be done at the local level, because in our federal system, they’re the ones with the legal authority to do this kind of building. Yep,

Fin

Got it. So, I might get some details wrong here, but I understand that the Center for Health Security hosted a meeting last year, and this is on the National Strategy for Improving Indoor Air Quality. Yeah, do you wanna say what that was and if anything came out of it?

Richard Bruns

Okay, so basically it was getting together a lot of experts in the field, Essentially, it was a kind of a cross between a research conference and a planning thing. We didn’t actually have poster presentations, but there were a lot of chat with people, talk about their work, get to know people. And it was really kind of setting the stage for producing the National Advisory Council and deciding, talking with people as a group and saying, okay, here’s what we know, here’s what we’ve been studying, here’s what we figured out. We saw presentations from a lot of different people. and deciding, all right, now that we’ve shared this knowledge and information, what’s the next step forward? And that was when we kind of decided, yeah, yeah, we can do this model law project. A lot of people who attended the conference then became the technical advisors for the model law. So you can think of it as the kickoff meeting or academic conference that got this model law thing going on.

Pandemic risk reduction from improving IAQ

Fin

Great, got it. All right, nice. So now I’m imagining… Like, just suppose this was successful. So, the model law turns into actual law in a bunch of states. This translates into, like, hopefully health outcomes, just fewer people dying from particulates, also from pathogens, like fewer people are getting the flu, getting pneumonia, getting endemic COVID. One thing we’ve talked about less, I noticed, is pandemics. Yeah, curious what you can say about like, if let’s just say in a big country like the US, it really gets its act together on indoor air. What does that do to the risk of a pandemic emerging from that country?

Richard Bruns

It does lower it pretty substantially, but it’s very much a probabilistic thing. So with an endemic thing, like if you could, with reasonable improvements to indoor air quality, reduce the annual risk by something like 25%. So we’re hopeful that if everybody did this, then like the number of people getting the flu or COVID each year would reduce by something 10 to 40%. Again, the confidence intervals are pretty big here, but that’s the kind of numbers we’re looking at. So it’s with a new pandemic, You basically have something starting off as being an outbreak. And will it completely blow up beyond our ability to contain? Or will it be something like the original SARS where we could actually control it? So if the reproduction rate is extremely high, then the air quality improvements that we’re talking about really aren’t going to change that much. It could slow it down a little bit. Maybe a super spreader event doesn’t have as many people. Where conversely, if the reproduction rate was very close to one, then this thing would stop it. But basically, it’s hard to know exactly how to model it, but to a first approximation, you could definitely do worse than saying it could reduce the chances of a respiratory virus pandemic by about that 25%. Okay,

Fin

Okay, awesome. And yeah, I guess it sounds like that’s just quite sensitive. Like, you need to know the distribution of, like, pandemic potential pathogens in the future. One thing I’m getting from that is that if you’re, like, especially pessimistic about engineered pathogens, engineered for like $R_t$ to be especially high; it’s going to be less effective than this sweet spot around $R_t=1$. So maybe it’s more effective for natural pandemics.

Richard Bruns

for a If most people who catch COVID, I think, actually catch it from the super spreader events, it’s a pretty big deal. So even with a high R-naught, it’s still… the super spreader events that really drive a lot of it. So you’re flattening, even if you don’t stop a pandemic, you’re still flattening the curve a substantial amount. So good indoor air quality, and again, things are different with different diseases, but the kind of indoor air quality improvements that we’re talking about can flatten the curve about as much as masking. Masking in a real world situation where a lot of people aren’t doing it properly, et cetera. Like, yeah, if things are done exactly right. masking is more cost-effective and more effective overall at flattening the curve than indoor air quality improvements. But in the real world, especially among a population that isn’t 100% cooperative, indoor air quality is at least as good, possibly better than a mask mandate at slowing down the spread of things or even stopping them from spreading. So if you think that masking is a good idea for any situation, indoor air quality improvements are an even better idea. because they’re an engineering control that doesn’t require everybody to comply and constantly be doing things. So, it’s definitely valuable in a lot of situations.

