It is difficult to ignore air pollution nowadays. Not because thick smogs descend on the city in wintertime, causing coughing and sore eyes as they did in the 1940s and 1950s, but because we read about it every day in our newspapers. We read of thousands of deaths, of children being carried choking into hospital, of threats to our hearts and even brains, and of the negligence of governments and councils in failing to control it. We are told that the risk involves exposure to air containing tiny invisible particles and a gas – nitrogen dioxide – and it is strongly urged that diesel vehicles are a large part of the problem. These messages are worrying, most notably to the mothers of young children, but also to the rest of us as we contemplate our cardiac and cognitive futures. And, like all curious individuals, we should ask, is it true?
I must first declare an interest. For the decade to 2002 I chaired the UK government committee that researched and recommended all the air quality standards adopted by our government and that subsequently became EU standards. All our recommendations were published and justified in short reader-friendly documents. Moreover, I was the person primarily responsible for the concept that the harmfulness of air pollution came mainly from very small, ultrafine or nanoparticles, which is now widely accepted. I thus have an interest in informing the public on the risks from air pollution and, importantly, in pointing out the uncertainties in the science. This is important, since the public in general still believe what they read in newspapers, and misinformation has the effect of causing anxiety where none previously existed. Most doctors are familiar with the anxious who come to our surgeries or clinics grasping a copy of the Daily Beast.
How do we know what air pollution does?
Our understanding was originally derived from simple observation, best exemplified by John Evelyn's 1661 petition to King Charles II entitled 'Fumifugium', in which he described the coal-burning sources and their perceived effects on health and well-being in London. Nothing was done about it until the late 1950s, following several notable episodes of pollution. The first was in the valley of the Meuse in Belgium in 1930, when 60 deaths from chest problems occurred over a few days, together with the deaths of many cattle. Investigations showed that coal-burning factories were responsible, together with a winter anticyclone that trapped the cold polluted air close to ground level. It was predicted by the investigators that a similar episode in London would cause some 3,200 fatalities.
Another episode occurred in 1948 by the Monongahela River valley in Pennsylvania, with 20 deaths and widespread symptoms of chest illness reported among the local population. Then came the great London smog of 1952, when again a temperature inversion trapped the smoke from factories and houses at ground level and at least 4,000 excess deaths occurred in the city over two weeks. It was this episode (which coincided with similar ones in other British cities, as I can testify) that led to the first UK clean air act of 1956.
These observations of association produced sufficient evidence that air pollution can kill people and put them in hospital, to persuade government to take preventive action. They also stimulated scientific research to answer the obvious questions: what is it in air pollution that causes harm and how much air pollution is necessary to cause these effects? This research took two forms – toxicology (the science of poisons) and epidemiology (the science of the health of populations) – both, as the names suggest, looking back as far as the ancient Greek philosophers. In the UK, the research concentrated more on the former, investigating the relative effects of gases and soot, whereas in the USA the most important advances came from epidemiology.
In the 1960s, the particles themselves were considered relatively harmless, comprising mainly carbon, whereas the sulphur dioxide derived from coal was of known toxicity. Since the source of the both pollutants was the same, action on coal-burning was very effective and the air concentrations of both soot (measured then by colourimetry as British black smoke) and sulphur dioxide fell steadily over two decades to levels considered safe.
In contrast, in the USA researchers looked at populations and started to relate deaths to levels of air pollution in different cities. They concluded that the relationship was still present at what were considered very low levels; for example, in the London smog, particle concentrations were measured in a few milligrams in a cubic metre of air, whereas the US researchers found associations with deaths at concentrations below
100 micrograms per cubic metre, less that a tenth of the smog levels and at a concentration widely present in our cities at the time. And it became apparent that there may well be no threshold below which effects could occur. In other words, in any population there are some people who are on the edge of a precipice owing to some susceptibility such as age, smoking, coronary artery disease, or genetics, and that pollution provides the final straw.
The early studies showed that the excess deaths occurring in pollution episodes were attributable almost entirely to worsening of chronic lung diseases and to heart attacks, and afflicted older people, especially smokers. This led to a conceptual problem; while the association with advanced lung disease was easily explained, how could such minute doses of particles composed of carbon, inhaled into the lungs over a day or two, kill people from heart attacks.
The plausibility of the epidemiology was thus challenged until we proposed that, rather than the mass of particles, it was the number
that was critical in causing the reaction; the lung reacts to the thousands of tiny particles inhaled in every breath as though they are invading bacteria. The more particles, the stronger the response – an inflammatory reaction in the lungs – and this is transmitted round the body by chemical messengers in a way that prepares the body's defences. It is now known that one such mechanism is the passage through the blood of small pieces of damaged cells known confusingly as microparticles. Thus, an inflammatory reaction can be initiated in distant organs such as the arteries to the heart or even brain. There is now sufficient evidence to believe that this hypothesis is at least partly correct, and that inflammation caused by inhaled nanoparticles is the basis of air pollution effects.
