How Do Scientists Know 9,500 Londoners Died From Air Pollution?

Sarah Jayne Bell
By Sarah Jayne Bell Last edited 16 months ago
How Do Scientists Know 9,500 Londoners Died From Air Pollution?

Figures from King's College London estimate that in 2010 up to 9,416 deaths in London could be attributed to exposure to air pollution. But how can scientists quantify the number of people who've died from air pollution?

Air pollution in London April 2015. Photo by David Holt from Londonist Flickr pool.


The report uses a new method developed by the World Health Organisation (WHO). This is why the number has risen from the previous estimate of around 4,300 — scientists are getting better at estimating deaths, rather than more people actually dying. The main change is the inclusion of deaths due to poisonous NO2, while previous studies included only deaths due to fine particulates, called PM2.5.

But how can they put those deaths down to air pollution? Epidemiologists convert air pollution concentration to number of deaths by using 'concentration-response functions'. These are statistical formulae produced by the WHO for use in Europe. They are based on hundreds of studies of the impact of pollution on health and mortality.

In the early days of the epidemiology of air pollution, counting deaths due to hideous air pollution events such as London's 1952 killer smog was relatively straightforward. The air was like pea-soup and a lot of people died. Estimating the longer term health effects of less severe air pollution, in the mix of everything else that affects our health, is more difficult.

Epidemiologists use long term data to correlate changes in air pollution with changes in the number of deaths and number of hospital admissions. Since the 1980s, thousands of studies have been done around the world on how air pollution impacts human health and lifespan. The WHO reviewed the best of these studies to derive its formulae for converting air pollution concentrations to deaths and hospital admissions. The researchers at King's College then used these formulae to calculate how many deaths in 2010 were due to NO2 and PM2.5 pollution in London, producing the figure of 9,416.

For more information, and to see the figures, you can read the report here.

Last Updated 23 January 2017

Greg Tingey

Better, much better, but still a little "wobbly".
I suspect the error-bars are still quite large.
Incidentally & in the opposite direction, the numbers usually quoted for the great smogs of '52 & '55 are UNDER-estimated or understated, which always puzzles me.

Maddforit Maddforit Maddforit

an estimation is still just a guess and does not reflect reality a bit like man made global warming