A burden-of-proof meta-analysis published in Nature Aging worked through the studies linking long-term fine particulate exposure to dementia and applied a method designed to report the weakest defensible version of an association rather than the average one. Across exposure levels from 4.5 to 26.9 micrograms per cubic metre, it found a minimum 14 percent increased risk of dementia, and a stronger association specifically with Alzheimer's disease. For vascular dementia it found nothing significant.

That is a careful result from a method built to be conservative. The obvious next question, whether cleaning the air you breathe indoors changes anything measurable, has a much shorter answer.

From the Lab to the Ledger

PM2.5 means particles under 2.5 micrometres across, small enough to pass the nose and throat and reach the gas-exchange surfaces of the lung. From there the plausible routes to the brain are two. Particles provoke inflammation in the lung that spills into the circulation, and systemic inflammation is one of the better established contributors to neurodegeneration. And the very smallest particles appear able to travel directly along the olfactory nerve from the nasal cavity into the brain, bypassing the blood supply entirely.

Neither route is speculative in animal work. What makes the human epidemiology hard is that PM2.5 exposure tracks with almost everything else that predicts dementia: income, education, traffic noise, green space, access to care. Researchers adjust for these, and adjustment is never complete. The burden-of-proof approach answers exactly that problem. Rather than reporting the central estimate, it reports the smallest effect consistent with the data, then rates the strength on a star scale. Dementia scored two stars, Alzheimer's three. Neither is the highest rating available, and the authors list the constraints plainly: few studies, geographically concentrated, PM2.5 considered alone.

Indoor air is where an individual actually has leverage, since most people spend the large majority of their time inside and indoor concentrations track outdoor ones closely. The intervention trials here are small. The most informative is a twelve-month double-blinded randomised trial in Hong Kong in adults over 70, with 47 participants split between real and sham purifiers. Indoor PM2.5 fell by about 28 percent. Diastolic blood pressure fell by 4.6 mm Hg in the treated group, which was the only statistically significant result. Systolic pressure, pulse pressure, carotid artery thickness and vessel function all moved in the favourable direction without reaching significance.

Forty-seven people over a year is a pilot. It is also close to the best evidence that exists, because a trial large enough to detect a change in dementia incidence from an air purifier would need thousands of participants and a decade.

Bio-Pipeline Ledger

Long-term PM2.5 exposure as a dementia risk factor: supported by a conservative meta-analysis at a minimum 14 percent increased risk, with a stronger signal for Alzheimer's disease and none for vascular dementia.

PM2.5 and cardiovascular disease: the longer-established half of this literature, with a much larger evidence base than the brain findings.

Indoor air purification with HEPA filtration: demonstrated to reduce indoor particle concentrations substantially, with a small randomised trial showing a modest diastolic blood pressure benefit and nothing yet on cognitive outcomes.

Purifiers sold with ionisation, plasma or UV modules: the added functions are not what removes particles, and some ionising designs generate ozone. Filtration is the part that works.

Simple exposure reduction, meaning closing windows on high-pollution and wildfire smoke days, using kitchen extraction, and not burning things indoors: cheap, immediate, and not dependent on any device.

Outdoor air quality regulation: the intervention with the largest documented population health effect in this field, and the one no individual controls.

Established dementia risk reduction including blood pressure control, hearing treatment, physical activity and not smoking: better evidenced than anything in this article, and unaffected by it.

The Clinical Reality Check

What is established is that living in more polluted air is associated with more dementia, that the association survives a deliberately conservative analysis, and that the biological routes by which it could happen are real rather than hypothetical.

What is not established is that filtering the air in one room changes a person's risk. The intervention evidence consists of small trials measuring blood pressure and vascular markers over months, and the one clear result in the best of them was a few millimetres of mercury on diastolic pressure. Anyone selling a purifier as dementia prevention is several steps ahead of the data.

The proportionate response depends on where you live. If your outdoor air is routinely poor, or wildfire smoke reaches you for weeks at a time, a HEPA filter in the bedroom is inexpensive, demonstrably reduces what you breathe for a third of your life, and is a reasonable thing to own on the general principle that less particulate is better. A box fan with a filter taped to it achieves most of it for a fraction of the price. If the air where you live is already clean, the same money spent on blood pressure control or a hearing check buys far more certainty.