An update on air filtration and COVID-19: Indoor air quality expert explains what we’ve learned to keep ourselves safe


Prof. Jeffrey Siegel with the Toronto skyline behind him. (Photo by Daria Perevezentsev)


CivMin sat down (virtually) with indoor air quality (IAQ) expert Prof. Jeffrey Siegel, as an overview and to update a March 2020 Q&A from the beginning of the pandemic.


In the last almost two years, we’ve learned a lot as a society, with the public developing vocabularies we didn’t have prior to COVID-19. Now we know COVID is airborne and can be transmitted via aerosols and droplets – previously we didn’t know there was a difference in those things. Now, thanks to experts like you, most of us do know a bit more. We now have vaccines, we’ve endured lockdowns, masking mandates, etc to combat the spread of this disease.

With all of this, can you tell us a bit about what’s worked and what approach, or approaches, we should continue to use?

 I want to begin by saying that we have learned a lot about taking precautions and we’ve learned a lot about this specific disease. I think one piece that’s been really interesting is, if you look back to the Q&A from March, there’s not a lot that I would change.

What was brought to light was the way we approach infectious disease; it’s very much from a public health, or from a medical context, yet the transmission is strongly affected by things going on inside our buildings. I hope, being most optimistic, there is now this understanding that it’s a building science problem. There were people who’ve known this for hundreds of years and they weren’t even building scientists. It has changed very little.

One thing in particular, this idea of droplets versus aerosols, is a hugely false dichotomy. The boundary used to be five microns- the idea that there is a threshold in particle size where there’s different behaviour is ludicrous. A 4.9 micron particle behaves very similarly to a 5.1 micron particle – there’s no difference there. My own research shows, and anyone who’s changed the [furnace] filter in their home knows, it’s covered with huge dust particles. Those things travel 10s of metres, go around 90-degree bends, and yet they end up on your filter.

So, even if the people who believed early on very large droplets were mostly responsible for the transmission of this disease, those big respiratory particles can travel for a long time indoors, and so we’ve known that for a long time there’s people who have shown that, for example, a sneeze or cough or even a breath can be like a cohesive cloud that moves for long distances.

That’s a long introduction by saying very simply that we have some new evidence, but we also have an appreciation by some who maybe didn’t fully appreciate it before. More of an appreciation for how important some of these factors can be. I recall in June 2020, someone who was on the radio a lot from the public health community said “indoor” for the first time as one of the risk factors for this disease. I turned to my wife and I said, “Thank God. That’s the first time I have heard someone in the public health community say ‘indoors’ aloud,” and that’s everything, right? We have seen there is very rare cases where outdoor transmission has occurred.

This is almost entirely in indoor disease, and yet people don’t talk about indoors. Now, just to follow up, I was listening to the radio again last night as I was cooking dinner and noted the expert they happen to be interviewing, again from the public health or the infectious disease community, is routinely listing ventilation as an important factor for transmission.

There has definitely been a change in the discourse. I’ve tried to never say “I told you so,” from the beginning of this, but we’ve known for hundreds of years ventilation is important for the spread of infectious disease. It’s nice to see that recognized.

To answer your question, I think the thing is what has been most useful to me to understanding the transmission of this disease is this idea there’s nothing magic here. I am well aware filtration and ventilation are solving one piece, but not even necessarily a very big piece in some environments.

There really has to be this idea of a multi-layered transmission protection model. In much the same way each layer could be represented as a slice of Swiss cheese, there are some holes in each one, but put together we hope to block most of the holes at some point.

A great layer, for example, is vaccines, but there are issues with vaccine hesitancy and vaccine availability. Not to mention, there is clearly  breakthrough going on. There’s the threat, or the reality and concern, of new variants the vaccine might be less effective for. I’m not qualified to evaluate vaccine effectiveness, potentially waning over time, and so this layer is a great layer, but imperfect. And we can kind of go through all the other things we think about for infectious disease.

