Indoor Air Quality Can Affect Sleep, but Filtration Is Only Part of the Fix

Michael Feldstein’s case for paying attention to indoor air begins with how continuously people breathe. People notice when they are thirsty or hungry; breathing happens without much conscious attention. Feldstein estimates that a person breathes about 17,000 litres of air a day and says the lungs account for much of the body’s detoxification. His practical point is that air is not an occasional input: it is the one people keep taking in while they sleep.

That makes a bedroom a particularly important place to consider air quality, he argues. A hot, stuffy or unpleasant-smelling room can make sleep worse; cleaner air, in his account, can make breathing feel easier and help the body settle. Better sleep can then affect recovery and performance. He calls air “sleep fuel” at night and “thinking fuel” during the day.

He also argues that people can adapt to uncomfortable conditions until those conditions feel normal. Someone may not notice that their breathing has become shallow in a stuffy room, or that a bedroom smells less fresh than it once did. The comparison he reaches for is being in the mountains, where people often find that breathing feels easier. At home, the air is invisible, and familiarity can make it easy to overlook.

Chris Williamson connects this to two kinds of concern: immediate effects on performance and possible cumulative effects on health. Feldstein answers first with examples he says surprised him. He cites research on baseball umpires, chess grandmasters and SAT scores, saying that measures of performance in each case were associated with air quality. He uses the examples to argue that air can matter for cognitive tasks, not only for people with allergies or an obvious respiratory problem.

Those examples are claims Feldstein makes about research; he does not describe the studies’ methods or results in enough detail to establish what caused the differences. His broader distinction is between fresh air and clean air. Carbon dioxide is one part of the picture: it rises when people breathe in an enclosed space with limited ventilation. Particles and other pollutants are another. A room can have elevated CO₂ because its air is not being exchanged adequately, while filtering particles from recirculated air does not, by itself, remove the CO₂ people are exhaling.

Feldstein gives approximate indoor CO₂ ranges: 600 to 800 parts per million as decent, above 1,000 as a level where brain fog may begin, around 2,000 as a point where people may feel the effects more strongly, and 3,000 as a level at which he says people can feel exhausted and unable to think clearly. He also says outdoor CO₂ is about 460 to 500 parts per million, and recalls it being about 395 when he began monitoring air quality in 2014. These are his figures and thresholds, not a substitute for distinguishing ventilation from particle filtration.

The distinction becomes tangible in spaces designed to keep sound out. Williamson describes a podcast studio with sound-dampening panels and limited opportunities to open the door during a long recording. With several people in the room, he says, CO₂ has reached 1,800 parts per million. The studio is doing its job acoustically, but the same design can make ventilation harder. Feldstein proposes a quiet air exchanger positioned away from the recording area, so air can be moved without disrupting the sound. The exchange makes the trade-off concrete: a room can be comfortable in temperature and useful for work while still becoming stuffy.

Feldstein also describes a community-led sleep experiment run by his company. One hundred and fifty people who had not previously filtered their air tracked sleep for four weeks using devices such as Oura or Whoop. In the first week they did not use a filter; in weeks two and three they ran a Jaspr in the bedroom at fan speed three; in week four they turned it off again. Feldstein says participants provided their data and came from a range of ages and occupations. He explicitly calls it a preliminary “beta test study,” not a randomized controlled trial.

Those are Feldstein’s company-reported results from that experiment. Williamson notes that the first and fourth weeks could be compared with the two weeks when the filter was on, and says the design made him want to run a larger version. Feldstein says he would like to repeat the work during allergy season.

Feldstein also reports changes in snoring among Jaspr users: 30% of people who snored stopped, and among the remaining 70%, more than half substantially reduced their snoring. He says the company receives accounts from couples who had been sleeping separately because of snoring and resumed sharing a bedroom after buying a unit for another reason, such as allergies. These, too, are company-reported outcomes rather than findings from a controlled study.

His broader practical argument is that the bedroom offers a manageable starting point. A person may not be able to control the air everywhere they go, but can take steps to improve the air where they spend many hours breathing it. Feldstein calls a bedroom with cleaner air a “clean air sanctuary”; he presents it as one environmental change that may support sleep, not as a guarantee of a particular health outcome.

