[MUSIC] My name is Miss Suzanne Condon, currently serving as consultant and advisor at the National Center for Environmental Health at the Centers for Disease Control in Atlanta. But prior to that, for some 26 or more years, I served as the director of environment health and Massachusetts Department of Public Health, where I was also the associate commissioner. During my tenure there, we conducted many cluster investigations. And so the subject that we're going to discuss today is, really, why are these types of investigations often unsuccessful in identifying an environmental cause or some other set of risk factors that might explain a disease elevation? The findings and conclusions in this presentation are those of myself and they do not necessarily represent the views of the Centers for Disease Control. And this presentation has not been revised or edited to conform to the agencies standards. So let's start with a discussion of cancer. We're going to begin the presentation with talking about what cancer is, the fact that it comprises many diseases, its common, it can take decades to development. And then we're going to move on with a discussion of two notable cancer cluster investigations that I led in the state of Massachusetts. That of childhood leukemia in Woburn, Massachusetts and looking at sarcoma in Ashland, Massachusetts, which was home to the Nyanza National Priority List Superfund Site. And then I really would like to talk at the end about why we need a nationally, state-based environmental health surveillance system that will help us improve and get beyond some of the barriers that we have when we're conducting these types of investigations. So first, let's start with, what is cancer? So cancer is really the result of a series of abnormal cells that divide uncontrollably and begin to take over normal cells or body tissue. And it destroys body tissue, and so that's how the disease process begins. While the term cancer is generally used and many people think it's one disease, it's actually more than 100 different types of diseases all generally referred to simply as cancer. This slide presents the five major causes of cancer deaths in the United States right now. So lung cancer, stomach cancer, colorectal cancer, liver and breast cancer are among those that you probably hear most about. One of the things that I think is really startling for many people to become aware of is the actual prevalence of cancer. So when you look at all of these 100 diseases together, 100 and more, it's surprising I think for some people to see or to understand that one in every three people will be diagnose with cancer at some point during their lifetime today. And so as we being to study cancer, cancer clusters and just understanding cancer in general, there are some terms that we need to familiarize ourselves with. So first, the etiology of cancer. Well, etiology is the study of causes of disease in general. How does disease develop? With respect to cancer, different cancers have different risk factors or causes. This slide presents some examples. So for example, in the case of lung cancer, tobacco smoking, exposure to asbestos or metallic dust or fumes, such as arsenic, beryllium, or chromium, as well as radon gas, can play a role in the development of lung cancer. With regard to breast cancer, DES, other estrogens used for menopausal therapy, X-ray and gamma radiation, alcohol consumption, smoking and exposure to ethylene oxide can all play a role. With regard to stomach cancer, tobacco smoking, again, exposure to radiation, an importantly, dietary factors play a role in the incidence of stomach cancer. And liver cancer, similarly, I think you're starting to see tobacco smoking plays a very big role in many of the cancers that we see today, but the environment does play a role. So exposures to arsenic, aflatoxin, alcohol, all of these other factors can contribute to one's diagnosis of liver cancer. The other thing that's important to keep in mind when we're trying to do cluster investigations is to consider the latency period. And latency is really the time between the first exposure to a cancer-causing agent and when symptoms or clinical presentation of a disease actually develops. Latency periods for many cancers may be 15 to 20 years or even longer, making investigation of cancer clusters that more difficult. And this slide presents a few examples of how that latency period or development period can vary from cancer type to cancer type. So in the case of lung cancer, it can take 20 to even 60 years before clinical presentation of the disease is established. With regard to prostate cancer, it can be 15 to 20 years. Interestingly, for childhood cancers, they tend to be considered more multifactorial, with both genetic, environmental and other factors playing a role. And so the latency period for childhood leukemia for example can be as short as 1 year or as long as 14 years. The other thing that presents a challenge to us when we're trying to investigate these reported clusters of disease is that sometimes disease elevations in a certain geographic area can really happen just due to chance. And the likelihood of chance observations is typically assessed by calculating a probability value, or p-value, and then estimating a confidence interval to help determine how confident you are that that might be a true rate. So a p-value is a number between 0 and 1 and is interpreted in the following way. So a small p-value, typically less than 0.05, indicates stronger evidence that your result didn't happen by chance alone. So what that means is that there's less than 5 chances out of 100 that that result happened purely due to chance.
