A Tale of Six Cities

by John F. Lauerman

Where did it all begin? Did it arise out of a deadly London smog that killed 4,000 people in the space of a few days in early December 1952? Could it have sprung in part from the long-standing interest in pulmonary physiology sparked by polio research? Was it Energy Crisis-induced anxiety over the impact of burning increased amounts of high-sulfur coal in a period of high-priced oil? Or a sense among 1970s scientists and policymakers that the science base for American air pollution standards was insufficient?

Whatever the reasons, in late 1973, Professor Frank Speizer found himself driving to Logan Airport while then-Department of Physiology head James Whittenberger read a proposal so fresh (in that pre-personal computer era) that some reordered paragraphs literally had been cut and pasted together. The proposal Whittenberger conveyed to Washington, composed by Speizer and his colleague, Benjamin Ferris, described what was to become one of the most influential, innovative, and longest-running experiments concerning the health effects of air pollution in America: the Harvard Six Cities Study.

The Six Cities Study was an expansion of Ferris’s pioneering effort to bring modern laboratory techniques and tools into people’s homes, both for sampling air quality and for monitoring health effects. For the first time in this country, public health researchers directly correlated environmental exposure to toxins with health impacts. The work, which continues today, set a standard for research in air pollution that is emulated throughout public health and around the world.

As the study’s principal investigator, Ferris was an important part of this hands-on research philosophy. A pediatrician by training, Ferris had been recruited by Whittenberger to what was then the DEPArtment of Physiology from Boston Children’s Hospital in 1948 because of Ferris’s interest in pulmonary function in children with polio. Prior to the invention of the polio vaccine, Ferris had instructed physicians in caring for paralytic polio victims using the iron lung.

"He had a very good background," recalls Whittenberger, now at the University of California, Irvine. "And I didn’t know how hard a worker he was at that point, but he proved to be a very thorough scientist."

When the Salk and Sabin vaccines all but extinguished the polio threat, Ferris expanded his horizons. On a 1957 trip with Whittenberger to a coal-mining region of Wales, he saw entirely new techniques being used to study the impact of air quality.

"They started out by selecting a probability sample of coal miners, and once they determined the sample size, they made an effort to get everyone in the group examined," Whittenberger recalls. "Previously, this kind of study would have been a matter of examining a number of people at certain exposure levels; that kind of data couldn’t be generalized to a larger population. So this was a new concept, and a major methodological improvement."

Impressed and energized, Ferris returned home where he immediately began a similar study not far from his wife’s summer home, in the New Hampshire paper mill town of Berlin. Initially, the Brown Paper Co. refused Ferris’ request for access to systematically sample air quality and monitor worker health in its factory. Ferris, reasoning that pollutants were still coming out of the smokestack, began studying the community surrounding the mill. When he demonstrated the adverse health impact of sulfur dioxide emissions–[including lowered lung function]–the paper company agreed to reduce them.

Ferris had been bitten by the field epidemiology bug, and he applied his own exacting standards to its execution. But when the Environmental Protection Agency (EPA) instituted a large-scale study of air pollution’s health effects in the mid-1960s, Ferris was one of many scientists who was highly critical. The chess (Community Health and Environmental Surveillance Systems) Study was an attempt by EPA to provide scientific justification for the air quality standards set by the 1967 Clean Air Act. However, Ferris and other critics in academe and industry charged that the study put the conclusion cart before the raw data horse. Specifically, exposure thresholds that had been set by the EPA for airborne toxins appeared to have limited correlation to the study’s observations.

"The criticism had to do with data collection problems, machine calibrations, training, and the use of standardized questionnaires," recalls Speizer, now Professor of Environmental Science, Edward H. Kass Professor of Medicine at Harvard Medical School, and Co-director of the Channing Laboratory. "And then, on the analysis side, some very funny things that none of us thought could possibly be there were forced on the data to fit the EPA’s threshold models. I don’t think there was any true malicious intent to defraud, but it was a very difficult problem."