Michael Feldstein’s list of everyday indoor-air sources begins with shoes. He says tests have found fecal matter on 95% of shoes, and argues that shoes can also carry glyphosate, tire material, pollen and other outdoor particles into the home. Once on carpet, he says, that material can be stirred back into the air by walking. He describes a “disturbed carpet sample” in which striking carpet with a particle counter produces a large reading, made up of a mixture that can include mold spores, skin cells, hair, pollution and pollen.

His proposed response is simple: take shoes off indoors. The point is not that every particle brought in on a shoe has a known health effect. It is that shoes are one source of material that can accumulate on floors and be disturbed again. Feldstein also notes that porous materials—including carpet, drywall, insulation and some furniture—can retain material. He contrasts these with sealed surfaces, which he says can often be wiped clean.

Cooking is another source he emphasizes. Even when someone uses good ingredients and cookware, Feldstein says, high heat and proteins can produce polycyclic aromatic hydrocarbons, or PAHs. In restoration work after fires, he says, his teams tested for PAHs and hexavalent chromium. His point is that a cooking smell disappearing does not prove that all cooking-related material has disappeared from the room. This is his account of the concern, not a claim that a particular meal or cooking method will produce a specific exposure or health outcome.

His suggested response is to use the range hood—the extractor above the stove—or open a window when conditions allow. The range hood is a form of exhaust: it can remove air from the cooking area rather than merely recirculating it through the home. Feldstein says many people do not use it. He also recommends using the bathroom fan after showers, but for a different reason: to remove humidity. He suggests leaving it on for at least an hour, or using a timer if the fan has one. Moisture left to dry indoors can contribute to a damp environment, he argues, especially when a bathroom door is shut and ventilation is limited.

These steps are about controlling sources and moving air, not just filtering particles after they have dispersed. Use the range hood while cooking; use the bathroom fan to exhaust moisture; open windows when the weather and outdoor air make that sensible. Feldstein also recommends cleaning the dryer’s lint trap and vent, both as a maintenance step and because he says dryer fires occur in homes. For household products that produce a smell or spray, his advice is to reduce exposure where practical: use them outdoors, near an open window, or with an exhaust fan running.

He places these sources in a larger account of buildings. Outdoor air moves and disperses; indoors, people live in enclosed spaces where cooking, cleaning products, building materials and occupants can add to the air. Feldstein estimates that the average home’s indoor air is five to ten times dirtier than outdoor air. He describes a home as a container that exchanges air through doors, windows, vents and small leaks. Even a tightly built house is not completely sealed: air that leaves through a vent or leak is replaced by air entering elsewhere.

That exchange matters because ventilation and filtration are not interchangeable. Opening a window or operating an air exchanger can bring in outdoor air and dilute air that has accumulated indoors, though whether opening a window makes sense depends on the conditions outside. A portable filter generally recirculates air through a filter; it can remove some particles from the air passing through it, but it does not create fresh air or necessarily lower CO₂. Feldstein’s proposal for the podcast studio illustrates the difference: the studio needs an air exchanger to address accumulated CO₂, while a scrubber addresses material that passes through its filter.

Feldstein says homes became more tightly sealed as building codes shifted toward energy efficiency, while their ventilation systems were generally designed to heat and cool rather than provide robust filtration. He calls the HVAC system the lungs of the home, but says standard furnaces and air conditioners are not necessarily sized to filter large volumes of air. He spent two years trying to design a filter for an HVAC system, he says, then found that restrictive filters could choke airflow. During COVID, he says, some people installed high-grade filters that contributed to HVAC motors failing. His distinction is between filtration efficiency—how many particles a filter captures—and how much air the system can move through it. A filter that captures more but severely restricts airflow may not be a practical fit for a system designed primarily to heat and cool.

For that reason, Feldstein does not treat air conditioning as equivalent to bringing in fresh outdoor air. In his account, typical systems mainly circulate and condition indoor air, and often run only when heating or cooling is needed. A filter added to the system may not solve ventilation or filtration problems if the system cannot move enough air through it. He says the long-term goal for his company is to develop HVAC products designed to account for filtration as well as heating and cooling, but describes that as work still to come.