A bit of a background in terms of the intentional attack. When I was at the Food and Drug Administration, I was on the team that did the intentional adulteration of food rule. So I actually have some experience with this kind of bio-defense and thinking through adversarial attacks. Very briefly, the Food Safety Modernization Act was passed in 2011. Congress ordered FDA to pass a bunch of new rules. And one of them was pass a rule telling food production lines how to harden their production facilities against being used as a vector for a terror attack. So, in every defense model is… Like making things difficult for the attacker has advantages. […] No defense is 100% secure, but the more you stack layers of defense, just the harder it is to get past all of them. So it is a much, well, it is a harder problem for someone to design a bioterror attack against a country that has good IAQ than less good IAQ. You have to do extra work to make sure it happens. And any extra barrier or difficulty that you add for your attacker has real gains when you’re thinking in this kind of adversarial situation. Yeah,

Fin

Got it. And so, I mean, taking masking, for instance, and improving IAQ — these are not like two alternatives, presumably the benefits more or less compound? You just have like two layers rather than one.They’re not entirely uncorrelated, but…

Richard Bruns

Exactly, especially if there’s a new pandemic or bioterror attack you want them both. One of the other things I’m doing at the center is next gen respirators; we’re trying to convince hospitals to get PAPR type things. Very briefly, you can right now buy yourself, like as a consumer right now, for about 1,400 bucks, buy a really cool welding helmet that’s like, it’s welding protective equipment, like, I mean, it’s not gonna be as good as a bike helmet, but it’s that same kind of thing. It’s like, you basically get your kind of Darth Vader helmet that will protect you against shock and impact and has a welding visor, and it has integrated into it a full popper that’s… like this same kind of protection against viruses that the, you know, medical people does giant hoods have. A PAPR stands for Powered Air Purifying Respirator. It’s there’s the filter inside the helmet and it’s got battery packs and fans and it just sucks the air in through the filters and gives it in to you. And we’re basically trying to convince hospitals that they should spend as much money to protect their workers as welding shops do. Like again, it’s back to the status quo bias or a cultural thing. Every Bubba Redneck welding shop in the country knows. that you’re just gonna spend a couple thousand bucks on proper PPE for every worker. That’s just what you do in that situation and you’re a hazard. But hospitals will not spend a trivial amount of money to buy similar protection for doctors and nurses and the people keeping us alive in a pandemic. It’s ridiculous.

Fin

Yeah, that’s kind of wild.

Richard Bruns

You can buy PAPRs, hospitals should buy PAPRs. They work, they’re cheap and getting cheaper all the time.

Fin

Yeah. And speaking of, like, especially bad pandemics… It strikes me that having a stockpile of very good PPE just is going to be extremely useful. You can just keep a country running with essential workers wearing these things. Yeah,

Richard Bruns

that’s one of the things that we’re trying to get in the strategic national stockpile. Of course, there’s other competing priorities, and they’re definitely in a better shape than they were pre-COVID. But given the other things that could be researched, like I think they wind up not doing PAPRs, but when you consider the alternate uses of their money, like stepping up platform technologies that allow you to rapidly roll out a vaccine for a new kind of thing, I’m actually sympathetic to be like, yeah, with new biomedical innovation, there’s actually better value for the money than stockpiling these things, because almost anybody can stockpile PAPRs. You don’t need the government to make that happen, but encouraging new kinds of biomedical innovation could be even more protective.

Fin

Can you imagine it’s getting much cheaper if people are trying to manufacture 10 times as many for a country because suddenly it’s not just welders but it’s essential workers in hospitals and stuff?

Richard Bruns

Yeah, I’m not an expert on this, but we think that getting it down to $400 or $500 is reasonable and achievable, even in the short run. If production scaled up even more, you might get it down to a bit cheaper than that, but…

How useful is the FDA?

