What the Study Found
- A randomized trial of more than 150 New York adults with prediabetes tested real versus sham bedroom air filters for about four weeks.
- Real filters cut tiny airborne particles 36 percent and were linked to a nearly 8 percent drop in daily blood sugar swings.
- After four weeks, extra glucose after breakfast was down about 17 percent on average and the peak 13 percent; no margins of error were given.
- Results come from a conference presentation, not a journal paper, and may not apply to people with diabetes or other US regions.
Air purifiers are usually bought to deal with pollen, pet hair or wildfire smoke, not to manage blood sugar. Adults with prediabetes who slept beside working air filters for about a month had smaller swings in blood sugar than those given look-alike units that cleaned nothing. The finding comes from a randomized trial in New York City that has been presented at a conference but not yet published in a journal.
Researchers at New York University’s Langone Health system assigned more than 150 adults with prediabetes, a state in which blood sugar runs higher than normal but short of diabetes, to one of two kinds of portable air cleaner for their bedrooms. None of the volunteers had known cholesterol buildup in their arteries. The results were presented on November 8 at the American Heart Association’s Scientific Sessions, a heart research meeting, and have not been through journal peer review.
Half the participants slept with real filters, built around high efficiency particulate air (HEPA) filters. The rest used sham units, machines that did not clean the air; the trial registry describes them as the same type of unit with the filter removed. Neither side was meant to know which it had. The team measured indoor pollution throughout, and the real filters cut levels of tiny airborne particles by 36 percent. Some of those particles are thinner than one thirtieth the width of a hair.
Volunteers wore continuous glucose monitors, small sensors that read blood sugar every few minutes, for one week before the filters arrived and again after four weeks of use. An artificial intelligence tool picked out mealtimes from the glucose traces so the team could zoom in on breakfast. Statistical models then compared each person’s readings over time, accounting for age, sex and body mass index.
Smaller swings, smaller breakfast spikes
The cleaner air was linked to a nearly 8 percent drop in daily blood sugar swings. After breakfast, the total extra glucose in the blood fell about 17 percent on average, and the peak fell 13 percent. The release does not say whether those figures compare the real-filter group with the sham group or describe change within the real-filter group, and it gives no margins of error.
Lead author Sharine Wittkopp, a cardiologist, said the findings suggest cleaner air at home “may help people with prediabetes keep their blood sugar steady and blunt morning spikes.” The registry lists the swing measure as the coefficient of variation of glucose readings, a standard gauge of how much glucose wanders. In a large real-world study of monitor users, higher variability went with less time spent in the target range, which is why researchers watch it.
How a filter could reach blood sugar
The Centers for Disease Control and Prevention says nearly half of US adults have prediabetes, according to the release. Fine airborne particles have long been suspected as one contributor. A meta-analysis published in 2025 that pooled 46 studies tied long-term exposure to fine particles to a higher risk of type 2 diabetes. Those are associations from observational studies, not proof that cleaning indoor air reverses anything.
In a video interview about the study, the team’s suggested explanation was that inhaled particles interfere with insulin signaling. Insulin tells muscle and fat to pull sugar from the blood by moving glucose transporters to the cell surface, and that step goes awry in insulin resistance. Reviews also point to inflammation and oxidative stress as routes from polluted air to diabetes. The trial itself did not test any of this.
What is not yet known
The caveats are substantial. The result is a conference presentation, not a peer-reviewed paper, and the trial ran for about a month in one city. Senior author Jonathan Newman cautioned that the findings may not apply to people who already have diabetes or to people in other regions, where the pollution differs. The registry lists 150 enrolled volunteers, against the release’s more than 150.
Wittkopp described HEPA filters as an easy, low cost addition to healthy eating, regular exercise and prescribed medicines. The trial did not compare filters with any of those, and it did not test cost. The team plans longer studies to see whether filtration can slow the slide from prediabetes to diabetes and lower the risk of heart attack and stroke.
Reference
trial registry record NCT05994937 on ClinicalTrials.gov. The results were presented at the American Heart Association’s Scientific Sessions in November 2026; no journal paper or DOI exists yet.
- Study type: Randomized, double-blind, sham-controlled trial, presented at the American Heart Association Scientific Sessions; not yet published or peer reviewed in a journal.
- Sample size: More than 150 adults with prediabetes (release); the trial registry lists 150 enrolled.
- Intervention: Portable air cleaner with a HEPA filter running in the bedroom for about a month.
- Comparator: Sham unit that did not clean the air; the registry lists the same type of machine with the HEPA filter removed.
- Duration: About four weeks of filtration; glucose monitors worn for one week before and again after.
- Funding / conflicts of interest: Doris Duke Foundation. Conflicts of interest not reported.
- Data availability: Not reported.
- Preregistration: Registered on ClinicalTrials.gov (NCT05994937).
- Main limitation: One city, one month, no margins of error and no journal review yet; may not apply to people with diabetes or other regions.
FAQ
What is a sham air filter?
A sham filter is a unit that looks and runs like a real air filter but does not clean the air. In this trial the sham units did not clean the air, and volunteers were not meant to know which kind they had. The registry describes the sham as the same type of machine with its filter removed.
Has this study been published in a journal?
Not yet. The results were presented on November 8 at the American Heart Association’s Scientific Sessions, and no journal paper has been named. That means the work has not gone through journal peer review.
Do the results apply to people who already have diabetes?
The study enrolled adults with prediabetes, not diabetes. Senior author Jonathan Newman cautioned that the findings may not apply to people who already have diabetes. They may also not apply to other regions where pollution differs.
What kind of air cleaner was used?
The trial used portable air cleaners with HEPA filters, placed in volunteers’ bedrooms. The release does not name a brand or model. It also does not say whether other filter types would work the same way.
Cite This Page