The home itself, Feldstein argues, can retain what enters it. He says smoke can pass into porous materials such as carpets, furniture and drywall, and that the smell may fade even while material remains. In his experience, homes affected by wildfire smoke can be recontaminated when outdoor air remains poor. A filter can help manage air passing through it; it does not remove contamination embedded in materials or fix a source of moisture inside a building. That is why his advice about scrubbers sits alongside, not in place of, source control, ventilation and remediation when a building problem is present.

Feldstein’s work in disaster restoration shaped his view of what air cleaning can and cannot do. His teams handled floods, fires and mold remediation. He describes remediation as isolating a contaminated area, removing the physical mold and scrubbing the air, with measurements taken before, during and after the work. One of his strongest impressions from that work was that the air could feel unusually fresh after several days of scrubbing.

A wildfire in Fort McMurray, Alberta, gave that experience a more consequential turn. Feldstein says the 2016 fire forced the town’s 100,000 residents to evacuate for a month, and that homes were contaminated by smoke from burning houses, cars and building materials—not only by the familiar smell of a campfire. His team cleaned a home and was called back after a baby was hospitalized. Their testing, he says, found that the house had become contaminated again because the outdoor air remained poor.

The episode also illustrates why he insists on comparing indoor air with an outdoor control sample. A measurement from inside alone may reflect what is in the surrounding environment rather than a problem unique to the building. Feldstein gives mold testing after rain as an example: outdoor spores may be elevated, so a high indoor count does not by itself establish that the home is contaminated. In his view, a test without an outdoor comparison can be misleading. He says air testing should compare inside and outside conditions at the same time, and that weather can affect what is found in an outdoor sample.

In Fort McMurray, Feldstein says, the insurer declined to pay for another cleaning. His team left industrial air scrubbers running in the home instead; he says the air measured well again within a few hours. That result led him to a product-design problem. Industrial scrubbers could move enough air but were loud and unattractive. Small consumer purifiers were quieter and easier to live with, but, in his experience, several inexpensive units did not clean the air sufficiently. He set out to make a quieter, more domestic-looking machine with the capacity of a scrubber.

That experience also informs his caution about smells. A fragrance can make a room seem fresh without removing what caused an odor. Feldstein says plug-in scents and sprays can mask smells, including the smell of mold, and recommends treating a suspiciously perfumed home as a reason to look more carefully—not as evidence that the air is clean. When viewing a home, he advises asking for fragrances to be removed, airing the place out, and then assessing it again. He says some sellers use plug-ins to conceal odors.

The same distinction applies to household products. Feldstein’s rule of thumb is that burning or spraying substances indoors adds something to the air. He advises using hairspray outside if possible, or at least near an open window or with a bathroom fan running. He dislikes incense indoors and treats insect spray as something to avoid when possible. For odors in fabric, he recommends washing or taking items outside rather than relying on a fragrance spray. He says the sun and fresh air are preferable ways to deal with fabrics that can be washed or taken outdoors.

Feldstein’s claims about fragrance extend beyond masking smells. He says scent is connected to taste and digestion, and describes fragrance sprays as making people “nose-blind” by blocking their ability to smell an odor rather than cleaning the source. Those claims are his explanation for why he avoids synthetic fragrances; he does not give supporting study details in the discussion. Williamson pushes back on how much fragrance remains in the air hours after someone applies it, and asks whether heightened sensitivity to smells could itself become fragility. Feldstein says he wants to be sensitive to his environment without allowing every exposure to ruin his day.

Candles are a qualified exception for him. He says the cleanest air choice would be not to burn one, but he uses candles himself and likes the light. He prefers beeswax candles with a suitable wick and says he usually extinguishes them outside, to avoid the smoke released at the end. He is still testing brands and does not name one he can recommend.

That distinction—pay attention without becoming fearful—is important to his advice. Feldstein says he often sees highly health-conscious people become preoccupied with finding the next problem, and argues that the stress of constant vigilance can become its own burden. He does not suggest that people need to eliminate every product or avoid ordinary life. His aim, he says, is to make healthier choices passive: improve the environment once rather than relying on repeated willpower.

