The bikes those riders park outside the hospital have no exhaust pipe. In July, Vietnam's MotoAir study moved into a second phase after screening 102 professional motorbike drivers with personal exposure sensors, silicone wristbands, spirometry and low-dose lung CT — and referring two of them to specialist care with signs of serious disease. That cohort drives for Green SM, an all-electric ride-hailing fleet on VinFast Feliz S and Evo 200 machines. They had already removed their own tailpipe, and they are still under a scanner. The short answer to the question every rider is now asking: going electric does not meaningfully change what enters your helmet, because roughly 85% of vehicular PM2.5 was never tailpipe in the first place.
Key takeaways
- Vietnam's MotoAir study screened 102 professional motorbike drivers with personal exposure sensors, spirometry and low-dose lung CT and referred two of them to specialist care with signs of serious disease — and that cohort rides electric bikes for Green SM.
- Non-exhaust sources — brake, tyre and road-surface wear plus resuspended dust — account for roughly 85% of vehicular PM2.5 and over 90% of vehicular PM10, so removing an engine removes the minority slice (OECD, Non-exhaust Particulate Emissions from Road Transport).
- Measured head-to-head, an electric vehicle emits only about 3% less PM2.5 than an average petrol vehicle, and being roughly 20% heavier it produces more non-exhaust particulate per kilometre (Journal of Hazardous Materials, 2022).
- Hanoi motorcyclists were measured at 95 µg/m³ PM2.5 in traffic against 34 µg/m³ ambient — 2.8× the city's own air — a gap set by where the rider's head sits rather than by what the rider's engine burns.
Six Years of Measuring the Air, and Almost None of It Measured the Rider
For six years the public conversation about Southeast Asian air has been a conversation about ambient numbers: city AQI tiles, national annual means, WHO breaches. Hanoi's PM2.5 annual average climbed from 37.9 µg/m³ in 2020 to 45.4 µg/m³ in 2024 — a fifth consecutive annual rise, and more than 9× the WHO annual guideline of 5 µg/m³. Every one of those figures describes the sky. None of them describes a rider.
The gap matters because the rider's number is not the city's number. A Hanoi commuter-exposure study put motorcyclists at 95 µg/m³ PM2.5 while riding, against 34 µg/m³ ambient at the same time — 2.8× the concentration the city was reporting. The same body of work found Hanoi motorcyclists carrying much higher black-carbon exposure than car, bus or bicycle commuters moving through identical traffic.
Then, in April 2026, the first hard clinical comparison landed. Writing in Scientific Reports, researchers compared 64 commercial motorcyclists with 64 age- and sex-matched controls and found cough in 37.5% of riders against 15.6% of controls (p = 0.009) and phlegm in 31.3% against 12.5% (p = 0.019), alongside lower lung-function indices. Riding for a living showed up in the airways.
2026 is the first year the instruments were pointed at the rider instead of at the sky. That is what makes the MotoAir cohort worth reading carefully — and what makes its drivetrain the most interesting unreported detail in the whole study.
Why Didn't Going Electric Fix It?
Because the rider's own engine was never the main thing poisoning the rider. Non-exhaust sources — brake wear, tyre wear, road-surface abrasion and the resuspension of dust already lying on the tarmac — now account for roughly 85% of vehicular PM2.5 and over 90% of vehicular PM10. Deleting the tailpipe deletes the minority slice. Measured head-to-head, an electric vehicle emits about 3% less PM2.5 than an average petrol vehicle, and because it is roughly 20% heavier, it generates more non-exhaust particulate per kilometre travelled.
The second reason is geometric. A rider sits low, directly in the wake of the vehicle ahead, inside the roughly one-metre band above the road where resuspended dust is densest. At a red light there is no forward motion at all, so passive helmet vents deliver nothing while the concentration around the head is at its highest. That multiplier is a function of where the head is. No drivetrain change touches it.
The third reason is that helmet ventilation is a heat device that has been quietly mis-sold as an air device. Vents are sized and shaped for convective cooling. They carry no rated media, no seal and no pressure differential of their own. Air arrives unfiltered by design — the same conclusion we reached from the other direction in what a closed visor does and does not stop.
One practical note for riders comparing filter classes, since it comes up constantly: the difference between grades is efficiency at the most-penetrating particle size, not a pass/fail line. An H11-class medium rated ≥95% on PM2.5 removes the overwhelming majority of the traffic-borne mass fraction, and a higher class buys efficiency at the cost of static pressure the fan then has to overcome.