Fin

Probably among the Twitter accounts I follow, that the FDA tends to over-regulate and therefore cause these kinds of deaths by omission.

Richard Bruns

Yes, I’m very familiar with the invisible graveyard argument.

Fin

Right, right. So that kind of meme, that the FDA almost always over-regulates, is that roughly correct? Because I guess we’ve been talking about a sense in which there’s a hole for useful regulation around air quality, so is there a more nuanced story about what the FDA does right or wrong?

Richard Bruns

Sure, I can guide you through several layers of the galaxy brain meme. So if you are not exposed to this EA [invisible graveyard] thing, the thing you should know is that the government regulating things too much does have costs, it harms innovation. And in general, there’s a constant dynamic everywhere in the world where if something goes bad, people call for the government to regulate it, to make sure nothing like this ever happens again. And that process of regulation just makes it harder to do new things that help people out. We see this in housing. We see it in a lot of situations. In general, there’s powerful public forces that encourage people in all kinds of contexts go too much into precautionary principle, not enough consideration of the costs of regulating things. That’s just a standard thing we see everywhere. Now, once you have that understanding, the next thing you should ask about is how tractable is this problem? And that’s where international comparisons are extremely useful. Like we know that the US sucks at zoning laws because we have the examples of Japan and Singapore who are just doing obviously better. Like, if the US just adopted zoning laws like Japan, like that would be a huge win. We can just point to say, yes, we know our country is screwing up. Look at this country. They do better.

With something like the FDA, there’s no example of any country anywhere in the world that does the FDA’s job obviously better than they do. If anything, the Europeans tend to be even more precautionary and even more hesitant to recommend new things. Even other places like Japan, Singapore, the usual suspects. they basically have a system that’s pretty much the same as FDA. So even if you think that there is something wrong with FDA that needs to be fixed, that tells you this is going to be a much harder problem. If you cannot point to any developed economy anywhere in the world that is doing obviously better, then first of all, you should be a lot more skeptical that something bad is actually happening, and you should be skeptical that you can fix it. Now. I definitely, there are tweaks that I would like to make. I would like to encourage more things to be done using cost-benefit analysis. I mean, you know, that’s kind of, you know, my personal hobby-horse, and most people in EA would agree. Like the current FDA approval system, it isn’t based on an actual monetize the benefits, monetize the side effects and risks and come up with a, “well, what’s the cost-benefit ratio of this new medicine?” It’s all just this. fuzzy, “do we feel that this justifies that” in a very uncertain way? And I think that if you could get them to do a more rigorous cost-benefit calculation, that would be a huge improvement. And if you could have some kind of tiered approval, like Scott Alexander talked about in that blog post of, you know, assign one through five stars, I would actually recommend one through seven of, like, allow something to be legal after it’s passed phase two clinical trials, but just kind of have this staggered innovation so it’s not all or one. that could be good.

Fin

Yeah, so maybe there’s a couple clarifications. I feel like I did a bad job or at least a quick job of explaining this kind of stage one in the galaxy brain of, well I guess the invisible graveyard is the word I guess given. Yeah, curious how you’d explain this invisible graveyard idea in your own words.

Richard Bruns

Okay, so everyone is probably aware of new medicines that can do a lot of good for people. Again, without endorsing any one particular thing. It seems like some of these new weight loss drugs, for example, were seeing very good results in clinical trials. We’re reducing weight, we’re reducing the risk of heart attacks and strokes. The invisible graveyard argument is, think of all of these medicines that are doing a huge amount of good for people. If the FDA was faster at approving the good ones, then you would be fine. They would be doing good years ahead of time. So you would have several years of these things saving people’s lives. So, there’s the invisible graveyard is all of the people who are killed because the delay in approving the drug was such that they died before the FDA approved the drug.