Physical distancing, masks, hand washing are all great layers, but they all are imperfect. Of course there are different qualities of masks, different fits of masks and so on. This is to emphasize all these layers are imperfect. They have holes in them.

You need lots of layers to have a small chance of the holes lining up, and you also have to do those layers well to have as few holes as possible. For example, you can say having a mask policy, is a great layer. But it’s got a lot of holes in it, because people often wear masks that don’t fit well or the efficiency of the masks varies. Or a problem I face, especially after a long lecture, is my mask is wet, and a wet mask is actually not that effective. Masks are a great layer, but they really have to be done well, coupled with good education, and potentially with enforcement. You also have to provide masks for people – not everyone has access to good masks. We have to do the education about what makes a good mask, when to replace your mask etc.

And so you put all that together and you want to do the many layers as well as possible, so there are a few holes as possible.

Prof. Jeffrey Siegel with a filter in a lab setting. (Photo by Daria Perevezentsev)


You bring up the point about indoors, as that’s your area of expertise. Here in Canada, it’s gotten colder and we’re seeing [COVID] case numbers rise recently. It’s really not a surprise, because we are gathering and we’re doing more things now like returning to classrooms returning to school, workplaces, and also unmasked indoor dining. As well, playing team sports, attending sporting events like hockey games, basketball games and concerts. Is there anything we can do in these instances to help ourselves? What should people do, or look out for? Feel free to use some of your own personal judgment as recommendation, as you are the expert.

I’ll begin by answering, and I want to be really careful here, because there are times where I think my advice aligns with people in the public health and medical community. There is also times where my advice is, potentially, a little bit divergent and I want to be really clear: I’m not a medical or public health professional, so I say this kind of cautiously. I do think the most important thing everyone needs to do is a good assessment of their own risk before any event.

I am a big fan of food and of eating out, and I think local, small restaurants are just enormously important to Toronto. Yet, I’m still not going out to eat. I’m still doing takeout when we get food and that’s partially because of people in my family and my community who have a higher risk, either because of their age or because of some underlying medical need. Also, there are people I interact with who were not vaccinated, because they’re too young to be vaccinated.

The most important thing is everyone should do their own assessment. What I don’t want people to take from anything I say is, “don’t do that.” Instead, evaluate the risk; how important is it to you to attend, for example, a sporting events?

Before we get to the spectator aspect, if you’re participating there’s enormous benefit with physical activity – it’s important for people’s physical health or mental health. Earlier this year I was saying people should go play hockey/other team sports, but you just have to do a risk assessment. If you have someone who’s vulnerable at home, you might change your behaviour. You might have to think differently about things like changing rooms, carpooling to get to the facility, or public transportation and all those kinds of things.

I’m certainly not saying don’t do something, only that you have to do a good evaluation of the risk. That’s the real challenge in the public residential environment. You often don’t have control over things like filtration and ventilation. You don’t even know how to ask about it, and so the only thing I tell people to do is, ask what measures are being taken. While you won’t necessarily be able to fully understand the answer, but you can assess the level of engagement. To me, that’s the most important thing. If there’s someone who can’t stop talking to you about the new ventilation system they put into the restaurant, then you know you’re probably in good shape. If someone says, “I don’t know, the landlord did something,” then that’s probably a good indication there’s not been a lot of attention paid.

Beyond that, I think what we’re dealing with is much less of an individual physical space problem, and much more a kind of society evaluation problem. Would you ever have thought about this two years ago? And the answer is no.

As an example, a kitchen renovation might involve selecting what countertop material you should use, but no one cares about the kitchen range hood fan. From an exposure point of view with respect to things that are likely to cause you harm, the kitchen range hood fan is enormously important to your health and it costs a tiny fraction of what your countertops do.  But, it gets very little attention.


