July 10, 2013

expert reaction to 2 studies on air pollution and lung cancer/heart disease

A paper published in The Lancet Oncology found prolonged exposure to particulate air pollution increases the risk of lung cancer, even at levels below the European Union limit values. Another systematic review in The Lancet found short-term exposure to most major air pollutants appears to increase the risk of being hospitalised for and dying from heart failure. A before the headlines analysis accompanied this roundup.

On the lung cancer paper:

Prof Jon Ayres, Professor of Environmental & Respiratory Medicine at the Institute of Occupational and Environmental Medicine, University of Birmingham, said:

“Lung cancer has been associated with air pollution exposure for many years but this European study is a strong reinforcement of the message from the earlier US studies producing similar effect sizes and pointing the causal finger at fine particles.  The study is powerful because of its multiple cohort design with an appropriate and better exposure estimate than in previous studies.  There is now no doubt that fine particles are a cause of lung cancer and the finding of higher effect sizes for adenocarcinoma (thought to be a more common horrific type in non-smokers who develop the disease) adds internal biological credibility to the findings.

“The meta analysis on air pollution exposure and heart failure produces a message which is not new and has been reported before.  Four years ago the Dept of Health’s Committee on Medical Effects of Air Pollutants (COMEAP) produced an extensive report on the association between heart disease and air pollution, a report to which two members of this meta-analysis contributed.  It is therefore surprising that this report was not cited in this paper but not surprising that the findings are broadly similar to those reported by COMEAP.  Air pollution contributes to heart disease both through long and short term exposure and there is no doubt that reducing air pollution, particularly the fine particle fraction (the associations with nitrogen dioxide being in my view likely to be due to co-exposure to fine particles) will reduce the burden from heart disease.  Costing this is more difficult although the simple approach used by the authors will have wide confidence limits around these calculated central estimates.”

On the lung cancer paper:

Prof Malcolm Sperrin, Royal Berkshire Hospital, said:

“This is a well-designed paper looking for possible correlation between pollution concentrations and the incidence of various types of lung cancer.  Whilst it has been a reasonable supposition for some time that pollution is a risk enhancer for disease, a statistically robust correlation has been elusive and this paper does go a long way to providing such evidence even at pollutant concentrations at or below so far accepted safe limits.  The authorship of the paper is very broad and draws on expertise for many of the highest calibre organisations in Europe and the findings are therefore worthy of further attention.  I would have preferred some more information on how the patient cohort was selected, why only fruit was selected as being a potential correlate and also the statement that there is an enhanced correlation to non-smokers needs elaboration.  I suspect the answer to the latter point is that smoking is a sufficiently high risk that any further correlation to pollution is masked but such commentary could be usefully added.

“This paper is clearly a springboard for further work which is eagerly awaited.”

On both papers:

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

“Both of these papers relate a well-known and dread disease to air pollution, and both are based on observing and recording data about many people over time, but apart from that they are very different. The paper on lung cancer is looking at a possible association between long-term exposure to air pollution and lung cancer, and it found limited evidence of such an association for particulate pollution – small particles in the air arising from traffic, some industry and some kinds of domestic heating. It found no association between lung cancer and some polluting gases (nitrogen oxides), or between lung cancer and traffic levels on the roads near where people lived. But lung cancers take years to develop.

“The paper on heart failure, by contrast, looks at quick-acting effects on people who would generally already be suffering from heart disease. It investigates associations between air pollution levels and hospitalisation or death from heart failure. They found an association with several polluting gases and also with particulate pollution, with the strongest association being on the actual day when pollution levels are high. That is, if the air is particularly bad on one day, people with heart failure are more likely to have to go into hospital on that same day.

“But, even though the studies are very different in some ways, both sets of researchers have had to deal with a major bugbear of investigating associations between air pollution and disease. Oversimplifying quite a bit, poorer people tend to live and work in places where the air pollution is worse, and poorer people also tend to suffer more from many diseases than their richer fellow citizens, including lung cancer and heart disease, for a lot of reasons that don’t all relate to air pollution levels. Poor people also tend to smoke more, and smoking is strongly related to both lung cancer and heart disease. So if people in an area with high air pollution suffer more from lung cancer or the effects of heart disease, that might be because of the air pollution, or it might be because the people who live there are poorer, or smoke more tobacco, or both.

“So researchers in studies like these must try to record and allow statistically for the effects of smoking, the socio-economic status of the people being studied, and so on. They can never do this perfectly. For instance, in the lung cancer study, the analysis allowed for how much people smoked, but the researchers had data on this only at the start of the study, and some people do change their smoking habits over time. The researchers allowed for the general socio-economic status of the area where people lived, but not of the people as individuals. They did some clever statistical modelling to get good data on pollution levels where people lived, but not where they worked. Even with studies like these involving hundreds of thousands of people, there remains considerable doubt about how big the effect of air pollution is. Indeed, in the lung cancer study, after all these statistical adjustments, most (but not all) of the relationships between air pollution and lung cancer that they studied were not statistically significant, that is, they could be explained by chance variability between people, so that we can’t be certain there is really a relationship between lung cancer and many of the air pollutants at all.

“That said, both studies do provide some evidence that air pollution is associated with particular diseases. In my view, the evidence in relation to heart failure is considerably stronger, but the evidence in relation to lung cancer shouldn’t be ignored. There is already evidence from many previous studies that air pollution increases the risk of several other diseases, including heart attacks (as distinct from heart failure) and several lung diseases (apart from lung cancer). These new studies simply add more diseases to the list of the possible effects of air pollution. And air pollution, of the kind studied here, doesn’t just affect the people working in one factory or one industry, but potentially affects everyone living in the area where the air quality is bad. So the potential public health gain from improving air quality could be very great.”

‘Air pollution and lung cancer incidence in 17 European cohorts: prospective analyses from the European Study of Cohorts for Air Pollution Effects (ESCAPE)’ by Ole Raaschou-Nielsen et al., published in The Lancet Oncology on Wednesday 10 July, 2013.

‘Global association of air pollution and heart failure: a systematic review and meta-analysis’ by Anoop S V Shah et al, published in The Lancet on Wednesday 10 July, 2013.