Reader is shocked by cancer information Editor's note: This letter to Doneen Hollingsworth, South Dakota secretary of health, was written Oct. 11 by Dr. Gregory Schalliol of Vermillion, the father of a childhood cancer survivor.
Dear Ms. Hollingsworth:
In response to the concern of some Vermillion residents that cases of childhood cancer in our community seem too high, you recently wrote a letter printed in the local Vermillion newspaper (Plain Talk, September 10, 1999) which contained a long list of reasons why your office declined to investigate the matter further. When I read your letter, I was shocked to see the false, out-of-date, and irrelevant evidence your office purportedly used to reach its conclusion. To make matters worse, two weeks after your letter appeared, the editor of our local paper (Plain Talk, September 24, 1999) blindly regurgitated the false, out-of-date, and irrelevant evidence from your letter to accuse the local citizen who first raised the issue of "innocently misusing much of the statistical information she has gathered ?", such that "her stats, I fear, could easily be found to have little foundation upon close scrutiny." The case is, sadly, quite the reverse, as the following analysis of your purportedly authoritative data shows.
You first point out that the State Epidemiologist, Dr. Sarah Patrick, "has spent considerable time and effort reviewing the cancer data available for Clay County." That sounds admirable, until one reads exactly what data Dr. Patrick has apparently consulted. First, you say that Dr. Patrick "has had telephone interviews with a parent of each child with cancer for whom correct contact information was provided ?" That statement is, in fact, a lie. On August 19, 1999, Dr. Patrick's office received a fax that included the names of all recent cases of childhood cancer in Vermillion, along with current telephone numbers and addresses. My daughter's name was on that list, along with our correct and current phone number and address. To date, neither my wife nor I have been contacted in any way by Dr. Patrick or anyone else in your or her office. That does not sound like very thorough investigating to me, not to mention the blatant dishonesty of the statement itself. How many others did Dr. Patrick ignore?
Next, you assure us that Dr. Patrick "has reviewed all cancers among children younger than 20 years in Clay County as reported to our Cancer Data Collection System between 1992-1996." That's nice, but I thought our concern was with recent cases. The last time I looked, the current year was 1999. Why are you quoting data covering such a short time period in the past? Four of the seven most recent cases were diagnosed after 1996, so the data you refer to will mislead by omitting most of the recent cases causing our concern. Or are you telling us implicitly that your office cannot be expected to respond to data until it becomes at least three years old? Moreover, how do you know that even the data in your out-of-date system is complete? Since reporting of childhood cancer cases is not mandatory in our state, as it is in many others, how can you even rely on that data?
This question, in fact, leads me to the next assurance you give us. Dr. Patrick, you say, "has talked with medical school faculty and practicing Vermillion physicians and nurses to determine if any childhood cancers had not been reported to the system." What about those Vermillion parents who take their children to doctors elsewhere? Since we have only one pediatrician in town, many parents (my wife and I included) visit pediatricians elsewhere. Moreover, when my daughter's cancer was finally diagnosed correctly, it was in Minneapolis, not in South Dakota, because our South Dakota pediatricians were unable to figure it out. Hence, Vermillion health care providers cannot be a completely reliable source for knowing of all cases of childhood cancer in Vermillion. And before you chastise me for not being loyal to South Dakota doctors, may I remind you that our governor also seems to prefer Minnesota health providers for severe health problems. Many of our local kids who got cancer eventually received treatment out of state. Are those out-of-state providers required to report South Dakota cases they treat to your data collection system? No. The state's Cancer Data Collection System, therefore, is hardly a reliable source to consult, even with the very limited efforts you claim have been made to bring it up to date in this case.
Next in your letter, you assure us that Dr. Patrick "has worked closely with the Centers for Disease Control and Prevention ?" which has developed "extensive, scientifically based guidelines" for use in cases like this one. You then quote an excerpt from those guidelines. (http:www.cdc.gov/epo/mmwr/preview/mmwrthml/00001797.htm) which notes that few, if any, requests for investigations ever lead to an actual investigation. That may be true, but it certainly does not lesson your obligation to treat the request from Vermillion seriously. It also does not explain why you did not mention from the very same section the "breakthroughs and triumphs in infectious disease control ?" which actually came about because of further investigation of reported disease clusters. Moreover, the "Stage 1" phase of the very guidelines you have presumably followed call for a rather extensive procedure to be used when gathering initial data from relevant persons when a perceived cluster is reported. No one in your office followed that procedure for my daughter's case, so what good does it do to invoke "scientifically based" guidelines when you do not follow them closely yourself?
The central part of your letter lists the "Department of Health analysis of available data" which contributed to your "scientific" decision against further investigation of the Vermillion cases. Let's look at those points one at a time. First, you point out that cancer represents several diseases, not just
Continued on page 5A
one, and that the seven children in Vermillion had six different types of cancer. It is true that the CDC guidelines suggest that "a variety of diagnoses speaks against a common origin." But since no one yet understands very well what causes these types of cancer, no one can say whether we are talking about a single disease with different consequences or completely independent diseases.
