Lasting legacy of the Six Cities study into harms of air pollution

Photograph: Justin Lambert/Getty Images

This month marks 30 years since the publication of research that completely changed our knowledge of the harm from air pollution. Known as the Harvard Six Cities study, its impact still resonates in public health actions including London’s ultra-low emission zone and the tightening of EU air pollution laws.

Prof Dan Greenbaum, a former president of the US Health Effects Institute, said: “In the wake of the serious air pollution incidents of the 50s, 60s and 70s, many people thought we had solved the problem. But then the Six Cities study came along and found strong relationships between air pollution and mortality at much lower levels.”

The origins of the study can be traced to the oil embargo and price shocks that followed the 1973 Arab-Israel war.

Prof Doug Dockery, of the Harvard TH Chan School of Public Health, became the Six City study’s principal investigator in 1988. He recalled the circumstances that led to the start of the study. “The price of oil on the international market doubled almost overnight. Anticipating a degradation of air quality over the next decade, a call was put out for a long-term study of health effects associated with a switch from oil to coal for power production,” he said.

In 1974, six US cities were chosen for the study. They each had different concentrations of air pollution. More than 8,000 people were randomly selected to take part and they were then contacted annually for up to 16 years.

Dockery said: “Every year we sent a postcard to each participant asking ‘has the person named below died? No or yes’. The plan was to conduct mortality follow-up for a sufficient period and then conduct statistical survival analyses.”

The team also collected air pollution data from each city. There was a 51% difference in rates of people who died between the best and worst city. Each participant had been interviewed to find out about their age, diet, occupation, smoking history and other things that could affect their long-term health. These factors accounted for over half of the differences in death rates but there was still a 26% difference in survival. When plotting this difference against air pollution, a clear straight line was seen. This was especially clear for small particle pollution, now termed PM2.5.

Dockery said: “We showed that long-term exposure to air pollution had much more significant effects on mortality risk than previously believed. It changed the dynamics of the debate.”

Two years later, a further study was published that also showed association between death rates and long-term exposure to low levels of air pollution.

The Six Cities study was plunged into controversy from the moment it was published and this increased when new air pollution regulations were proposed to protect public health.

Dockery explained how the tightening of regulations was firmly resisted. “The American Trucking Association, various private companies and the states of Michigan, Ohio and West Virginia filed lawsuits challenging the Environmental Protection Agency’s authority to issue these standards. These legal challenges ended up at the US supreme court,” he said.

Greenbaum said: “The controversy swelled and began to revolve around the fact that no one else had had access to these data. There were protests and full-page ads reading ‘Harvard, release the data!’.”

The researchers were in an impossible situation since the medical information had been collected in confidence. There were also concerns about protecting the scientists from the types of industry-led attempts at character assassination inflicted on those who had exposed the dangers of lead additives in petrol.

The solution was an independent reanalysis. It was funded by the US Congress and convened by Greenbaum’s team at the US Health Effects Institute.

Prof Anna Hansell, of the University of Leicester, said: “The reanalysis supported the original analysis and findings. In fact, they went further, to show that air pollution effects were greater in deprived individuals, which is something we still don’t pay enough attention to.”

In the past three decades, many studies have confirmed the results of the Six Cities study, including one on more than 60 million people. These have led to air pollution guidelines from the World Health Organization and underpin the frequently quoted estimates of about 4,000 early deaths yearly in London due to air pollution, 29,000 to 43,000 across the UK and more than 400,000 in the EU.

Prof Barbara Hoffman, of the University of Düsseldorf, said: “Now, after decades of high-level research, follow-up analyses show that the strongest reductions in air pollution lead to the greatest health gains.”