Does carbon monoxide protect against Parkinson's? Or is there a simpler explanation?

By Josh Bloom — Sep 25, 2026

"Carbon monoxide" and "brain health" aren't terms normally found in the same sentence, at least not in a good way. But a massive new study reports that people with higher levels of exhaled CO were less likely to develop Parkinson's disease. The finding isn't as crazy as it sounds, but neither is a very different explanation.

Image by ACSH using AI

Here's a strange one. Researchers from the University of Oxford and the Chinese Academy of Medical Sciences just reported something that should make neurologists take note: people who exhale more carbon monoxide are 29% less likely to develop Parkinson's disease (PD). Seriously?

Maybe.

Unlike chintzy studies with small study populations, which can easily lead to spurious results, the new study, published in JAMA Neurology, goes in the other direction. Researchers followed more than 512,000 adults in 10 regions of China for about 12 years. Among people who had never smoked, those with exhaled carbon monoxide levels of at least 3 parts per million had a 29% lower risk of developing Parkinson's disease than those with levels below 3 ppm (hazard ratio 0.71).

That's a big study, a long follow-up, and a statistically convincing association. But does it mean carbon monoxide actually protects against Parkinson's?

I don't question the data, but a plausible source of reverse causation isn't addressed and could make the association look more causal than it is.

It's not nicotine. Or is it?

At first glance, nicotine would be an obvious suspect in modifying PD. Smokers are exposed to both carbon monoxide and nicotine, and there is some biological plausibility for nicotine affecting Parkinson's disease.

Nicotine activates nicotinic acetylcholine receptors in the brain, including those involved in regulating dopamine, the neurotransmitter whose loss is central to Parkinson's. Laboratory studies have also suggested that nicotine may protect dopamine-producing neurons from damage. So nicotine has long been a plausible explanation for the strange observation that smokers are less likely to develop Parkinson's. There's just one problem: when nicotine has actually been given to people with Parkinson's disease, it hasn't worked very well.

But what about the nonsmokers?

The failure of nicotine to treat Parkinson's doesn't necessarily mean that long-term exposure couldn't help prevent it. Treatment and prevention are two different animals. But there's another problem. The inverse relationship between carbon monoxide and Parkinson's was also seen in people who had never smoked. So nicotine may still help explain the curious relationship between smoking and Parkinson's, but it can't explain this finding.

Which leaves us with carbon monoxide. Maybe.

CO isn't just a poison

Carbon monoxide has a well-earned reputation as a poison. It binds tightly to hemoglobin, interfering with the blood's ability to carry oxygen, and at sufficiently high concentrations it can kill you.

But CO has another, less sinister side. Our bodies naturally produce small amounts of it during the breakdown of heme, and at low concentrations it acts as a signaling molecule. Experimental studies suggest that low levels of CO can affect inflammation and oxidative stress, two processes that have been implicated in Parkinson's disease. In animal models of Parkinson's, low-dose CO has even been reported to protect dopamine-producing neurons from damage.

So the idea that a little carbon monoxide might be protective isn't as nuts as it sounds at first. But biological plausibility isn't proof. And there's another explanation for the association that has nothing to do with CO protecting the brain.

Driving

Although the researchers controlled for physical activity, passive smoking, geographic location, and the use of solid fuels such as coal or wood for cooking or heating, one potentially important variable is not mentioned: driving. People who drive are exposed to carbon monoxide from automobile exhaust. (On the demonic Cross Bronx Expressway in NYC, the average atmospheric composition is roughly 100% CO.)

Here's where things get interesting. Parkinson's can be brewing for years before anyone knows it's there, and subtle changes in movement, thinking, sleep, or mood can occur during this period. As these changes develop, people may begin driving less. Less driving could mean less exposure to carbon monoxide from automobile exhaust and, consequently, less CO in their exhaled breath.

Hence the problem. Does more CO lead to less Parkinson's, as the authors suggest, or does early Parkinson's lead to less driving, less CO exposure, and consequently less exhaled CO? Either explanation could produce the same result: people with higher exhaled CO subsequently developing Parkinson's less often.

The association is real. Which direction the arrow points is another matter, as is often the case with observational studies.

Bottom line

This is an intriguing study, and with more than half a million participants, the association between exhaled carbon monoxide and Parkinson's disease deserves to be taken seriously. There's even enough biological plausibility to make the idea that low levels of CO could be protective considerably less crazy than it sounds.

But association isn't causation, especially when Parkinson's can be developing for years before it's diagnosed. The authors may ultimately be right about carbon monoxide. Or the causal arrow may be pointing backward.

Carbon monoxide may be protective. Or it may simply be along for the ride.

Tags:

Josh Bloom

Director of Chemical and Pharmaceutical Science

Dr. Josh Bloom, the Director of Chemical and Pharmaceutical Science, comes from the world of drug discovery, where he did research for more than 20 years. He holds a Ph.D. in chemistry.