TL;DR: You can achieve much greater reductions than 50% of PM2.5 with a HEPA filter if you are clever about how you use it. Using a “concentric shells” approach in a 100m^2 apartment has allowed me to enjoy air with up to 99% less PM2.5 than outdoors.

Some of you may recall that QRI held an event called “Cause X” a year and a half ago. Indeed, Dony Christie (whom I recently credited for sounding the alarms in mid-January about the coming coronavirus pandemic) presented: “The researching of possible Cause Xs as itself Cause X“. I think this is a great idea, and it is already bearing some fruit…

Recently, Lukas Trötzmüller reached out to me to follow-up on one of the causes I proposed at the event. Namely, to subsidize HEPA filters as a potentially EA-level intervention. Here was my justification:

[Subsidizing HEPA filters] might be a highly effective way of improving the health-span of a country’s population in a cost-effective fashion. As Robin Hanson has argued over the years, if we truly cared about the health of people, we would be spending more resources on the top 4 drivers of health (diet, exercise, sleep, and clean air) rather than on extravagant medical interventions designed to convince us that “an attempt was made.” Clean air, in particular, seems easy to influence at a rather minimal cost. HEPA filters capable of providing clean air to entire apartments (reducing by 10X the PM2.5 concentrations in the apartment) can cost as little as $70, with an upkeep of about $30 a year for renewing filters, and about $20 a year for electricity. Fermi calculation would indicate this would cut the average person’s daily PM2.5 exposure by half. I haven’t worked out the math concerning the amount of micromorts prevented per dollar this way, but the numbers seem extremely promising.

After stumbling upon the Cause X post, Lukas and some of his friends decided to do a more thorough cost-benefit analysis, which was recently posted on the EA Forum: Cost-Effectiveness of Air Purifiers against Pollution. I mostly agree with his analysis (and I commented on the post with what I thought was missing). His conclusion?

Air pollution is one of the biggest public health problems of our time. Simple air purifiers are surprisingly effective at reducing the harm. In our sample calculation, the intervention easily meets WHO criteria for a “highly effective” intervention in Austria, and the criteria for an “effective” intervention in India. With just a few small improvements to cost-effectiveness, it would qualify as “highly effective” in India too.

There are many ways in which effectiveness could be improved: If the bedroom is shared by two people, effectiveness doubles. Our calculations were made for 10 hours per day of use. Many people stay home for longer than that, and would correspondingly benefit more from an air purifier in their home. It is plausible that we could find more energy-efficient devices and optimize location, placement and timing. Furthermore, devices could be preferentially given to individuals which are at special risk of pollution-induced illness.

Buying air purifiers for people to place in their homes is probably not a promising EA intervention: Cost-effectiveness is two orders of magnitude worse than GiveWell-recommended charities. That being said, there might be much more cost-effective ways of helping people get access to air purifiers. We might lobby governments to subsidize those devices, or to make HEPA filters mandatory for public buildings and vehicles.

I’ve been quite surprised by the results. It seems that using an air purifier has solid health benefits, both in very polluted and in averagely polluted locations. It is surprising that in affluent countries, where people can easily afford these devices, air purifier use is not commonplace. The health benefits are clear and well-studied. I have installed a homemade device in my bedroom, together with a PM2.5 sensor, and plan to place a second device in the office.

If you’re interested in air pollution, air purifiers, or would like to collaborate on future research please get in touch.

I take this as a sign that encouraging people to sketch out new EA causes can be of very high-value*. Indeed, holding semi-regular “Cause X” events might be a great way to find completely new cost-effective interventions to improve human and non-human welfare across the globe.

Now, the topic of air purifiers has been on my mind again lately. And the reason should be quite obvious. Namely, that in September of 2020 we in the West Coast are currently experiencing one of the worst air qualities in the entire world due to the forest fires. I’ve been recommending friends and acquaintances who live in this area to get a HEPA filter NOW, since the fires are expected to continue for a month at least (and the worst is likely yet to come as we are only in the middle of the fire season). Unfortunately, a lot of HEPA filters are now out-of-stock, in what mirrors perhaps the spike in demand for PPE at the onset of the COVID pandemic. As they say, the prevention is better than the cure – I happened to have both air filters and N95 masks on hand before these two events, but not everyone is as lucky. Ideally, everyone in the West Coast would have at least two air filters from the start (I’ll get to why in a minute) but we do what we can with what we have.