Fin

And the idea is, it’s really not very salient. That’s, for a particular person who died for want, for lack of a drug which could have been approved earlier. It’s not so salient that in some sense the regulatory agency which delayed that drug is responsible, counterfactually responsible, for that death or that accelerated death. It’s an omission rather than a commission.

Richard Bruns

Well you pointed at something that’s very good. It’s like… The regulatory age, you’re blaming the FDA for essentially not being perfect. And the invisible graveyard, and I tried to be very specific when I said this, of if they had only approved the good drugs very fast, and there wasn’t, and holding everything else in the system the same. So here’s where I’m going to get to the part that is going to sound more like, it’s more of a defense of the current system, which is before you had… a real drugs that FDA approves, you actually have proper scientific knowledge and standards. Like the system works in such a way that it teaches doctors what can be used and what can’t be used. So when people talk about the invisible graveyard, they’re assuming that you can get all of the benefits of this current system if it runs much more quickly. And that’s kind of an untested assumption. Again, you can’t point to any other place in the world that is working like that. I do think that there are improvements, but I don’t consider it a certainty. Like, it’s definitely plausible to me that the only reason we get these drugs at all, well, we know the only reason you’re going to get a very expensive drug is if there’s really good evidence that it actually works.

Insurance companies, especially Medicare and Medicaid, and of course you’re familiar with the NHS, deciding whether or not to do something — you’re giong to need a big mountain of evidence before people are giong to spend money for it. And the FDA is part of the system by which we generate the proof that makes people willing to pay for it. And in that way the FDA is responsible for creating the entire system of innovation and new drugs. It kind of pattern matches, like a lot of people I think by now are familiar with the arguments against naive communism. Like naive communism is, well you’ve got all of this money, why don’t you just take it away from the rich people and give it to the poor people and that way the money will be spread equally and like everybody will be so happy. The obvious counter argument is if you… You only have this money because the system of free markets and democracy and rule of law is what generates the wealth. The system that generates the wealth generates it in an inequitable way, and you can certainly do the European social democracy to use some of it, but if you break the foundations of the entire system, you’re left with nothing. You’ve destroyed the system that’s producing all of the benefits. and it’s certainly plausible to me that if you made too many changes to the FDA system, it would be kind of like communism, and that the FDA is an essential part of the system that makes the drug development happen, because it’s the Giant Seal of Approval that will make people pay the giant amounts of money for the drugs that makes the innovation happen. So I would be very hesitant for massive changes to the system for exactly the same reason you should be very hesitant to support a communist revolution. Like you’re breaking the thing that actually delivers the goods because you think that it could be delivered perfectly, but where’s the proof for that?

EAs, most of whom are very young, don’t realize how novel the entire concept of cost-benefit analysis is. There are older examples of it, but almost all of the old examples really suck because they’re not actually measuring the benefits properly. You still see remnants of this in a lot of government agencies. This cost-benefit analysis done badly is basically just an accounting exercise in the service of an inhuman entity. It’s only looking at the profitability to this inhuman thing, be it a corporation or a government, without considering the actual human cost and benefits. So a lot of times in the past, they would only look at medical bills and lost productivity. There would be no valuation for people actually like being alive and healthy. and such in the bad old days, if something killed a retired person without generating any medical costs, then there would just be no cost aside of that, that they would only measure, like they would consider the value of someone’s life to only be their contribution to the economic engine. And it’s only been in the past 20 or 30 years since like the 80s and the 90s where the foundational research on actually properly valuing life and health, et cetera. And there’s really, EA’s kind of attached themselves to this, and oh yeah, that seems obviously correct, why are you not doing it? But the entire culture of cost-benefit analysis as a thing that one automatically does just really doesn’t exist outside economists and EA’s. It’s slowly starting to spread. And the only way to make it spread is to show that it’s actually a thing of respecting humanity more and that we’re not doing the bad old cost, inhuman cost-benefit analysis that reduces people’s lives to how much you contribute to the machine.