There’s now more of an awareness of the benefits of HEPA filters. [The Faculty of Applied Science and} Engineering has installed large, mounted HEPA filters in the main lecture halls and provide measurement of air changes per hour, which you have mentioned at times as “ACH”, right? And you mentioned the level of six ACH, or higher, as being good.

ACH equivalence is usually done for particle removal, and six is a really good target. We don’t have a ton of evidence that the number should be six, or seven, or five, but some number much higher than it usually is [on its own] we know is beneficial.

The ACH equivalence idea is a way of saying that we can’t always achieve what we want with ventilation, because we might not have enough heating or cooling capacity, might not have a big enough fan, or we might have an outdoor air quality issue. Effective air exchange considers the impact of the fresh air from outside, any recirculated air that goes through a filter, and then any portable filters, and kind of puts them all on the same scale. They all look at removing particles, including these respiratory particles that might contain the virus.

The U of T case is an interesting one. In the interest of full disclosure, I’ve been working extensively with the U of T Faculty Association (UTFA), which has been in opposition to the administration on this issue. I think the target of ACH six, or higher, is right and the University certainly got that right.

The problem is, and this gets to the layers and the holes in the layers, you really have to measure it to understand what you’re actually achieving. I’ll give you some simple examples of why you don’t. You might of your six air changes, and let’s say two of them are coming from outdoor air. Lots of things happen this time of year. Things can occlude the outdoor air grills with leaves or something. Block the grills and then you will have less outdoor air or there can be HVAC controls that limit airflow. As one example, there are often a lot of control issues around the dampers on outdoor air supplies.

With a filter there are whole bunch of things that are really important and some of them are quite easy to understand.  A lot of filters in big commercial buildings, and also in our homes, go into some sort of rack. If there are gaps around the filter, air goes around the filter, rather than through the filter. Ironically, the better the filter, the more air bypasses the filter. Unless you go through an active process of looking for, and limiting those call bypass gaps, you’re going to get a much lower performance.

It’s also important to point out the filtration world is kind of divided into charged media and uncharged media filters. The efficiency of uncharged media filters tend to improve, or stay about the same performance, as they age. Charged media filter efficiency can decline, depending upon many details. For the classroom situation the concern is how much the filter has declined in performance.

We did a study in 20 Toronto homes for a year, a few years ago, where we looked at different filters in each of the homes and their long-term performance. We found in a lot of cases a MERV 14 filter, which is a very good filter, better even than it is in most of the buildings at U of T. In some homes it was performing like a MERV 5 filter when it was new out of the box, and that’s because of the bypass gaps. And then you take it to the end of its life. Again, it depends on what it’s loaded with, how much a bunch of things that we only kind of partially understand from aerosol science perspective. You can understand filters decline in performance as they age, so at some point in their life they become much less effective.

There’s lots of things in the literature that shows a MERV 13 or 14 filter is performing worse than half as well as it should. In some cases, much, much less, as it as it gets a few months into its life. Thus it’s really important to do measurements of the effect of air change rate in in classrooms to capture degradations in filtration and ventilation performance.

Despite the fact that I’ve been working extensively with   [University of Toronto Faculty Association], there is a lot to be gained from measurement.  You might find that the effective air change rate is lower than intended in a classroom. By separately looking at the decline of respiratory sized particles and CO2 we can say we can start to understand where is the failure? Is it something related to filtration? Is it something related to the portable cleaners? Is it something related to ventilation and then we can fix the issue and actually maintain performance.  This is fundamentally why measurement is so important.

One final comment I’ll make, that I think is really important, is we often think about the cost. It’s expensive and I’m sure that the institution will tell you how much more money they are spending on energy as well as the capital costs for better filters, and so on – I’m sure it’s a big number.

The problem I have with valuations, which we’re seeing over and over again and have historically why HVAC systems are so poor, is we’re not looking at the benefits. Things like reduced health care costs – improved cognitive function and decision making, reduced absenteeism, and improved productivity. There is a long list of benefits.

