Last week we reviewed the origins and rationale for the Long Island Breast Cancer Study. Per the dictates of the congressional bill that funded us, it was required to be a population-based case-control study. In such a study, you compare cases of breast cancer to controls for exposures. The case group consists of all histologically proven cases of the disease that occur in a given population, in our case Nassau and Suffolk Counties, over a given time frame, August 1, 1996-July 31, 1997. We included both in-situ and invasive breast cancer, and all cases underwent pathologic review by a single pathologist, ensuring consistency. Physician consent was required before a patient could be contacted; this was followed by a recruitment letter and follow-up letters and phone calls. Ultimately a compliance rate of 82.1% was achieved among cases (235 DCIS, 1273 invasive = 1508 total).
To identify these breast cancer cases, I visited and established connections at all of the 28 hospitals on Long Island as well as at 10 hospitals in the five boroughs that we identified where substantial numbers of Long Island residents went for their medical care. This necessitated that we obtain institutional review board approval from each of these hospitals, which would permit them to notify us immediately of the diagnosis of any patients from Long Island with breast cancer. Our goal was to do rapid ascertainment, to be able to contact new patients within a month or so of diagnosis.
In the meantime, any epidemiologist will tell you that the more difficult part of a case-control study is to properly identify appropriate controls. In this particular instance, we chose to utilize a mechanism known as random digit dialing. For each breast cancer case, we would take her 10-digit telephone number and randomize the last two digits. These calls would give us households within a few blocks of the index case. We would call each household in order and inquire as to the presence of a person in the household who matched our case within five years of age and in race/ethnicity. If there were none, we would move on to the next random numbered household until we found a match. The compliance rate with joining our study was an absolutely extraordinary 77.9% of those eligible controls contacted (N=1556). How many of you say yes to an anonymous caller who asks you to participate in a survey? Those were the good old days. One cannot do random digit dialing anymore since landlines have gone the way of the dodo bird.
Interviewers visited the homes of each case and control and conducted two-hour interviews. In addition, blood and urine were collected from each subject; dust, soil, and water samples were collected from each home. We measured the electromagnetic forces (EMF) coming from telephone and power lines over homes as well as the EMF coming from electric blanket use and their potential effect on breast cancer risk. Countless exposures were assessed by our investigators in relation to breast cancer incidence and survival.
As I sit here today, almost 30 years and 140 published papers later, what can I say we found? Mostly I can say what we did not find. There was little or no evidence to tie DDT or other pesticides or polychlorinated biphenyls to breast cancer. There was some suggestion of an increased risk of breast cancer with increased exposure to polycyclic aromatic hydrocarbons, but it was inconsistent and not terribly strong. Most other dietary and smoking or alcohol risk factors were also not terribly important.
So, in the end, why was there an elevated risk of breast cancer on Long Island? Probably for the same reason that the risk of breast cancer is similarly high in the 10021 zip code of the Upper East Side—breast cancer is a disease of the White Jewish upper socioeconomic status group and that is who inhabits Nassau and Suffolk counties. If you ask what puts them at risk, it generally reflects the well-known reproductive hormonal risk factors—early menarche, late menopause, late age at first pregnancy, fewer pregnancies, etc.
Our study was, in my view, extremely well done and high quality. Certainly, however, many scientists continue to believe that environmental factors play a significant role in breast cancer etiology. Certainly, the women on Long Island, despite our study, continued to adhere to that belief.
Alfred I. Neugut, MD, PhD, is a medical oncologist and cancer epidemiologist at Columbia University Irving Medical Center/New York Presbyterian and Mailman School of Public Health in New York. Email: [email protected].
This article is for educational purposes only and is not intended to be a substitute for professional medical advice, diagnosis, or treatment, and does not constitute medical or other professional advice. Always seek the advice of your qualified health provider with any questions you may have regarding a medical condition or treatment.
