The Harvard Six Cities Study

When the EPA tightened its fine-particle pollution standard in 1997, one study did most of the heavy lifting in the science behind it. The Harvard Six Cities Study, published in the New England Journal of Medicine in 1993, was the first large prospective study to link long-term exposure to fine particulate matter (PM2.5) with shortened lives in the United States.

Background

Benjamin Ferris at the Harvard School of Public Health launched the cohort in the mid-1970s. Douglas Dockery, C. Arden Pope, Frank Speizer, and colleagues followed 8,111 adults across six cities chosen to span a range of pollution levels: Portage (Wisconsin), Topeka (Kansas), a section of St. Louis, Steubenville (Ohio), Watertown (Massachusetts), and Harriman (Tennessee). The researchers tracked respiratory symptoms, lung function, and deaths for 14 to 16 years while measuring local air pollution.

What the study found

In the 1993 paper (Dockery et al., NEJM), people living in the most polluted city (Steubenville, an Ohio River steel town) had a 26 percent higher risk of dying during follow-up than people in the cleanest city (Portage). The strongest associations were with fine particles (PM2.5) and sulfate aerosols, not with coarser dust or with ozone. Deaths from cardiopulmonary disease and lung cancer drove most of the excess.

The result was controversial enough that the Health Effects Institute, an independent research body funded jointly by the EPA and industry, commissioned a full reanalysis. Daniel Krewski's team audited the raw data, reproduced the calculations, and ran sensitivity tests against alternative statistical models and city-level confounders. Their 2000 report confirmed the original numbers and concluded the association was real.

In 2006, Francine Laden and colleagues published an extended follow-up through 1998. PM2.5 had fallen substantially in the dirtier cities over the intervening years, and mortality risk fell with it. Each 10 micrograms per cubic meter drop in PM2.5 was associated with a roughly 27 percent reduction in overall mortality risk. A later extension by Lepeule et al. (Environmental Health Perspectives, 2012) carried the analysis through 2009 and reached similar conclusions.

Why it matters

Six Cities, alongside the American Cancer Society's larger CPS-II cohort, supplied most of the evidence the EPA used to set the first PM2.5 National Ambient Air Quality Standard in 1997 (15 micrograms per cubic meter annual, later tightened to 12 in 2012 and to 9 in 2024). The cleanup that followed has been credited with adding measurable years to U.S. life expectancy in the dirtiest counties.

The 2006 follow-up was also one of the first studies to show the inverse: when air gets cleaner, people live longer, and the effect is detectable within years, not decades.

Open questions

Researchers still debate how much of the mortality signal comes from specific particle types (sulfates vs. nitrates vs. organic carbon vs. ultrafines) and whether there is any threshold below which PM2.5 stops causing harm. Recent work in the Medicare population (Di et al., NEJM 2017) suggests harm continues well below current U.S. and WHO guideline levels.

Sources

• An association between air pollution and mortality in six U.S. cities (Dockery et al., NEJM 1993)
• Reduction in fine particulate air pollution and mortality: extended follow-up of the Harvard Six Cities study (Laden et al., 2006)
• Chronic exposure to fine particles and mortality: extended follow-up of the Harvard Six Cities Study from 1974 to 2009 (Lepeule et al., EHP 2012)
• Reanalysis of the Harvard Six Cities Study and the American Cancer Society Study (HEI, Krewski et al.)
• Air Pollution and Mortality in the Medicare Population (Di et al., NEJM 2017)
• Harvard Six Cities study (Wikipedia overview with citations)