When people have tried to assess the benefits of things like better filters, or improving ventilation, the cost-benefit ratio depends a lot on the context and the particular thing you’re trying to deal with. All of those estimates are in the range of 10-100 to one in benefits versus costs. For every dollar we invest in indoor air quality, we make back as a society somewhere between $10 and $100 – most of that’s avoided health care costs.

At some point COVID will be in the rearview mirror, though I can’t say when, because I don’t have a crystal ball. When that happens, I hope we don’t relax back to the “this is costing us so much money” thinking. Instead, let’s do this right; let’s keep it up.

While we do have to think about energy, as we have a climate crisis, we have to do this well and in some cases that’s going to take thinking about schools across Ontario. We’re talking probably billions of dollars of investment to fully redo systems.


Just to verify, you referenced “MERV” numbers – a MERV 5 is poorer filtration than the MERV 12, just to be clear, right?

Yes, the MERV [Minimum Efficiency Reporting Values] scale goes from one to 16.

You can think of MERV 1-4 as what I call “boulder catchers,” but they’ve nothing to do with our health. The filters are there to protect the equipment – keep coils clean, fans clean and ducts clean from big pieces of dust. Then, once you start getting a little higher, maybe MERV 8 and above, you start seeing some potential health benefits, although a lot of it has to do with the context and how it’s used.

The high end of the scale is a MERV 16. And even higher efficiency filters are available, things like HEPA filters.   You might say, why don’t we just put HEPA filters in all of our HVAC systems? And the answer, as we already talked about in the original Q&A, is there’s a real pressure drop associated with such filters and our systems can’t take it. Not only that, you can get most of the benefit from a well performing non-HEPA filter.

ASHRAE [American Society of Heating, Refrigerating and Air-Conditioning Engineers] has decided MERV 13, or so, is an appropriate balance – it’s still doable in a lot of systems but has a high efficiency. I’ve said many times over the course of the pandemic, and I’ll say it again, I’d much rather someone put in a lower MERV filter and use it well, than a higher MERV filter used poorly.

How does somebody who’s looking for a furnace filter in their local store ascertain its rating? Does it provide a MERV number on the filter?

Great question. The short answer is many filters list MERV; the slightly more complicated answer is some filter manufacturers don’t like these ratings, so you sometimes have to rely on other information like equivalency scales between whatever the manufacturer uses and MERV. What I can say definitively is that you shouldn’t think of the MERV number, or whatever filtration rating number is on the filter, as a magic elixir. Think about it as you’ve got a filter that testing says MERV 13 or an equivalent, that’s a good filter, and then to get that performance you have to use it well.

A big issue in furnaces, and we saw this in our filtration study, is your furnace fan doesn’t run continuously. The average runtime over the course of the year in our 20 Toronto homes was about 10 per cent. Meaning 90 per cent of the time it doesn’t matter what filter you have installed, as air doesn’t go through it. You’re not just solving a filtration problem, you’ve got a bigger system problem you have to address.

One solution is, if you have a smart thermostat, it can often cycle the fan if it hasn’t needed to run for conditioning purposes, then you can get the benefit of filtration. A lot of filters and thermostats have manual fan operation. You can run the fan manually, but again, it’s part of this whole picture of making sure the filter is installed correctly in the first place.

With our filtration study, which to my knowledge is the largest study of residential filtration that’s ever been done, we found no benefit to increasing from MERV 8 to MERV 14, Including some both charged and uncharged media filters,. In our sample of homes, we found no discernable difference in the particle concentration within. Over the course of a year, each filter was in for three months, at different times of the year and a random order in each house. It was pretty solid design and we didn’t see any difference in the particle concentrations based on the filter installed.

However, our sample was mostly houses in downtown Toronto, and were predominantly older homes. Older homes tend to be relatively leaky, and so another truth about filtration is the more air leakage, but less relative difference the filter makes because some particle removal is already happening because of the air leakage of the home, and so you’re adding a small removal on top of the leakage and it doesn’t make as much difference as in a more airtight home.