The complexity of air pollution
Air is polluted by gases as well as particles and particles themselves may differ greatly in chemical and physical characteristics. The common gases – nitrogen dioxide and sulphur dioxide – are unlikely to cause lung inflammation at the concentrations now prevalent in the UK. However, epidemiology still shows associations of cardiac deaths and some respiratory illnesses with the former. This is widely accepted in the media as evidence that current levels of nitrogen dioxide pose a serious threat to the health of children. If the association is causative, it must therefore be mediated by an unknown mechanism; this is possible but unsupported by toxicological evidence. More likely, the association is due to confounding, since the combustion sources of nitrogen dioxide also produce equivalent numbers of nanoparticles.
From a regulatory point of view, assuming this relationship holds, reducing NO2 will reduce nanoparticle numbers correspondingly, so it may not matter save in one respect. Burning gas in kitchens is a potent source of NO2, leading to concentrations much higher than occur in UK streets. It also produces very high numbers of particles, yet kitchens are not (yet) recognised as places of sudden cardio-respiratory death. This is probably explicable; low concentrations of NO2 are not toxic and the particles from gas combustion differ from those produced by petrol and diesel combustions, which brings me to the differences between particles.
Particles are conventionally measured by weighing the mass of them in air drawn through an orifice that selects those below a certain size, as particulate matter less than 10 or 2.5 micrometers in diameter (PM10 and PM2.5). This says nothing about what they are. Collected in a school assembly hall, for example, they are likely to comprise mainly dust and fibres derived from clothing. In the London Underground they comprise largely particles of iron from rails, brakes and wheels with very few combustion nanoparticles. In the street they usually comprise particles from vehicle combustion with some sea salt and debris from the action of wheels on roads. But all the evidence we have on their effects on human health relates to particles derived from combustion, primarily now by petrol and diesel vehicles.
It is ignorant or dishonest of scientists to claim that the effects of particle matter are similar wherever they are measured. Most of the evidence points to a specific toxic effect of vehicle combustion particles from chemical contaminants on their surfaces, specifically metals and complex organic chemicals that can initiate inflammation in the lungs.
What is known about the effects of pollution?
There is no scientific doubt that outdoor pollution by particles can cause excess deaths and hospitalisations from heart attack and chronic lung disease, sometimes in previously asymptomatic people. The risks to any individual are low but increase with age, smoking and other known risk factors for these diseases; because the risks apply to a large population, large numbers of people are at a low risk, but the large majority are not at risk even when pollution is high. However, there is good evidence that the cumulative effect of exposure over a lifetime will make individuals increasingly susceptible to the acute effects later in their lives, so continuous reduction of particulate pollution will continue to have public health benefits. So will NO2 reductions, though probably because nanoparticle numbers will also be reduced by the same action on sources.
It is very unlikely that ambient air pollution is a cause of childhood asthma in the UK. Indeed, prevalence studies have shown the highest proportion of asthma is in the least polluted areas such as the Highlands and Western Isles of Scotland. However, exposure to high pollution can certainly cause worsening of asthma and initiate attacks. Most evidence on healthy children has shown little or no effect on respiratory health, and even severe pollution episodes in London in the late 20th century had no effect on hospitalisation in children.
Whether air pollution can cause other effects is still open to question. The evidence that it increases the risk of stroke and clotting problems is strong and plausible. There is some evidence that it increases the risk of cognitive decline in the elderly and that it might impair development in children, but at present this is far from certain. Dementia has some established environmental risk factors, most of which coincide with vascular risk factors; smoking, obesity, lack of exercise and, now, air pollution. I regard this as plausible, but through a vascular mechanism. Suggested effects on the foetus are possible but unlikely; many other risk factors such as poor maternal diet, obesity and blood pressure, may confound any relationship and most observations seem to rely on passage of pollution particles to the placenta, which is unlikely in the large numbers required to initiate a reaction. If there is a causative relationship, it is likely to be through passage of inflammatory messages.
What do we need to do?
The reduction on air pollution in the UK since the 1950s may be regarded as a public health triumph, and this has occurred despite an exponential increase in the use of motor vehicles. However, concentrations of particles and NO2 have now reached a plateau and further reductions require restriction of individual liberties. It is important to remember that we
are responsible, not the government, and it is our
thoughtless use of personal transport, be it petrol or diesel, that is largely responsible.
The health effects of further reductions on urban pollution in the UK would be small but significant across a range of diseases, but there is another, and in my view overriding reason for prohibition of fossil fuel powered vehicles: climate change. All combustion produces carbon dioxide and, as I show in my book, 'Farewell King Coal', we are already witnessing the effects in wild fires, desertification, floods, storms and migrations. Urgent action by all of us is necessary; getting rid of our addiction to oil and its products is a major part of it.