I don’t know what your patience or tolerance is for getting philosophical quickly, but I think if you’re doing any kind of good cost-benefit analysis, very quickly runs into philosophical questions of what do you value and why. And you mentioned, again, in the past people were running on deontology. The FDA does run on deontology and a bureaucracy run on deontology is an amazingly good social invention. Like, you know, it’s a good idea. Bureaucracies are really incredibly valuable machines for mass producing nice things in life. Like you can think of them as factories and bureaucrats of these lies, these machines that are like on a factory assembly line. And in the COVID, like… A lot of EAs have pointed out, you know, bureaucracies like the NIH or FDA do have obvious flaws. They’re not able to adapt very quickly to changing environments, but that’s because they’re never really meant to. A bureaucracy is a factory for churning out reliable, good enough judgments and rules and approvals. And this is an incredibly valuable social technology that we really didn’t even have until the last 100 years. Before then, it was all just individuals making judgments. I don’t know how familiar you are with the idea of why common law legal systems are better and historically function better. Very briefly, The reason that we like a common law legal system is precedent and predictability are extremely important. People need to be in a system where a lawyer can look at the facts of the case and say, OK, given these facts, if the judge is following precedent like they should, this is how I predict the case will happen. That way, you can go to arbitration and you don’t actually have to go to the court because if you have a court that makes predictable outcomes and everybody knows what the outcomes are. Then you can save yourself the expense of going to court and your lawyers hash it out and say, yeah, here’s what’s gonna happen. Here’s how you resolve the dispute. And you just coordinate better on it. And that relies on precedent and predictability, which means that the individual initiative of the court has to be taken away.

Like you can’t, and in pretty much every other system ever, a judge has been like, “you are a paragon of Confucian virtue. Well, you will do what is right in every case according to the rules of virtue.” Nobody has any clue what you’re going to do. It’s a recipe for corruption. Like, you’re going to do it. There’s no predictability, you can’t do commercial contracting, it just doesn’t work. So you gain a huge amount of value in society by constraining people and forcing them to follow precedent. And a bureaucracy kind of expands that out into a new area. Knowing what to do to make the FDA approve your drug is really valuable, because that way everybody knows this is what you have to do. And as long as there’s a reasonable correlation between what you have to do to get approval actually generate good scientific knowledge, it works really well. Yeah, great. So when you have an engine that’s specifically designed to be reliable and steady and predictable and do the same thing, it’s always like that’s not something that can or should change its mind quickly. And people who, again, they’re complaining about, oh, you should have this engine of predictability that also automatically does the right thing. No, the bureaucracy being slow and stupid is an inevitable side effect of… to bureaucracy being a predictable bedrock foundation upon which economic activity is built.

Fin

I guess if you’re building some software products, you want to build it out of parts which are made out of parts and so on. Yes. And all of those parts need to be predictable. Like if you have any element of like, well, it’s kind of, you know, kind of whimsical and fickle sometimes and it’ll, you know, just say something unpredictable. That’s just going to like percolate to the rest of the system. So if you’re trying to like build complex big things that work together and don’t stall, you might need to pay this cost of like sometimes having something which is slow or like overly rigid, but then you just realize that it’s like more or less necessary.

Richard Bruns

Exactly.

Do a culture’s stories matter?

Fin

I like this point. Yeah. So I just had some questions unrelated to what we’ve been talking about. I stumbled on this blog post you’d written and you mentioned that a bunch of, let’s say, original seeming thinkers, contemporary thinkers, seem to have been raised to like an unusual extent on old sci-fi books rather than Newer fiction, at least they’re kind of citing these books more. Curious what you think that is.

