New study: air pollution is deadlier than cigarette smoke or AIDS

Originally published at: New study: air pollution is deadlier than cigarette smoke or AIDS | Boing Boing


… and now we’re learning that antibiotic resistance is correlated with fine particulate concentrations.

“The study authors acknowledged limitations in their own data and called for further research to better understand how air pollution affects antibiotic resistance. But based on the available information, they concluded that every 10% rise in PM 2.5 increases antibiotic resistance by 1.1%. By 2050, with no changes to current policies, air pollution could trigger a 17% increase in drug resistance, leading to 840,000 premature deaths in that year.”


This data is limited to large air particulate pollution that is detectable from satellite imagery, so this is a low-end estimate of the effects of air pollution.

For instance, more acute pollution like in Cancer Alley which greatly shortens lifespans in South Louisiana doesn’t show up in to this kind of data.
Clean air and clean water are always the most important things for our health


Firsthand cigarette smoke reduces life expectancy by 1.9 years, on average, according to the report.

That’s not what the CDC says

Life expectancy for smokers is at least 10 years shorter than for nonsmokers.

The sources it cites:

To be clear, air pollution causes a variety of severe problems & we must start working to correct the harm brought about by air pollution but it’s not fair to underestimate smoking’s harms.

1 Like

Conflating two different stats. Smoking is responsible for reducing the average human lifespan by 1.9 years, but that includes all the humans who don’t smoke. Likewise, that stat about how “conflict and terrorism” only reduces the average human lifespan by 7 days is only true because most people aren’t in war zones or the victims of terrorist attacks.


CNBC screwed up the description there, because it’s not what the EPIC report is claiming. They were trying to convey it is the effect on average life expectancy of everyone (smokers and non) only factoring in firsthand health effects (since it’s really hard to get clean stats on reduction in lifespan due to secondhand smoking).
As brainspore wrote, it’s a question of both how harmful it is and how many are expose to it.


It’s (relatively) easy to measure mortality of two non-overlapping groups such as smokers and non-smokers and announce that smokers live 10 years less on average. The hard part is figuring how much of that difference is actually due to the smoking and how much is due to other risk factors that correlate with smoking. For example, smoking is correlated with higher alcohol use and is correlated with lower standard of living, which correlates with less access to medical care. The 1.9 years figure has tried to estimate how much of the 10 years is due to cigarettes as opposed to the other correlated activities.

And because it’s hard, it’s not uncommon to find vastly different numbers from different studies. As an example, a couple of decades ago there were estimates of the number of deaths in Australia caused by cigarettes that range from 12,000 to 30,000 p.a. (Though if you include estimates from the tobacco industry the lower number drops to zero.)

Allocating deaths to a single cause also gets very subjective when the correlated risk factors have synergistic effect. For example, there is a theory that tobacco causes more damage to the cardiovascular system when alcohol is present. (My reading is out of date, but last I heard I think the status of the theory was that a plausible mechanism existed, but data was inconclusive.) That is, the extra risk due to smoking and drinking is greater than the sum of the extra risks due to smoking and drinking individually.

Another thing to remember here is that life expectancy isn’t the only relevant statistics. We should also consider morbidity rates, meaning rates of disability and illness. It’s easier to study mortality than morbidity. Someone is either alive or dead, but whether someone is disabled can be more subjective. There are degrees of disability, but dead/alive is a binary status.

Personally, I feel that trying to quantify the relative effects of smoking and air pollution is an academic exercise. Both numbers are sufficiently high that we need to address both. In general I strongly support funding of epidemiological studies, but I think this is one area where we have enough data to know air pollution is a serious problem, and we need to start redirecting the funding from epidemiology to the engineering and material science research trying to solve the problem.


Okay that answered my next question because it’s weird to try to calculate the total burden & specifically leave out the secondhand effects. Personally, professionally, I’d have made some attempt to calculate the global burden of smoking, firsthand & secondhand, as there are some studies at least attempting to describe the impact.

1 Like

See also the Kanawah Valley in Charleston, WV. I suspect pretty much every state has an area like this. Everybody knows about it, no one does anything about it because it affects the poors, and the cycle goes on.

1 Like

This topic was automatically closed after 5 days. New replies are no longer allowed.