Tom Chivers

Tom Chivers is a science writer. His first book, The AI Does Not Hate You is out now.

June 27, 2019

Everyone is freaking out about air pollution at the moment. Particulates, nitrogen oxide, sulphur dioxide, ammonia. Diesel fumes. Buses on Oxford Street. Schools in London apparently handing out backpacks that measure air pollution, like Geiger counters on recovery workers after a nuclear disaster. Is it giving you cancer, is it giving you cardiovascular problems, are women producing fewer eggs, is it making you lose your memory?

And it’s getting worse, apparently. The “death risk from London’s toxic air” – steady on, now – “sees ‘utterly horrifying’ rise for second year running”, shrieked the Evening Standard a week or two ago. According to the paper, the “rate of fatalities linked to breathing in killer particles went up from 6.4 per cent to 6.5 per cent in 2017”, having gone up from 5.6% to 6.4% in the previous year.

Which sounded very strange to me. Because, firstly, that figure seemed very high. If one person in every 15 died from air pollution, I think we’d know about it. Second, I thought that the air was, in general, getting cleaner.

Luckily for me, a new paper is out in the journal Environmental Research Letters which helps make sense of it all. I popped along to a briefing on it on this week.

Suggested reading
Why more people die as the economy grows

By Peter Franklin

The paper looked at the health impacts of air pollution with a snapshot every 10 years, starting in 1970 and ending in 2010. It found that for the main pollutants – sulphur dioxide (SO2), nitrogen dioxide (NO2), and particulate matter smaller than 2.5 micrometers (PM2.5) – things are a lot better.

Emissions have consistently dropped for SO2 and PM2.5, while emissions of NO2 rose slightly until 1990 and have dropped to well below 1970 levels since. Only about 55% of UK air pollution is directly linked to UK emissions (the rest is mainly from nearby countries), but still, the same pattern is visible. This pattern is not true for ozone, which got worse until 2000 and has declined a bit since, but that’s a smaller contributor.

And it has had an effect on public health. Using the same metric that the Evening Standard’s story used, the “percentage of fatalities linked to” PM2.5 in 1970 was about 12; in 2010 it was more like five. Similar drops are seen in deaths linked to NO2 and SO2.

The paper links this, convincingly in my mind, to air pollution control policies, in the UK, at European level, and internationally, such as the 1997 National Air Quality Strategy and the European vehicle manufacturing standards that were issued in the 1990s. Stephan Reis, one of the authors, pointed out that, in the 1970s, German manufacturers used to build cleaner cars to export to California, where the regulations were stricter, and sell more polluting ones in the domestic market.

Still, though, 5% of all deaths caused by particulate pollution alone sounds pretty bad, right? And that’s only one contributor, albeit the biggest. When you add in NO2 (about 3%) and ozone (about 2%) it gets even worse – 10% of all deaths caused by air pollution!

Suggested reading
Trust conservatives to save the planet

By Giles Fraser

Except, of course, you can’t add them up like that. That’s not how it works. What they’re reporting is the ‘attributable fraction of mortality’, or AF. The AF is a statistical measure that describes “the proportional reduction in population disease or mortality that would occur if exposure to a risk factor were reduced to an alternative ideal exposure scenario”.

Insofar as I understand, it’s essentially relative risk, but in reverse – it’s saying, if you took this factor away, cleaned the air perfectly, how much would the annual mortality rate go down by? Which is an interesting question, but it’s not the same as “this causes 5% of all deaths”.

The risks overlap. So, if you added all the possible contributing factors of death together (smoking, ageing, obesity, diabetes, stress, car crashes, murder, meteorite strikes, Godzilla, whatever) it might add up to many thousands. It’s a very widely used and accepted metric, but I really don’t like it, because I think that when you give percentages, they should add up to 100. Otherwise people will hear “air pollution kills 6% of people”, which I’d say it definitely doesn’t. It contributes to 6% of deaths, but that’s not the same thing.

According to Sotiris Vardoulakis, one of the authors of the Environmental Research Letters paper, the effect of air pollution is roughly equivalent to, on average, a six-month reduction in the national average lifespan. Last year, a professor of respiratory medicine told BBC More or Less that in London, that figure is about nine months. Which is quite a big deal! And, of course, it won’t be evenly spread: some will see smaller effects, while some, especially those with preexisting lung conditions, are likely to suffer much worse.

Suggested reading
$10 a barrel for oil? The age of the electric car is closer than you think

By Michael Liebreich

But, for context, the AF caused by smoking in 2005 was 19% (27% among men). That’s not among the 25% of Britons who were smokers – that’s for the whole population. Obesity and alcohol are both significantly larger contributors as well. Vardoulakis pointed out that air pollution is the largest environmental factor, and it’s certainly not negligible, but “London’s toxic air” is getting a bit carried away. I’m concerned that over-fretting about it will stop people from doing things which definitely are good for you, like cycling, because they’re worried about particulates or nitrogen dioxide.

And there’s another point. This is a good news story! Legislation to reduce air pollution appears to have worked. People have had to bear some part of the burden of that regulation, either through increased costs or through bans on products, but it appears that we are all significantly healthier because of it. If we then all kick off and complain that our air is toxic and we’re all dying of the black lung – which we’re not; this is not the Great Smog of 1952 – then people might, understandably, feel that we’ve been wasting a lot of time and effort regulating all this stuff.

Air pollution is bad. That’s fine, we can all agree about that. And we all want a future world of electric cars powered by offshore wind, and London air that smells like primrose and freshly cut grass rather than piss and diesel. And maybe we’ll get it one day. But in the meantime, let’s have a little perspective: things are getting better, we’re all living longer, and it’s not going to kill you if you go outside and go for a run.