Richard Bruns

Okay, so this is… definitely a different, I really like the rationalist habit of putting epistemological status on the front of each different essay. So this epistemological status of this is nothing like what I’ve been talking about in the past. So pretty much, well, when I started talking about the FDA, that’s more of like how I’d fix things, but like most cost benefit analysis area, people would agree. And the indoor air quality stuff is like, just about any expert would agree. And yeah, this is going to get into a bit more. So other people in the EA community have said similar things, but people are socialized by narratives. And that’s an important part of how culture is created. And like throughout cultures, people are just going to copy cultural patterns. Like humans, I’m sure you’ve heard of, you’ve seen the experiments where orangutans are actually better than human children at solving logic puzzles. Like orangutans are better shape rotators than human children.

Fin

I know they have better short-term visual memory.

Richard Bruns

And that is why we rule the world and orangutans don’t, because humans deliberately followed an evolutionary strategy of being more like an ant and less like a monkey. Humans are machines that copy cultural and behavior norms. An anthropologist might say it differently, but that is the human killer app. We have culture and other animals don’t. So even if we’re actually stupider than animals Don’t have culture, we still out-compete them because our crystallized intelligence is vastly larger than their crystallized intelligence could ever be, even their fluid intelligence might be. Are you familiar with these terms? Yeah, it should be clear from context. So, like the orangutan may be smarter than you in terms of just figuring out things, but you’ve talked to an entire, like you’re the heir to human culture since Buddha and Aristotle and like the first person who realized that these berries will kill you if they eat them and like grunted warnings to the other monkeys to not eat the berries. Like there’s so much stuff that we have and most people. will spend almost all of their cognitive effort just absorbing the myths and stories and role models of their culture. And if you look throughout the history of humanity, the culture of engineering and innovation is extremely rare. Like most people don’t have them, it’s not a natural thing. The Industrial Revolution, the Enlightenment, European ideas of enlightenment, free speech, freedom of religion, industrial revolution. This is one very particular culturally driven meme space that didn’t exist throughout most of history. And there’s an extent to which in the past there was much more of a culture of exploring the frontier, tinkering in your garage, whereas today, It’s nowadays, and there’s often a culture of, well, just kind of memorize the teacher’s password and go along with things. So that’s kind of what I was pointing to. There’s a lot more resources for people available if they have the engineer innovator culture. But it’s plausible that mainstream society is less good than it was in the past at passing on the engineer culture, unless you’re reading fiction that’s specifically designed to do that.

Fin

Nice, that was a fantastic answer. Yeah, one detail I remember which just like, raised an eyebrow. I haven’t watched the Avenger movies, but you mentioned that, like, the story arc involves, you know, them going on some quest and defeating the guy. And then, like, restoring planet Earth or whatever to the state it was originally. But presumably if they have the powers to defeat these galaxy-spanning villains, then they could at least end malaria or something! I liked that point.

Richard Bruns

A lot of genres of anime where the characters are like “you know what, I’m just gonna go out and make myself more powerful for the hell of making myself more powerful because I want more money and power. Like anybody who does that in American fiction is the bad guy, but there’s popular genres of anime and other things where none of the good guys get to do that. Like the good guys should be allowed to say, yeah, I wanna be better. I just wanna improve things. And you don’t really see that as much. And like Lord of the Rings, I think, was one of the main cultural tipping points where we move away from the, I’m using Mark Twain as an example because it’s more famous, but I think it’s emblematic of Connecticut Yankee, the difference between Connecticut Yankee and King Arthur’s Court and Lord of the Rings is very huge and instructive. I think My rational everyone’s rationalfic reading list should start with Mark Twain’s Connecticut Yankee and King Arthur’s Court It’s the original like you get thrown back into a backwards environment You use modern science and technology to make it better now. It is true that The narrator is quite obviously kind of a greedy bastard, but he still makes things better despite being a greedy bastard. And that’s kind of part of the joke.

Fin

Yeah, I like this point that there are these features of… So having a drive to self-improvement and generally being ambitious about the effects you’re having on the world. And then also having those ambitions be in some sense not narrowly constrained to your people or whatever. And these are like marks of a villain, right? At least in contemporary popular fiction. I think this point that these properties, none of them are necessarily bad. And if they’re pointed in a direction of doing good things, then good to just the same extent. And it’s a shame in some sense that the bad guys currently have a monopoly on these properties, again, at least in the kind of popular imagination.

