The Link Between Exposure and Disease
The investigation in Woburn evolved in an unusual fashion. There were certainly suspicions by Ann Anderson and others that something in the environment was responsible for the leukemia cases that were occurring, but their concerns were largely dismissed by town officials. The contamination of Wells G and H with TCE and PCE was discovered somewhat accidentally when an unknown person left 184 barrels of waste on a plot of land. Even just north of the wells. The barrels were discovered and removed before they had contaminated the soil, but a town engineer submitted samples of water from the wells that lead to discovery of the contamination of TCE and PCE. In fact, it was this discovery of TCE and PCE in the wells and heavy metal contamination at the site of the old Merrimac Chemical Company that prompted Reverend Young to search for additional cases. These events drew the attention of the Massachusetts environmental officials and the EPA who conducted the series of studies described previously in this module. They concluded that the incidence of leukemia in Woburn was higher than expected based on the statewide average, but now several questions remained to be answered:
- Who was responsible for contamination of the wells?
- To what extent had had the water supply been contaminated prior to 1979?
- Did the elevated concentrations of TCE and PCE in the wells cause leukemias in the children?
In 1982, Rev. Bruce Young and Anne Anderson approached staff at the Harvard School of Public Health and asked for assistance in some kind of further health study. Dr. Marvin Zelen and others responded and planned a cross-sectional interview study to see if childhood leukemia, birth defects, and other childhood illnesses were associated with exposure to water from Wells G and H. A major volunteer effort was mobilized, and over 5,000 households were contacted and queried about their health. When the results were presented in 1984, the first link to the Wells G and H was made in a scientific study. They concluded that the children who had leukemia received twice as much contaminated water as the children who didn't. This study also led to the recognition that ordinary citizens could play an important role in the design and conduct of epidemiological studies. Results of the study were published in 1986 (Lagakos SW, Wessen BJ, and Zelen: An analysis of contaminated well water and health effects in Woburn, Massachusetts. J. Am. Stat. Assn 1986;81(395):583-596.)
The authors stated,
"Our objective was not to compare these rates [of leukemia] with those elsewhere but to determine whether space-time distribution within Woburn of the 20 cases was correlated with access to water from wells G and H. Using a test derived from Cox's (1972) regression model, we found that both a "cumulative" and "none-some" metric of G and H exposure were positively associated with leukemia rates. It does not appear that Woburn's entire leukemia excess, based on national rates, is statistically explainable by wells G and H."
In the discussion, they offered the following comments:
"A difficulty in trying to determine whet in her wells G and H caused any adverse events is the lack of knowledge about the contaminants in wells G and H and the effects of these contaminants on human health. Several of the chemicals detected in 1979 have been shown to cause cancer in laboratory animals when given in high concentrations, and others found in the surrounding groundwater are suspected of causing cancer in humans. The effects on humans, however, of exposure to these chemicals at lower concentrations and in combination are not well understood. Furthermore, the types and levels of contaminants in wells G and H prior to 1979 are not known. Thus wells G and H might have become contaminated shortly before they were tested in 1979, or might have been contaminated for many years, at much higher levels and with chemicals other than those found in 1979. The EPA is currently conducting geological studies to determine the sources of the contaminants found in wells G and H and the surrounding groundwater. These studies might make it possible to predict the types and levels of contamination in earlier years. Without this information, it is unlikely that the role of wells G and H as a cause of adverse health outcomes will ever be fully understood.
A key question among citizens is whether individuals who were formerly exposed to wells G and H might still be at elevated risk. One way this can be investigated is by monitoring adverse health outcomes in Woburn over the next few years."
The animation in the video below shows a model of how TCE from 5 sources was likely to have moved in groundwater from 1960 to 1986 and ultimately contaminated Wells G and H. Keep in mind that movement of ground water was affected by the composition of the soil and other geologic materials (e.g., bed rock) in the buried valley beneath the Aberjona River, pumping of Wells G and H (which were operated intermittently as shown in the animation), and by drought conditions and flow in the Aberjona River. Note also the persistence of the TCE plumes in groundwater.
Animation by Scott Blair, Ohio State University, Department of Geological Sciences
Source: Carleton College; http://serc.carleton.edu/NAGTWorkshops/hydrogeo/activities/10688.html
The water from Wells G and H was mixed with water from other water sources in Woburn, so the degree of contaminated with TCE and PCE varied. The maps below examine the spatial relationships between TCE and PCE levels in the water supply to various parts of Woburn and the occurrence of leukemia. The map on the right is adapted from a map constructed by FACE (the citizen organization For A Cleaner Environment). The red circles identify the residence of 28 leukemia victims from 1960 to 1986. Wells G and H are shown in gray circles. The orange circles show the location of some of the contaminated sites. "R" is the Riley Tannery; "IP" is the Woburn Industri-Plex; "U" is the plant; and "G" is W.R. Grace.
What conclusions do you draw from these maps?
Note that there is a paucity of cases in the northeastern part of Woburn despite high concentrations of contamination along the eastern border. What might account for this?