select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to air pollution and risk of dementia

Reactions to a cohort study published in BMJ that show evidence of a positive association between residential levels of air pollution across London and diagnosis of dementia.

Prof Martie Van Tongeren, Professor of Occupational & Environmental Health, University of Manchester, said:

 “I think the press release accurately reflects the science. The conclusions are backed up by solid data as the research is making use of a large database collected through GP practices and using an air pollution dispersion model.  The authors highlight some of the main shortcomings of the study, including the level of under-diagnosis of dementia and Alzheimer’s disease in general practice and on death certificates.  The use of only one time period for air pollution modelling don’t allow for analyses of what time periods of exposure are critical in the development of neurodegenerative disease.  However, the conclusions are valid and backed up by the data.

“There is a growing body of evidence of the link between air pollution and brain health, including dementia and Alzheimer’s.  This study adds to this body of evidence and fits with some of the previous studies.  Subsequent studies should focus on identifying critical time periods of exposure that are of importance to health effects in later life (ie is this cumulative exposure across the whole life, or is exposure in certain periods in life more important than others in terms of brain health?).

“Some of the important confounders are adjusted for.  Information on occupation and indoor pollution levels (eg from passive smoking) may also play a role in the development of neurodegenerative disease.  However, the CPRD doesn’t contain any information on these factors, so the researchers could not test or adjust for these exposures.

“This study provides further evidence of the adverse health effects that are caused by increased levels of air pollution.  As most people in the UK live in urban areas, exposure to traffic-related and other air pollutants is ubiquitous.  Hence, even a relatively small increase in risk will result in a large public health impact.  I agree with the authors that burden of disease of dementia and other neurodegenerative disease is substantial and growing.  Hence, action is required by authorities to reduce air pollution levels as well as increase our understanding of the exact role of air pollution in relation to brain health.”


 Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

 “This isn’t the first study to report a possible link between relatively high air pollution levels and a small increase in dementia risk. But it’s a good, well-conducted study that adds to what is known. That said, one of the things that the study clearly demonstrates is how difficult it is to research these things, and that’s why the study inevitably had several limitations and why it’s not straightforward to interpret its findings.

“The main problem with this kind of research is that people who live in areas with high air pollution differ, on average, from those who live in places with cleaner air in several ways. So if those in the more polluted areas are more likely to be diagnosed with dementia, that could be directly because of the pollution, or it could be because of one or more of these other differences. It’s possible to adjust the comparison for such differences, but only up to a point – it can be done only for differences that the researchers have data about. One thing that has been found to affect people’s chances of being diagnosed with dementia is their socio-economic position, and a potential issue here is that the researchers did not have data on individual people’s socio-economic position and instead had to use a measure of deprivation for areas containing about 1500 people on average. In several parts of London, people who are relatively rich and relatively poor live close to one another and one of these areas might actually contain people in very diverse circumstances. But at least the researchers could adjust for socio-economic status to some extent – there may be other influential factors that they could not take into account because they had no data. This is why the researchers, correctly, say that the study can’t establish whether air pollution causes dementia. Maybe it does, alongside many other factors, but maybe it doesn’t.

“Another issue is that the researchers couldn’t actually measure the levels of air pollution that the people they studied were individually exposed to. They did use pretty precise methods of finding the average pollution levels outside their home address, but that doesn’t take into account exposure to pollutants at work, or inside the home, or how much time they spent outside actually using polluted roads. Again this is a common limitation of research on the effects of air pollution on health. It doesn’t in itself invalidate the results at all, but it can make them harder to interpret. And the researchers point out clearly that they can’t be sure the same patterns would be observed outside London.

“Finally, it’s worth pointing out that, in the middle-aged and elderly people studied, the risk of being diagnosed with dementia, in the average period of seven years for which they were studied, was not very high. Under two in every hundred had a dementia diagnosis during the study period. The clearest associations between air pollutant levels and dementia were for comparisons between the areas with the top fifth and bottom fifth of pollutant levels. Evidence of higher dementia risk for people living in less polluted places than the worst fifth was generally not statistically significant. Even for the top fifth, the fact that the great majority of people still didn’t get dementia means that there remains considerable uncertainty about exactly how big the increase in risk is.

“Overall, this good study has moved the science on considerably, but there’s still much uncertainty about whether dementia does play a role in causing dementia, and if so, exactly what that role is, how it fits in with other factors causing dementia, and exactly how much the risk of dementia is increased in polluted areas.”


Dr Stefan Reis, Head of Atmospheric Chemistry and Effects, Centre for Ecology & Hydrology (CEH), said:

“This important study highlights the wide range of negative effects air pollution has on human health, beyond increasing respiratory and cardio-vascular health risks. It is worth noting, that the increased health risks associated with air pollution have been identified in this study at ambient concentration levels below (PM2.5, currently at 20 µg m-3) or around (NO2, currently at 40 µg m-3) UK air quality limit values* (for annual mean concentrations). These  findings illustrate the need for integrated research into the public health burden arising from air pollution, even at low ambient concentrations.”

* Source.

All our previous output on this subject can be seen at this weblink:

 Declared interests

Prof Martie Van Tongeren: “I sit on the scientific advisory group to the International Carbon Black Association, which is an industry association of the carbon black manufacturing industry.  The fees are paid to the University, however.  I was also a member of the expert panel of the health effects institute (HEI – in the US, to evaluate the epidemiological evidence of the link between diesel engine emission and lung cancer.  The HEI is part funded by EPA and part funded by the motor vehicle industry.  Again, fees for this activity were paid to my employer at the time (Institute of Occupational Medicine).”

Dr Stefan Reis: “No competing interests to declare.”

Prof Kevin McConway: “Kevin McConway was the lead author of a chapter on Measurement and Communication of Health Risks from Pollution for the 2017 Annual Report of the Chief Medical Officer ‘Health Impacts of All Pollution: what do we know?’. He is a member of the Science Media Centre’s Advisory Committee.”


in this section

filter RoundUps by year

search by tag