Feldstein does not describe any region as free of air-quality problems. He says he considers air quality when travelling and names Mexico City, Bangkok, Delhi and Beijing as cities where he has concerns. In the United States, he says, conditions vary by region and every place has its own problems. These observations reflect his experience in restoration and air testing; he does not offer comparative measurements for the places he names.

He describes Florida as having substantial mold concerns, and says hotels may switch off air conditioning between guests, reducing airflow. For Austin, he points to humidity and attic-based HVAC systems. He argues that cold air moving through ductwork in a hot attic can create condensation, and says he sees mold concerns in attics there. He also criticizes flexible ducting, saying it can be damaged during cleaning. These are Feldstein’s explanations of regional risks, not a general survey of homes in those states.

Dry outdoor conditions do not necessarily rule out problems indoors, in his account. Feldstein says people sometimes move to Arizona to get away from mold, but that he has seen mold concerns there too. His explanation is that a home can remain a cool, damp enclosure in a dry environment, and a leak can create an indoor problem. He also refers to dust storms and older material in desert sand; he does not establish how much either contributes to indoor exposure.

When asked where he has breathed especially good air, he names Costa Rica’s rainforest and Vancouver Island. He says the air in the rainforest measured lower in CO₂ than elsewhere and describes the air beneath trees near Austin’s Lady Bird Lake as cooler and fresher. His larger point is that geography, building design, moisture and ventilation interact. A dry or green setting alone does not determine what is happening inside a home.

For someone buying a home, Michael Feldstein recommends environmental assessment alongside a conventional inspection. He says standard home inspections often exclude mold and other environmental issues, focusing instead on problems such as a failing roof or electrical fixtures. A home can therefore pass an ordinary inspection without the inspector assessing whether its air or moisture conditions present a concern.

He also warns that credentials alone may not tell a buyer much. Feldstein says some mold-remediation certifications involve only a short course, and describes the testing and remediation business as vulnerable to alarmism and financial incentives. He calls the industry a “road to hell paved with good intentions”: a doctor may encourage testing, a tester may interpret a result as a serious problem, and a contractor may then recommend expensive work, even when the underlying evidence is incomplete.

His example is ERMI, a dust test that can indicate whether mold has been present over a period of time. Feldstein calls it a useful test but says it can be alarming if treated as a simple yes-or-no answer. Detecting some mold does not by itself establish the extent, significance or source of a problem. He argues for considering which species are present, how much is present, and what further testing shows. He also returns to the importance of outdoor control samples, particularly when weather conditions could affect the result.

Feldstein says remediation contractors may be among the people best placed to identify a problem because they know how to remove it, while also acknowledging that it is difficult to know which contractor to trust. Chris Williamson describes asking friends and his network for recommendations and choosing a company that came up repeatedly. The exchange preserves a real tension: the person assessing a problem may also have a financial interest in fixing it, but avoiding all practitioners who do remediation could leave a buyer relying on someone with less direct experience.

Feldstein offers a practical warning for house viewings: if a home has plug-in fragrances or other strong scents, ask for them to be removed and reassess the space after it has been aired out. He argues that fragrance can conceal an odor, including one associated with mold. He also says conventional inspections may not cover environmental conditions, so a buyer concerned about air quality may need to ask specifically about air and mold testing rather than assume a standard inspection includes it.

The purpose of testing, in this account, is not to turn every detectable particle into a diagnosis. Feldstein says mold can make people very sick, but also argues that the mold industry can make test results sound more conclusive than they are. A result has to be understood in relation to the sample, the conditions when it was collected, and what the test can actually establish. That caution applies to both alarm and reassurance: a high reading without context may not show that a home is contaminated, while a general home inspection may not have looked for the relevant problem at all.

Williamson says he spent more than two years trying to regain his health after what he describes as exposure to poor air quality. He connects that experience to his concern that some people are more sensitive to their surroundings, and says that environmental problems can sound implausible to people who have not encountered them or known someone who has. He also points to lower-cost steps, including a tabletop water filter or improving the air someone is breathing, for people who cannot afford to move or undertake a full remediation.

Feldstein distinguishes the air scrubber he sells from smaller consumer purifiers largely by how much air the machine can move. He says his unit uses a larger filter and can move more air at a lower fan speed. His warning to consumers is that an advertised room-size rating does not necessarily mean a device is moving enough air to make a noticeable difference. He describes the goal as filtering material out of the air rather than adding anything to it.

