PU580 Unit 9 Cancer Cluster Letter Example

Eleven Cancers Is Real and the Cluster Analysis Cannot Tell You Why: A Unit 9 Risk Communication Piece Written for Parents Who Counted the Cases Themselves, With the Four Different Diseases Their Eleven Includes

[Student Name]

Purdue University Global

PU580: Environmental Epidemiology

Unit 9 Risk Communication Piece

[Instructor Name]

[Date]

The street, the closed plant and the eleven cases are composite; the cluster investigation method, the statistics and the research cited are real.

The Letter

To the families of Rowley Street We have completed the review you asked for. This letter explains what we found, what we could not find out, and what happens next. We are not going to tell you that your count was wrong, because it was not.

What You Counted, and What We Confirmed

You reported eleven children diagnosed with cancer among families who have lived on or near Rowley Street since 2009. We confirmed ten of those eleven in the state cancer registry. The eleventh was diagnosed in another state and the record has been requested. What we also found, which you did not have access to, is what the ten cases were. They are four different diseases: three acute lymphoblastic leukemias, two brain tumors of two different types, three lymphomas and two solid tumors unrelated to each other. That matters for the question you asked, and we explain why below.

Why Four Diseases Changes the Question

Childhood cancer is not one illness. A chemical exposure that causes leukemia does not generally cause brain tumors, and no known exposure causes all four of these at once. When a group of cases includes several unrelated cancers, it usually means we are looking at the ordinary pattern of a common group of diseases in a defined place rather than at a single cause. That is not a reason to dismiss what you found. It is the reason our review cannot conclude that one source produced these cases, and we would be misleading you if we said otherwise. If the eleven had been eleven leukemias, we would be writing you a different letter. They are not, and we think you should know exactly why that matters.

What the Statistics Say, and Their Honest Limit

We calculated how many childhood cancers would be expected among the population of your area over fifteen years, using statewide rates by age and sex: 7.8 cases. You have 10 confirmed. That is higher than expected and it is within the range that chance produces: the probability of 10 or more when 7.8 are expected is about 23 percent. The limit of that calculation is real and we want to state it rather than hide behind it. Any boundary we draw changes the answer. Your street alone has fewer expected cases and so looks more unusual; the surrounding neighborhood has more and looks less so. Because clusters are usually noticed first and bounded afterward, around the cases, the arithmetic tends to overstate how unusual they are. Reviews of this kind across the country have rarely identified an environmental cause, and this is the main reason why (Kingsley et al., 2007; Thun & Sinks, 2004).

What We Found About the Plant

The electroplating plant operated from 1961 to 1998. Its permitted discharges included chromium and nickel compounds. We reviewed state files and found two violations, in 1983 and 1991, both for wastewater discharge rather than air emissions. We also did something the file did not contain. We sampled soil at eight locations on your street and in the two yards nearest the fence line, and we sampled three private wells still in use. Chromium in soil at the fence line was elevated above background and below the level at which we would recommend any action in a residential yard. The wells were below drinking water standards for every metal tested. Those results are attached in full, including the ones we are describing as reassuring, because you should be able to check them.

What We Cannot Tell You

We cannot tell you what any individual child was exposed to between 1961 and 1998, because nobody measured it. We cannot rule out that exposures then were higher than what we measured now. And we cannot tell any one family whether their child's cancer was caused by the plant, which is the question most of you actually asked us. No study of any size can answer that for an individual.

What We Are Doing Next

1. Obtaining the out-of-state record to complete the count.

2. Keeping Rowley Street under annual registry review for five years, and telling you the result each year whether or not it changes.

3. Asking the environmental agency to complete a formal site assessment, which carries authority ours does not, and sending you the request and the response.

4. Holding a meeting at the library on the second Tuesday of next month, where the epidemiologist who did this work will answer questions. The soil and well results will be on the table.

Why This Letter Is Written the Way It Is

A note on the letter itself, for the record. We have led with your count rather than with our statistics, given you the soil and well results in full, and said plainly what we cannot answer. That is deliberate. Concern about a hazard follows whether it was chosen, whether it is understood and whether somebody else imposed it, far more than it follows a calculated probability (Slovic, 1987). A plant you did not choose, chemicals you cannot see and a file you were never shown score badly on all three, and no arithmetic from us will change that. What can change it is that you hold the same numbers we do.

One Request

If you know of a family who lived on the street and has moved away, please tell us. Our review can only see people the registry can find at an address, and the cases we are most likely to have missed are the ones who left.

References

Kingsley, B. S., Schmeichel, K. L., & Rubin, C. H. (2007). An update on cancer cluster activities at the Centers for Disease Control and Prevention. Environmental Health Perspectives, 115(1), 165-171. https://doi.org/10.1289/ehp.9021

Slovic, P. (1987). Perception of risk. Science, 236(4799), 280-285. https://doi.org/10.1126/science.3563507

Thun, M. J., & Sinks, T. (2004). Understanding cancer clusters. CA: A Cancer Journal for Clinicians, 54(5), 273-280. https://doi.org/10.3322/canjclin.54.5.273