Senate hearings followed, and before long the National Institute of Environmental Health Sciences (NIEHS) was founded to assume the EPA’s biomedical research burden. While the discredited chess study’s findings were never released, strong national interest in air pollution persisted. Large amounts of high-sulfur coal were being mined for electric power production, and it appeared that even more coal might be burned in the wake of the Arab-Israeli War of October 1973, which severely damaged US relations with many oil-producing countries in the Middle East and played a role in precipitating the late-1970s Energy Crisis.

A key research objective was to gain more knowledge about sulfur dioxide; many attributed the deadliness of the London smog and another pollution disaster that occurred in 1948 in the mill town of Donora, Pennsylvania, to this chemical.

"The big thing was to decide whether it was sulfur dioxide or smoke," that had precipitated the deaths, Whittenberger says. "That’s basically what the Six Cities Study was supposed to do."

Then-NIEHS director David Rall believed that such an important study should not be done in a government laboratory, but in a university setting with support, peer review, and funding from NIH. When he asked whether Ferris and Speizer would perform the study, Whittenberger assumed that the answer would be a polite "no."

"I didn’t think they would be interested in such a large undertaking," said Whittenberger. The study as proposed called for at least 10 years of work, and it would be three years until any descriptive data could be published, five years or more before any papers concerning the impact of air pollution would be ready. As a career move, it was a gamble.

"I suppose we were a bit naive," says Speizer, "but we assumed it was worth doing, and it never crossed Ben’s or my mind that we wouldn’t take a shot at it."

Speizer and Ferris soon discovered that a national air quality network they had hoped to use had been dismantled. This, however, turned out to be a blessing in disguise, since it allowed them to conceptualize and design their own monitoring systems and measurements, with the help of other faculty from the School.

"I was an assistant professor just starting out," says John Spengler, S.M.’73, now Professor of Environmental Health and Director of the Environmental Science and Engineering Program. "I was doing some work on the exposures of commuters in Boston, and Ferris asked if I would play a role."

Working on the 13th floor of Building 1, Spengler and his team of doctoral students and young engineers designed and built a variety of personal air monitoring systems and home sampling equipment. A key challenge was keeping the disruption to a minimum, particularly for those subjects who were asked to wear personal air samplers.

"Essentially we put the batteries, electronics, and pumping system into a box about the size of a Tupperware container and added some shoulder straps," Spengler says. "It was very quiet, and it had to be because people took them to church, work, their bedrooms, everywhere they went."

Soon after coming to the School in the Fall of 1974, epidemiologist Douglas Dockery became involved in the study. His days were spent traveling to the study sites–Watertown, Massachusetts; Portage, Wisconsin; Topeka, Kansas; Kingston/Harriman, Tennessee; St. Louis, Missouri; and Steubenville, Ohio–setting up air pollution monitors in people’s homes and doing personal exposure measures.

"It has always amazed me how interested and cooperative people were," says Dockery, now Associate Professor of Environmental Science and Epidemiology, "We would go into people’s homes and set up equipment; we would have all kinds of boxes and noisy pumps, and they were still very happy to participate. There were people who would drill holes in the side of their houses just to make our sampling easier, and they would stay with the program for a year or more."

But no researcher or subject was more enthusiastic about the study than the principal investigator himself. Benjamin Ferris–who died in 1996–was a man of Renaissance proportions: a physician, a photographer, an enthusiast of food and wines, and, by all accounts, a gentle and caring person.

"He was the calm one," recalls Speizer, "and I was the excitable one. I would come into his office and yell and scream about problems, and after we talked for a while, everything seemed all right."