The next three statistics you offer, though, again mislead. You say "a review of Clay County child deaths since 1985" did not show anything unusual. What, indeed, does the number of deaths have to do with the number of cases of childhood cancer? As of today, only one of our seven local cases has ended in death. Are you trying to perpetuate the myth that anyone who gets cancer dies from it? Or do you mean to say by using this statistic that your office, supposedly a Department of "Heath," only takes seriously citizens who end up dead? I am sure my daughter � who endured four major surgeries, lost a kidney, and suffered through nine months of hellish chemotherapy on a cancer ward � would feel very comforted that her state Department of Health thinks so little of her disease since she did not die, not to mention what state employees and taxpayers might feel about the almost $750,000 her treatment cost them.
Next, you say "a comparison of overall cancer incidence rates ? found Clay County cancer incident rates for 1992-1996 were lower than the South Dakota rates, which were lower than the U.S. average." But I thought the issue here was child cancer rates, since many of the cancers we are dealing with here occur almost exclusively in children. By lumping together children and adults in this statistic, you again mislead. In addition, you once again quote data that excluded all cases from the last three years.
The last part of your "in-house" analysis is "a comparison of overall cancer mortality ? in Clay County for 1992-1996 ?" This supposedly relevant statistic really takes the cake! Again, at issue is whether recent cases of cancer in kids here warrant further investigation. What is the relevance to that question of (1) out-of-date data which (2) lumps together numbers for both adults and children and (3) only includes the people who have died? I did not need my college statistics course to see that this statistic is irrelevant in three ways. And yet you provide it to us as a reason for no further investigation?
Would it not be better to deal with the concerns of Vermillion citizens by dealing with the present facts of the situation rather than trying to assure us with a smoke screen of out-of-date, incomplete, and irrelevant data? Below, in the chart, are the cases we are concerned about. They are all children, not adults. They are not all dead. They cover the period from 1992-1999, not 1992-1996. You are welcomed to confirm all of these cases again, even though Dr. Patrick already received a list from us. On the face of it, these data are clearly disturbing. In a small town with less than 1400 children under the age of 15, any one of these cases is extraordinarily improbable. Even if you do treat each cancer type as a separate disease, the two cases of Ewing's sarcoma are surely extraordinary. According to the National Cancer Institute's best and most recent estimates, the incidence of all forms of cancer in children under 20 nationwide in 1995 was 150/million. Even if one considers only the six most recent cases in Vermillion, that makes the cancer rate for children under 15 in our community approximately 28.57 times the national average. That rate is much greater than he rates that already triggered an extensive investigation of the well-known children's cancer cluster reported in Toms River, New Jersey (http://tr-teach.org/). The above data are not from different or conflicting sources. The incidence rates, for example, come from the publication cited above which can be reached from a link at your own South Dakota Department of Health web site. The local cases can be confirmed by talking with the parents involved, who could even refer you to the physicians responsible for treatment.
Could this be just an accidental coincidence of several diseases having the same general name? Yes, that is possible. But how will we know without a more persuasive examination of the data than what you have provided in your letter? What, for example, will happen if more cases arise, even if they, too, happen to be different types of cancer? Will there be any number of cases that would cause your office to reconsider the need for an investigation? Are you reluctant to investigate because you don't have enough money, or because you have no idea where to look for a cause? If one or more of those is the reason you chose not to investigate further, why did you not say so? Why did you, instead, mislead our local newspaper editor and his readers with authoritative looking statistics and assertions which, in fact, are false, out-of-date, or plain irrelevant so that he ends up injustifiably criticizing a local, concerned citizen?
Given the quite limited nature of your department's own information in this matter, it seems to me you should be thanking profusely the concerned citizens of Vermillion who have brought your data base up to date, even though they are not getting paid for it. You also owe them, I think some constructive account about what quantity of disease would warrant an investigation. That way, you at least sound more sincere in your concern about the children in Vermillion than the substance of your previous letter demonstrates. You might also explain to everyone why South Dakota was the only state in the entire United States in 1997 not to receive any funding from he National Program of Cancer Registries for updating and improving state cancer data collection (http://www.cdc.gov/nccdphp/dcpc/npcr/reg98.htm). Has your office done anything to respond to that dubious distinction? For the Vermillion cases, perhaps you should consult someone who specialized in he epidemiology for childhood cancer. In case we have no such expert in South Dakota, I would suggest the one in the Pediatric Hematology/Oncology Department at the University of Minnesota Hospital in Minneapolis.
Lastly, you should also not forget that since many people in our community are state employees through The University of South Dakota, any child of a state employee who gets cancer probably will cost state employees and taxpayers several hundred thousand dollars. For every such case your office might prevent, you would save the State of South Dakota a lot of money, and surely more than it would cost your department to do a little better job in responding to reports of suspected cancer clusters.
We realize the danger of creating unnecessary fear about this matter, but you only make fears grow when you try to assure us of no danger with statistics that seem designed to mislead.
Dr. Gregory Schalliol
Associate professor of
classics and philosophy,
University of South Dakota,