Where the particles come from when nothing in the queue has an exhaust
Side view of a stopped traffic queue, travel direction to the left
What Do Rider Lungs Look Like in 2031?
Here is a falsifiable projection with the working shown, because a number without its arithmetic is just an opinion.
Start with the measured base: 95 µg/m³ for a Hanoi rider in traffic. Road traffic contributes 58–74% of Hanoi's emissions by the city's own accounting; take the 66% midpoint. Tailpipe is roughly 15% of vehicular PM2.5, since non-exhaust is the other ~85%. So the tailpipe-attributable share of the rider's inhaled dose is about 0.66 × 0.15 ≈ 10%. Remove it entirely — Hanoi's low-emission zone is scheduled to reach Ring Road 3 by 2030 — and you land at 95 × 0.90 ≈ 85.5 µg/m³.
Now add back what that arithmetic leaves out: a heavier electric fleet raising non-exhaust particulate per kilometre, and non-traffic background compounding at roughly the +4.6% per year implied by Hanoi's 37.9 → 45.4 µg/m³ move between 2020 and 2024.
Projected in-traffic PM2.5 for a rush-hour Hanoi rider in 2031: 85–100 µg/m³ — statistically indistinguishable from the 95 measured today, and still about 6× the WHO 24-hour guideline of 15 µg/m³. That is a synthesis, not a measurement, and it is offered as one.
The unusual thing about this prediction is that the experiment to test it is already running. The MotoAir cohort electrified before it was enrolled. If drivetrain were the dominant variable in rider exposure, those would be the healthiest working riders in Vietnam. They are instead the first to be given lung CT scans, and two of them have already been referred onward. We argued the same physics from the policy side when Hanoi announced its petrol-motorbike ban; this is the clinical version of that argument, written by a scanner rather than by a model.
What Actually Changes the Number
If the dose is set by geometry and by non-exhaust particulate rather than by the rider's own engine, then the only intervention that moves it is one placed between the traffic and the rider's face. That is the entire design brief for Easi Breezi: it draws air through an H11-class medium rated ≥95% on PM2.5 and delivers it into the helmet under positive pressure, so intake does not depend on forward motion. It works at a red light, which is exactly where the concentration peaks. It is IP67, patent pending, and vehicle-powered with no battery, so it runs as long as the bike does. Filters are consumable and sold in multi-packs. Pre-order is $199.00.
For everything you can do without buying anything, our guide to protecting your lungs when you ride all day covers position in the lane, timing and following distance.
Frequently Asked Questions
Does an electric motorbike reduce what I breathe while riding?
Barely. Roughly 85% of vehicular PM2.5 comes from brake wear, tyre wear and road dust rather than exhaust, and measured head-to-head an electric vehicle emits only about 3% less PM2.5 than a petrol one. Electrification is a genuine fix for a city's emissions inventory and a very small one for an individual rider's inhaled dose.
Is it safe to ride a motorcycle in heavy traffic pollution?
It carries measurable risk. The April 2026 Scientific Reports comparison of 64 commercial motorcyclists against 64 matched controls found significantly higher rates of cough (37.5% versus 15.6%) and phlegm (31.3% versus 12.5%), plus reduced lung function. Riding occasionally is a different exposure from riding for a living.
How toxic is the air actually inside my helmet?
In Hanoi, motorcyclists were measured at 95 µg/m³ PM2.5 while riding against 34 µg/m³ ambient — 2.8× the city's own reading. An open helmet does not lower that figure, because its vents carry no filter media and no seal.
Why is pollution worse when I am stopped than when I am moving?
Because there is no forward airflow to displace what the queue is producing, and you are sitting inside the roughly one-metre band above the road where resuspended dust is densest. Passive vents rely on movement, so at the exact moment concentration peaks they deliver the least.
Are lung CT scans normal for motorbike drivers?
They are not, which is why the MotoAir study is notable. It screened 102 professional drivers with exposure sensors, spirometry and low-dose lung CT and referred two with signs of serious disease — the first imaging cohort of working riders anywhere.
Ready to Breathe Cleaner on Every Ride?
The fleet under the scanner already went electric, and it did not save them. What reaches your lungs is decided a metre above the tarmac, in the wake of the vehicle you are sitting behind, not under your own seat. Filter what you actually breathe: pre-order the EB unit at $199.00.
Written by Ash — mechanical engineer and founder of Easi Breezi, building an active HEPA filtration system for motorcycle helmets (patent pending). Based between Hong Kong and Bali, riding daily in the traffic this blog writes about.