Personalising the Health Impacts of Air Pollution: Interim Statistics Summary for a Selection of Statements

Dimitris Evangelopoulos, Klea Katsouyanni, Heather Walton and Martin Williams Environmental Research Group, King’s College London


The public are understandably interested in the size of the effect of air pollution on health and, in particular, the risks to them as individuals, or at least to individuals like them. Typical statements about the impact of air pollution on health have been communicated in terms of numbers of premature deaths or life years lost. This is mainly because (i) it is assumed that people are most concerned about the most severe endpoints and these usually have a dominant influence on costbenefit analysis (ii) most places collect mortality statistics routinely, there are a lot more studies for this endpoint (iii) there are also more studies on all-cause mortality because it is a clearly defined endpoint without confusion as to whether a disease has been diagnosed correctly (iv) overall impact on the population as a whole is the output of interest for public health practitioners.

As a measure of population impact, an input into cost-benefit analysis and a general headline for the media, the above types of statements remain influential. However, the acknowledgement of the risks of air pollution and the motivation to change behaviour may be increased by summary statements with which individuals can more easily identify. Life years can seem a rather abstract concept and deaths may seem too distant in time for many in the population. So, there is a role for summary statements on more common adverse health effects of air pollution and in particular groups (due to susceptibility or travel behaviour).

The aim of the project is to develop statements on the effects of air pollution on health outcomes that may be more familiar to the public or specific groups of the public than life years, life expectancy or numbers of deaths. A supporting aim was to ensure a clear route from simple statements, understandable to a wide range of the public, to the detailed technical justification for the numbers quoted. This document is one link in the chain of evidence from simple statement to full technical details. The latter will be available in a full research report, currently in preparation.

There are many scientific studies on the effects of air pollution on a wide variety of disease outcomes but their conclusions are written for scientists rather than the public, and it could be difficult for a member of the public to judge its quality or put a particular study into context. There are, however, documents that pull together consensus positions on the evidence including Committee reports (e.g. COMEAP, 1998; WHO, 2013; US EPA various dates) and also systematic reviews/meta-analyses (e.g. Brook et al 2010; Mills et al, 2015; Hoek et al 2012). Meta-analyses pool quantitative information across studies so are useful to give a sound basis for estimating the size of the air pollution effect. Finally, there are health impact assessments which have quantified effects in particular places or for particular policy scenarios. Some of these only cover mortality (e.g. EEA, 2018) and some are becoming outdated (COMEAP, 1998) but other more recent publications do cover disease outcomes (Holland 2014; APHEKOM 2011; Walton et al 2015). Their methods vary substantially, and they are not necessarily written in language accessible by less specialist readers. The work in this project is based on expert Committee positions, pooled results across several studies or large studies that themselves cover results from many different places. The project did not do its own pooling of the most up to date data, its own review of the latest evidence on causality (if we understand how a pollutant has a particular effect, it strengthens the evidence), or detailed mapping of air pollution exposures at a small spatial scale. (This would have required a very long project given the range of outcomes addressed!). Nonetheless, it is substantially more robust than statements based on single studies or by non-experts, relying on past detailed work by others.

Full Report


© King's Environmental Research Group. All rights reserved 2016.