Similarly to how early official advice for how to protect against COVID was deadly (e.g. downplaying the importance of masks early on), I am dumbfounded by the low-res and relatively useless listicles of advice I see for how to deal with the smoke from the wildfires. Yes, staying indoors and avoiding exerting yourself is better than nothing. But we could do much better than that. In particular, I think that a little cleverness in positioning can go a long way in optimizing the effectiveness of HEPA filters for maximum health. Below you will find a brief account of my experience with HEPA filters in a fairly-insulated ~100m^2 (~1100 square feet) apartment in California over the last few years (YMMV) and what I recommend you do if you find yourself in a similar situation:

The basic claim I want to put forth is that the ~50% reduction of PM2.5 quoted in most studies can be massively improved upon with a little work. In particular, with two air filters intelligently placed I’ve been able to achieve a reduction of PM2.5 in the 95 to 99% range. But first, what equipment am I using?

• We’ve owned four air quality monitors over the last few years – one I sacrificed by giving it to a friend who went to Burning Man to get some sense of what the air over there is like (answer: really bad), and another one I forgot at the house of a cousin pre-COVID which I haven’t had a chance to retrieve. The two I currently own are: Temtop LKC-1000S+, and BIAOLING’s JSM-131. They both make very similar measurements for PM2.5 and PM10, which also agree fairly closely with the values I get from PurpleAir and AirNow. That said, Temtop seems to be better on the whole – it is even more correlated with the values I get online, is faster (it takes only 1 minute to adapt to a new environment vs. 3 or 4), and is far more granular (it does not jump in increments of 3). So below I will just report the values from the Temtop devise.
• We’ve owned two air purifiers for about 3 years – the LEVOIT LV-H132 with H13 filters. It’s really similar to other LEVOIT models, and as far as I can tell, indistinguishable in its performance:

So here is my experience over the last couple of years:

• With no air filters the difference between the PM2.5 outside and inside is small, typically in the 20% range. So let’s say the air outside is 20PM2.5, then inside with all the windows and doors closed for many hours, the reading will be no less than 16PM2.5. This is one of the reasons why I feel like the advice of “staying indoors with all the windows closed” is not really that useful (and in my experience people tend to believe they are staying out of harm’s way that way, which is very much not the case).
• With one air filter on in the living room, with all of the inside-doors open, I get a stable state where the reduction is in the 60 to 80% range. With two air filters on in separate rooms but with all the inside-doors open, the reduction is a bit higher, in the order of 80 to 85%.
• When I have only one air filter on inside one room with no door to the outside (and closed windows), the reduction is more dramatic, in the vicinity of 90%. This makes sense, as both the “surface area” to the outdoors, and the volume of air to clean are smaller.
• The best outcome seems to come from having (a) one air filter in the living room, and (b) a second air filter in the room I’m in with its door closed. This is what takes the PM2.5 value to be as low as 1 to 5% of what’s outside. For instance, by using this method, when the air quality outside reached 400 PM2.5 outside a few days ago, I was able to enjoy a 5PM2.5 inside my room (!!!). The living room would be at 60PM2.5 and the smoke still somewhat noticeable, but once inside the bedroom there would be no way to detect there was anything off about the air.
• I’ve generally found that the biggest point of entry for PM2.5 is not the windows (which seal pretty well) but the doors that lead to outside. The air also gets in more efficiently when a bathroom vent is on (obviously, as this causes a slight pressure differential with the outside).
• Once a room has been cleaned (i.e. the filter has been running inside it for one or two hours), it takes up to 6 hours for the room to become half as dirty as the rooms it’s connected with through doors. I was really surprised by this, actually. But yes, even though most doors do not work as airlocks, they do dramatically reduce the airflow between rooms.

