So, I'd like to move on a little bit and talk about some of the investigations that I've been involved with and notably two of them where we actually had some luck in identifying associations with environmental factors that we believe played a strong role in cancer occurrence. So, the first one involved investigating childhood leukemia in Woburn, Massachusetts. In fact I might note that this was the very first study I ever worked on as a new researcher at the Massachusetts Department of Public Health in the early '80s and so, I learned through this. You'll see that the title of this slide says Childhood Leukemia Follow-up Study and that's because after the first study was done, many many questions remained. So, the State Health Department had to do several investigations to follow that. So, what do we know about Woburn, Massachusetts? Well, we know that for a long time, Woburn was very very friendly to industry, in terms of being a home for numerous types of industries. So, between the late 1800s and the early 1900s, they were tanneries across the city and there was also other leather-related industries and chemical industries. Greenhouses were very prevalent as well and in fact at one point in time, the city of Woburn was referred to as the greenhouse capital of the country. In 1964, drinking-water wells G and H were constructed. Those two wells were constructed because Woburn was becoming a community that was close to Boston, lot of residential development, people wanted to live in Woburn, Massachusetts. In 1979, hazardous waste was discovered near wells G and H. There were actually two hazardous waste sites one called Industrial plex where they dumped a lot of hazardous waste and that site wasn't too far away from wells G and H and it had an impact on those wells as well. A mixture of contaminants were discovered in the wells at that time and by the middle of 1979, steps were taken by the State Department of Environmental Protection to close the Woburn wells G and H. So, what was presented to the State Health Department at the time by a group of resonance was what appeared to be an unusual number of children that had been diagnosed with childhood leukemia. I might mention that the early days of Woburn in those first studies were actually the subject of a major emotion picture, as well as a book both titled "A Civil Action" and that story focuses on the families of Woburn in their struggles in the courtroom to try to find justice. But what we were presented with at the State Health Department was this map. So, as you can see, Woburn is a city that was composed of six census tracts. I might mention that in the northeast corner of the city where you see those six figures, that's northeast Woburn and that's where Wells G and H were located. So, at the time, there were 12 children reported with childhood leukemia in Woburn, Massachusetts diagnosed between 1969 and 1979, the majority of which were concentrated in a single census tract. So, a study was done at the time. Essentially the conclusions of that study that were released in 1981 documented that the incidence of childhood leukemia was statistically significantly elevated in Woburn as a whole. There were 12 children diagnosed with leukemia while slightly more than five would have been expected. The majority of access cases were male, six of the cases were diagnosed while residing in that single census tract in Woburn. The hypothesis though suggesting that the increase in leukemia incidence was associated with environmental hazards in Woburn specifically as it related to contaminated drinking water was neither supported nor refuted by the findings. As you might imagine, that did not sit well with the Woburn community. The State Health Department is telling them, "You're right. You have far more children diagnosed with leukemia than we'd expect to see, but we don't know what caused it. We're really not sure if the environment played a role." So, several years later I should say that the State Health Department kept a watch on Woburn. During that period of time, different evaluations of environmental data were done, a study looking at reproductive outcomes or birth outcomes was done, all sorts of things trying to see if researchers could figure out whether or not the environment played a role in impacting health of the Woburn residents. As you can see on this map, by 1986 that number 12 grew to 21. But what was unusual and what you can see here is that those cases continued to look slightly more odd in that Northeast section of the city, but then you started to see cases occurring in other census tracts with larger numbers as well. So, at one point in time, one had to ask whether the cluster of leukemia was really specific to that Northeast section of Woburn or was the entire city considered a cluster. So, at that time, myself along with others designed a follow-up study. We wanted to take another look at leukemia. We had developed tools from other studies that had been done in Woburn and thought well let's try to really focus in a different way. So, our objectives for the follow-up study involved interviewing the original childhood leukemia families, all 12 of them, as well as families of the new children, nine new children diagnosed with leukemia. We wanted to use a comprehensive water distribution model that was developed from that birth study that I mentioned that would help to determine what the generation of well water exposure data looked like. We also collected comprehensive information on occupational exposures of the parents, residential history, where else did the children live, reside, spent time at. We took a detailed health history collecting information for the affected and non-affected families. If you will, collecting information both cases and the control population. We also collected information on electromagnetic fields. The reason that we did that is because EMF exposure at that time was beginning to emerge as something that people felt might be having an impact on health. The ultimate goal was for us to assess the risk of childhood leukemia in relation to the opportunity for exposure to contaminated water from wells G and H. So, we worked with a fellow named Peter Murphy. Murphy helped us to develop this water distribution model and he developed an exposure index that was based on the fraction of time per month that contaminated water reached a user area, and the fraction of water supplied that came from the contaminated wells. This next diagram gives you an idea of what's called the user demand areas in Woburn and I would be remiss if I didn't mention my close colleague Elaine Krueger who passed away a number of years ago. But, it was Elaine that helped us test the Murphy model. So, Murphy's model estimated by guessing how much water based on the pumping of the Wells went from one area to another. But we didn't have a way to validate Murphy's model and I was sitting at my desk one day and Elaine came to my office and knocked at the door and she said "Susanne, I have a way to validate Murphy's model." I said, "What do you mean Elaine. How are we going to do that? We can't turn the wells back on with a contaminant." She said, "Woburn's water supply isn't fluoridated. We can add fluoride to the water, and pump the wells and the rest of the community and then we'll be able to establish whether or not Murphy's predictions for how much water went to a given area based on the pumping rates is actually valid." I couldn't begin to tell you if it wasn't for a Elaine coming up with that idea, I'm not so certain that we would have been able to demonstrate that this model worked. So, what did we find with the follow-up study? So, when we looked at the relationship between risk of leukemia and level exposure to mothers during pregnancy, we saw that those mothers that never received any water from the contaminated wells during pregnancy had a relative risk of one, basically no risk. Those mothers that had some amount of exposure but certainly not a lot had an increased risk of having a child diagnosed with leukemia about three and a half times the rate we'd expect. Those mothers who received the greatest amount of contaminated water had a risk 14 times greater of having that child develop childhood leukemia over time. I want to mention one thing here, we talked about P-values and confidence intervals earlier. In this particular case, neither of these elevated rates were statistically significant. The 14 times greater risk came close to achieving statistical significance. However, it was only when we did a test for trends that we saw statistical significance and p-value less than 0.05. So, when we looked at the trend for increasing exposure over time during the entire period that was investigated, that particular trend toward increasing consumption of contaminated drinking water was statistically significant. So, our conclusions of that study based on everything that was done was that the risk of developing leukemia was related to the amount of contaminated water received at the residence of the child's mother. Mother's exposure to contaminated water while pregnant showed the strongest association between water exposure and the development of leukemia in her offspring. Interestingly, the child's exposure to contaminated drinking water from the time of birth to diagnosis of leukemia showed no association to leukemia risk. I should mention I mentioned the birth defects study, the reproductive outcomes study that was done earlier on when we looked at the risk of having adverse reproductive outcome based on a child's consumption of water, they actually had a greater likelihood of having a normal birth weight if they drink more water. So, clearly there was no elevated risk to the child simply from the child themselves drinking the water. As with any epidemiologic investigation however, we caution that due to the small number of study subjects, our conclusions had to be interpreted cautiously.