Richard Bruns

Yeah, superhero fiction, like the whole hero’s journey of you only exist to restore the status quo. And it’s very weirdly interesting. and more. Yeah, this would actually suck if people had that power without earning it. And like the difference between proper sci-fi and superheroes, in proper sci-fi usually the people earn it through skills and technical knowledge rather than just randomly getting it.

Fin

Let’s do some final questions. Okay. We asked the same couple of questions to everyone. One is with respect to IAQ, Any kind of research or work that you’d be especially excited to see people just do soon? I mean, potentially including people listening to this?

Richard Bruns

So, if someone who is listening to this is still at a very early career stage and wants to contribute, Yeah, becoming an indoor environment engineer, EA does have a shortage of engineers. We have plenty of computer engineers, and that’s an important thing to be working on, but HVAC engineers, structural engineers, architects, as EA moves outside of the more elite universities and the super high-end mathematical and academic stuff, I would encourage, I see this a lot, I think EA needs to expand and it needs to expand beyond like the elite of the elite. And a lot of, there are probably a lot of people out there who don’t have what it takes to do technical AI alignment, who don’t have the elite credentials from high-end universities. but are still capable of doing a lot of good engineer-type work. And my work with the ASHRAE Standards Committee has told me there’s a lot of low-hanging fruit. If you have the technical expertise and are willing to serve on a Standards Committee, an Engineering Standards Committee, you can do a lot of good for the world. So, especially, and I’ve also noticed I’ve mentored a lot of people at EA Global and stuff. There’s often a cultural pressure in the EA community to drop whatever you’re doing, drop your existing resources and networks and knowledge and jump onto the hot new thing that EA’s decided it needs to do this year. That is so wrong, and it’s, especially if you’re a mid-career person who already has some kind of engineering degree, just do what you’re doing, follow your existing knowledge and connections and network, but just do it applying EA stuff. So if, don’t change career prospects, you can probably do a lot more good for the world by following your existing specialization, introducing people in there to EA, and just kind of moving the needle there, like Tyler Cowen said, like having an EA who’s like in a position of power in the American Navy would be vastly more useful than the marginal EA in a lot of things. So. I would discourage anyone who already has a good career in something else from working on engineering or air quality. But if you’re already in engineering school, if you’re good at engineering, even if you’re an HVAC technician somewhere who’s just listening in on EA because you’re intellectually curious and you’re not part of the cognitive elite, you’ll never be in the Oxford or Ivy League kind of people. Just, yeah, there’s lots of things that you can do with your engineering knowledge. Basic facts about how engineering systems affects people’s health, basic facts about the quality adjusted life years, the life and health costs of things. It’s very easy to move, to do more research on just knowing what’s going on. I mentioned earlier, we need to know the indoor environmental chemistry interactions with UV light. There’s a lot of little technical questions like that, and there will probably be more as we investigate more issues. Anything relating to how the built environment systems affect people’s life and health, especially in a global catastrophic risk system, we need to know more technical details.

Fin

That was an excellent answer. And then finally, yeah, if there are any books or any books that you’ve read, that you’ve read in the past, papers, just any resources, kind of, let’s say two or three of them that you’d be keen to recommend with respect to indoor eye quality, then now is your chance.

Richard Bruns

I’m constantly doing meta-analysis kind of things for like, I do a big literature review and I remember a few papers. I can send you a link to particular papers, like I’ll try to find a toll booth study afterwards. And honestly, your blog post was a very good summary of the available evidence. I’m very kind. I don’t think I can recommend, I didn’t notice that you were missing anything important. So yeah, you’re a good source there. You’ve done your homework.

Fin

And I’ll link to a bunch of reports that you’ve been involved with as well. But with that, Dr. Richard Bruns, thank you very much.

Richard Bruns

All right. Thank you.