Based on the above, I would offer a few heuristics for how to maximize the benefits from your HEPA filters:

First, use the fact that your place likely has “concentric shells” that separate the outside from the inside. If you have more than one filter, distribute them along the shells: it is much better to have one filter in your studio and another in the living room than having two in the living room. Each shell’s “outside environment” is the shell surrounding it, so make use of the fact that there is this “sequence of relative outdoors” and place a filter on each of them when possible for an exponential reduction of PM2.5 as you move inwards.

Second, if you only have one filter, place it in your bedroom when you sleep and in your studio when you work.

Third, if you are living with other people, you can basically do this iteratively: clean the air in your room, then pass the filter to the next person, who will clean the air in their room and pass it along to the next one and so on. If each person keeps the filter for one hour, by the time it gets back to you, the air inside your room will still be very good.

Fourth, if you have filters to spare, I would recommend placing one right next to each door that leads outside, as this is the place from which you will be getting most of the particles (assuming your windows are pretty good).

Fifth, if you have the stamina for it, you could in principle partition your place into more concentric shells using plastic sheets and tape. Then you can filter the innermost room, then filter the second innermost shell, then the third, and so on until you reach the shell that is connected to the outside doors. If done properly, this will have created highly clean air for the entire apartment. More generally, once you are done cleaning the air of one concentric shell, you can move the filter to the next one. Think of it as a “moving airlock” that cleans the air one step at a time from the center to the edge.

More generally, I would encourage you to notice that thinking in terms of more dakka really does pay off when the air outside is really bad. Don’t just say to yourself “well, I have a filter so I’m doing all I can already” because the above heuristics (and plain just getting more filters) can dramatically increase the effectiveness of the air filtration. Be thirsty for clean air – crave that 99% reduction!

Now, what about even smaller particles? A lot of people worry that particles smaller than PM2.5 may be the most harmful to health, and HEPA filters may leave them free to roam. Is this true? Thomas Talhelm wrote a wonderful answer to the Quora question “Do air purifiers remove PM2.5” where he shows that not only are HEPA filters effective at removing particles around 2.5 micrometers in diameter, but that they are in fact even more effective at removing even smaller particles. He shows the results of a few of his own experiments using a professional-grade air quality monitor capable of detecting PM0.5 particles.

And what about CO2 concentrations? Scott Alexander has written about the importance of maintaining a low CO2 environment for optimal cognition. I agree with this, and arguably the measures I’ve proposed above may lead to a lot of CO2 in one’s living space. I do happen to also have a CO2 monitor, which shows around 600PPM when the space is very well ventilated, but can reach upwards of 1200PPM if we’ve kept all windows and doors closed for several days (and it seems to stabilize around there). This is worth it, I think, when the choice is between 5PM2.5 and 400PM2.5. But I think you can actually have your cake and eat it too:

The solution I’ve found is what I call the “FF procedure”, which is “Flush and Filter”. Don’t just open a little crack in one window hoping that that the CO2 you are removing will make up for the PM2.5 you are getting. Even if you are filtering the air inside, when the air outside is outrageously bad (e.g. 200+PM2.5) you will notice a fast reduction in the quality of air inside. A formal treatment for how to approach this problem optimally will undoubtedly involve playing with differential equations, but I think my solution is pretty effective and I think minimizes the total exposure to PM2.5 while also giving you good CO2 levels. The idea is to wait for a time when the air outside is not as bad and you then open all the windows and turn all the fans on for an hour and a half or so, which will effectively replace the air inside with the air outside. You then quickly close all the windows and turn the air filters on to their maximum capacity. The drop in PM2.5 will be quick (and indeed exponential), while you will now enjoy fully restored CO2 levels that will last for a day or two.

*I would point out, as another example of this paying off, the light we’ve cast on the issue of cluster headaches and how easily they can be dealt with thanks to psychedelics; it is partly thanks to this work that now OPIS is lobbying for the decriminalization of psychedelics for migraines and cluster headaches in Finland – I have to say that being part of the effort to reduce the incidence of cluster headaches in the world (if successful) will give me tremendous joy. Fingers crossed that it works!

[Featured image taken from: NASA Earthdata on September 10th 2020]