The product emerged from Feldstein’s experience that industrial scrubbers worked but were too loud and unattractive for many homes, while smaller devices did not perform as he wanted. He says his wife objected when he brought an industrial unit into their house. The resulting design, he argues, had to move substantial air and be attractive and quiet enough that people would keep it in the room. He says the device uses sensors that respond when particles rise, such as during cooking or when a candle or incense is burning.

Those features do not make a portable scrubber a replacement for ventilation or remediation. A scrubber filters air that passes through it. It does not supply outdoor air to a crowded, poorly ventilated room, and it does not resolve a leak, remove mold from building materials or prevent a home from becoming damp. In Feldstein’s account, different problems need different responses: exhaust fans and air exchange move air and moisture; source control reduces what is added; filtration captures some material that passes through a device; remediation addresses contamination in the building.

For people who cannot afford a commercial machine, Feldstein recommends looking up the Corsi-Rosenthal box, a DIY air filter made from hardware-store parts. He estimates the materials cost at $60 to $120 and says it takes a few hours to assemble. The trade-offs, as he describes them, are that it is less attractive, louder and lacks a built-in sensor. He would still prefer it to a small consumer purifier that does not move enough air.

Feldstein says a bedroom is the first place he would put a unit, because it is where people spend hours breathing the same indoor air. If someone has only one, he recommends putting it there; a second can go in the living room or another place where people spend time. He says the device need not be beside the bed and can sit in a corner or against a wall. For an open kitchen and living room, he says it can be placed in the shared space rather than directly beside the stove.

His own company’s evidence is part of the product discussion, and he identifies it as such. Feldstein says the company makes home visits in Austin and measures indoor air before and after installing a Jaspr. He reports that some customers experience no allergy symptoms or a substantial reduction, particularly during the region’s cedar-pollen season. He also reports the sleep and snoring results described above. These figures are observations and reports from the company’s customers and experiment, not independent evidence that a scrubber will produce the same result for every user.

Feldstein does not present scrubbers as the long-term answer. Better-designed homes and less pollution would reduce the need for them; in his words, “the real air filtration is the trees.” He says he is planning further studies, including a larger sleep study and work with companies to measure employee performance and absenteeism when filters are used in workers’ bedrooms. He describes a school in Austin that has air scrubbers in every classroom as a “living lab,” intended to examine the effects of clean air alongside other environmental changes. He cites a study in Finland in which, he says, classrooms with filters had lower absenteeism among teachers and students. The larger studies he proposes are still planned.

Feldstein’s practical hierarchy is not to buy equipment first. He argues that air, water and lighting are environmental conditions people can improve without having to make a new health decision every day. The source-control steps he emphasizes are taking shoes off indoors, running the range hood while cooking, using the bathroom fan to remove humidity, and airing out a home when outdoor conditions are suitable. For sprays and fragrances, he suggests using them outdoors or with ventilation rather than treating a scented room as a clean room.

He also offers a low-cost filtration option in the Corsi-Rosenthal box, while acknowledging the noise and appearance trade-offs. If someone wants a commercial filter and has only one, his priority is the bedroom. That recommendation is about where the person spends time, not a claim that the device will correct every indoor-air issue.

The limits matter. If the concern is CO₂ in a crowded or poorly ventilated room, a portable particle filter is not the same as bringing in fresh air. If moisture is entering through a leak, filtration is not a building repair. If a test shows something concerning, Feldstein’s own cautions point toward interpreting the result in context rather than assuming that one number settles the question. And if a fragrance masks an odor, removing the fragrance and reassessing is more informative than taking the pleasant smell as proof.

Feldstein’s central appeal is to make the home environment easier to live in, rather than to ask people to monitor every exposure. His restoration work led him to value cleaning and maintaining air after a home has been contaminated; his product business grew from the difficulty of finding a quiet, domestic-looking device with the airflow he wanted. But the broader account also leaves room for measures that do not involve a product: remove sources, exhaust moisture and cooking emissions, ventilate when appropriate, and treat filtration as one part of the response.