I’ve never found this anywhere in the literature, but I just know it to be true because of my own life – older homes tend to be dustier. So we had these homes that were very clearly older, leakier, dustier and then fans that didn’t run that much, so it goes to show you really have to look at the whole picture of filtration and not just the filter itself.

One of the outcomes of that project funded by ASHRAE, that I think has a lot of relevance, is that ASHRAE is trying to decide do we need a residential filtration standard? It’s clear the commercial filtration standard doesn’t really apply well to residences, so should we do a residential filtration standard? And the answer, which I felt was correct, and that ASHRAE I think also largely agrees with, is that we don’t need a standard for residential filters – we need to provide guidance on people on how to use filters well. Things like when to change them, fan runtime, eliminating bypass gaps and more.


You touched on this briefly, and this is a good way to bridge HEPA filters and standalone units. Places like dentist offices have reopened, partitioned off rooms, and each examination room might have a standalone HEPA filter. Is that just for peace of mind, or do these things really work well? Should we think about them in other environments such as office spaces, indoor restaurant dining, or even our own homes?

As we look towards family gatherings, or other social gatherings for the holiday season there will be a lot more people in indoor spaces. Are these HEPA filters effective and worthwhile pursuing?

We have portable filters at home we use whenever there is someone in the house who’s not part of the household. So, absolutely, we should all be using portable filters.

I do have some hesitation – number one is it’s kind of the quick, easy solution. You know, buy something, plug it in and you’re good. It’s not a bad solution, given that overnight you can make a change, assuming you can find a portable filter (some are impacted by the same supply chain issues affecting many other products). But the thing I have some hesitation about is that I see it as a temporary solution. It only addresses some indoor air quality problems – particles, including respiratory particles. Things like odors, a HEPA filter doesn’t do anything for. I’d much rather people have a HEPA filter than not, and the scenarios you mentioned, such as for restaurants I think could very often benefit from it, but I feel like in a lot of cases places would be better served by a more kind of comprehensive solution. So, if HEPA filters are a bridge to that comprehensive solution, great.

HEPA filters in places like classrooms are a no brainer. As for the cost, some people have looked at the cost of these things are approximately a nickel per kid per day in terms of capital and operating costs. It’s just silly we’re even talking about it. The worst that happens is they don’t do anything for COVID, and the best that happens is they have all these other benefits we know about that kind of 10 to 100 to one benefit cost ratio. We should just do it, but it should be our starting point, not our endpoint.

Getting back to the family gathering for the holidays. Again, the first point is your risk assessment. Who’s there? How many people? From how many different contexts? How many are unvaccinated? How many are vulnerable because of age or other factors? Everyone has to make that decision themselves, but I would encourage people to be cautious.

To be very clear, we don’t have people over a lot, but social life is important. Then we also have people who are potentially vulnerable, so we take a combination of measures in our own house. Windows open, which I feel kind of mixed about because it doesn’t always provide ventilation, but certainly helps a lot. So, either windows open or HEPA filters operating – we aim for a target of somewhere between six and 10 air change equivalents with our HEPA filters alone in our space. We also have a decent filter in our central system, but I don’t rely on that alone for the reasons I just mentioned previously. And masks, particularly on any unvaccinated individuals which historically are kids under 12 in recent times. But, again, that’s getting better as they are becoming vaccinated.

Over the holidays, like a lot of people, I’ll be travelling and seeing family who I haven’t been able to see for over two years at this point.  It’s important to me to see them – some of them are in assisted living or other similar situations. We’re taking pretty extreme protection measures like everything I just said – filters, masks and improved ventilation as we can. We’re also doing a fair amount of rapid antigen testing – again, that scales with the risk. The combination of my family travelling and going to see people who are potentially vulnerable raises the risk level.  [important update: this trip was cancelled in light of Omicron – see note at end].