Ferris’s accomplishments as a mountaineer alone would have made him a memorable figure, and an annual report he produced on fatal mountain-climbing incidents contributed greatly to safety in the sport. He frequently combined his climbing ability and respiratory research when he single-handedly carried his 55-pound spirometer–an instrument for measuring lung capacity–to the top þoor of Watertown tripledeckers in order to test study subjects. His dedication to the study led him to personally interview thousands of the more than 20,000 children and adults enrolled in the study.

"He made a point of going to each of the cities once a year to interview subjects, working incredibly long hours far into the night for a couple of weeks," says Dockery, who accompanied Ferris on many field trips around the country. "His forte was getting people to participate. First, he would administer the questionnaire and talk with people; then, after he had gained a rapport, he would ask them if they would mind doing one more little test, and bring out the spirometer. I’m sure that if he started out with that huge piece of equipment, many of them never would have participated. He was very gentle and everyone really enjoyed working with him."

As Whittenberger had predicted, it was years until the Six Cities Study produced data on the health impact of air pollution. When it did, however, the results were different from what anyone had expected. First of all, the study showed that indoor air quality was far more important to overall health than outdoor air; most people simply spend much more of their time indoors, and pollutants often become concentrated inside homes and workplaces. The study also indicated a strong, positive correlation between levels of air pollution and mortality: deaths from lung cancer, pulmonary disease, and heart disease were 26 percent higher in Steubenville–a city so polluted that air-born sulfates turned lead-based housEPAint black within a year–than in Portage, the cleanest city in the study.

However, rather than indicating sulfur dioxide as the culprit emission, the Six Cities Study directed attention to so-called particulate matter, now recognized as a dangerous form of air pollution. Specifically, the study highlighted the hazards of so-called pm 10 particles, which are 10 microns in diameter and smaller. This lethal relationship between particulates and cardiovascular mortality has been replicated in studies performed in cities around the world.

Following the release of the study’s results in 1993, says Dockery, the American Lung Association brought suit against the Environmental Protection Agency to update its standards on particulate matter pollution, which had not been reviewed in accordance with the Clean Air Act in well over five years.

"The quality and quantity of the data were both very impressive," says Mort Lippmann, a New York University Medical Center professor of environmental science who served on the EPA’s Clean Air Scientific Advisory Committee. "This was the largest epidemiologic study of the effect of pollution on individuals that was available. It was done very carefully over a long period of time, and it had to be taken very seriously."

Six cities study data was also cited in the EPA’s standards for nitrous oxide and for ozone. At a recent School of Public Health symposium, current NIEHS director Kenneth Olden praised the study for laying a firm scientific foundation for regulatory policies. Numerous foreign countries have consulted the study when reviewing their limits for particulate matter pollution. The study developed important research tools that went into the hands of public health researchers everywhere. Biostatistical innovations–such as longitudinal data analysis, compensation for missing data and "smoothing" techniques–as well as air sampling techniques that were developed for the Six Cities Study are now used in a variety of research settings. In the laboratory of Joseph Brain, S.D.’66, Chair of the DEPArtment of Environmental Health and Cecil K. and Philip Drinker Professor of Environmental Physiology, dust from the study’s filters has been used in animal models to study pulmonary disease. Conversely, these animal studies have suggested further modifications in sampling and data analysis in the Six Cities Study.

"At the outset it was expected to be 12 years," says Dockery, "and that we did do. But what was surprising was how many offspring and spinoff studies of the Six Cities Study there are."

The Five Cities Study and Twenty-Four Cities Study have examined the long-term effects of acid aerosols on schoolchildren in the United States and Canada. All three studies continue to yield data that strongly inþuences the direction and content of American air quality policy, including the possible adoption of standards that may restrict emissions of smaller pm 2.5 particles.

"The debate on lowering the pm standard is based on this environmental epidemiology program, and I don’t think it’s overstating the case to say that the Six Cities Study data has been one of the bedrocks of the Clean Air Act," says Brain. "The research that has been done by the Ferris’ disciples has spawned a method and approach that’s been used all over the world, and all of us are the beneficiaries."