Are there any other final thoughts you’d like to impart or recommendations you’d like to make?  

Yes, there are. I think there’s two other really big picture points I would like address. Number one is one of the issues that’s come up as a potential risk is what about people who live in condos or apartments or other close settings where is there a risk of transmission? The short answer is we don’t have a good read on that. There does appear to be some risk, but I think it’s relatively a small risk. If you live in an apartment, or an attached home, you smell your neighbours cooking sometimes, or that sort of thing, so you know your air is shared.

There have been multi-family outbreaks, including in Ontario as well as globally. There does appear to be the potential, but you shouldn’t think about it as coming through your walls. It’s more likely things like vertical plumbing chases, where the air can go from the top to the bottom of the building and potentially interact with multiple apartments. It’s something I think we as a society have to think about going forward in terms of things like building codes – Prof. Marianne Touchie does a lot of work on airflow in multi-family buildings. I do want to want to say, however, I think the risk is quite small, but to still be cautious.

The risk we should all think about, no matter in what environment we live, is if someone in our household is exposed to COVID. If someone either gets COVID, or is in some sort of high risk situation, I do think we do have to think about how we can isolate people safely. You have to think about is there a bedroom they can go into? Is there a bathroom or a process for using the bathroom that we can minimize the risk? Under ideal scenarios, can we keep the space that they’re in under negative pressure? We need to think about these adjustments we make for isolation because it’s really important. We need to think about it before it happens if someone in the household is high risk for, or has, COVID.

The second big picture issue is that we obviously have a health disparity and inequity issue in society, and COVID is a perfect example that obviously has impacted some communities much more than others. What I would like to think about as a model going forward is can we start to think about addressing some of those issues by addressing indoor air quality. For example, can we address things like housing quality, which has enormous benefits to society and to the individuals?

It also has all kind of health issues, so I think that as a society a really good investment we could make would be in looking at how do we improve the quality of indoor air in congregate settings? We’ve just seen such devastation at long-term care homes and such a potential risk in  drop-in centres and homeless shelters.

We really need start to think about addressing these vulnerable areas and is there a viable model of improving indoor air quality and actually realizing those benefits? Use the 10 – 100 to one ratio, leverage that and instead of leaving money on the table, actually invest and improve and contribute to correcting those disparities.

I would say, more than anything else, what I hope comes out of COVID is a future where we’re looking at equity issues and addressing them by improving indoor air. I think that’s something that will draw the interest of lots of people, not just engineers, but a lot of people. That is what I’m most optimistic about; that’s the future I can see.


UPDATE NOTE: This interview occurred before the full details of the Omicron variant were as clear as they are now.  One follow-up question was whether omicron changes anything?

Short answer is it changes everything and nothing.  It changes everything, because first of all we have seen some very prominent individuals in Ontario acknowledge the transmission is airborne for the first time. It also changes everything because it is so transmissible. It is still too early to know the full impact of Omicron, but it is prudent to take strong measures until we have a better sense of how serious it is and what impact rising cases will have on our healthcare system.

It also changes nothing: Omicron is so transmissible that it will find a way through any holes in the layers of protections and so I would double down on the advice to have lots of layers and do them as well as possible.

I think it is important to not be alarmist: we are likely in for a tough winter and spring: we will get through it.  As with other waves, our priority should be reducing the risk as much as possible for those in racialized and marginalized communities who have borne the brunt of earlier waves.  It is also important to not be defeatist, the vaccines (with boosters) are an amazing layer, the medical community has developed a variety of treatment options, and we know a lot more than we did two years ago.  We will get through this and be in a better place.


By Phill Snel



Jeffrey Siegel is a professor in the Department of Civil & Mineral Engineering at the University of Toronto and an expert in indoor air quality. His research has looked at a wide variety of airborne particles and air cleaning approaches.