Tobacco Documents Online

459 Statement of Katherine McDermott Herrold, M.D. My name is Katherine McDermott Herrold. I am a certified pathologist and am presently retired from my most recent position as medical director of the United States Public Health Service. I received my medical degree from Women's Medical College of Pennsylvania in Philadelphia in 1948. From 1948 through 1949 I interned at George Washington University Hospital in Washington, D.C. Between 1952 and 1955 I was the chief medical officer at the Federal Reformatory for Women in Alderson, Va. Between 1955 and 1957 I was a resident in clinical pathology, and be- tween 1957 and 1959 I was a resident in pathological anatomy, both of these positions being at the National institutes of Health in Bethesda. Between 1959 and 1971 I was a laboratory pathologist at the National Cancer Institute in Bethesda, Maryland. I am a member of numercus professional societies including the American Society of Clinical Pathologists, the College of American Pathologists, the International Academy of Pathology and the American Association for Cancer Research. I am a member of the honorary medical society of Alpha Omega Alpha. I have published papers in the scientific literature dealing with pathology and cancer research. 95-077 0-82-30

460 Those who assert that cigarette smoking causes lung cancer rely on the data contained in several prospective studies, including one whose data was first presented by Dr. Harold F. Dorn in 1958. That study reported a statistical association between smoking and lung cancer. However, Dr. Dorn recognized the grave danger of relying upon death certificates for lung cancer diagnoses without pathological confirmation. It was my privilege to work with Dr. Dorn on this project in providing the clinical and pathological review of the data. The Dorn study involved a total of almost 300,000 U.S. veterans who were policyholders of O.S. Government life insurance. I reviewed the clinical and pathological materials available from the 2,241 patients with the diagnosis of lung cancer on their death certificates.. Histologic material was available for review in about 651 of those cases. Extremely important from a biological standpoint is the fact that only a small percentage of even heavy cigarette smokers develop lung cancer. The frequency of the histologic types of cancer was also established. Of the 472 patients who were "current cigarette smokers" at the time of their deaths, and for whom histologic sections were available for my review, I found no correlation between the various histologic types of lung cancer and the amount of tobacco smoked. Further, the age at death from lung cancer was 461 not related to the age at which smoking ste years smoking, or the number of cigarettes These findings were in agreement with tt some ten years earlier. Dr. Passey wrote t without hesitation that the age at death fr that which is associated with chemical carc findings, as well as those of Doll and Hill to mine - that the age of diagnosis of lung to the age at which the patient started to time smoking, or the amount smoked. Those who make claims against smoking ass, dose response relationship between smoking . is, the risk of developing lung cancer incrt cigarettes smoked and the earlier one begin: the risk. My findings, like those of Dr. Pe otherwise. In recent years, there have been reports "epidemic" of lung cancer, which some have a tobacco consumption. However, scientists ha apparent increase may well reflect changes i rather than in the actual incidence of the d diagnostic problem is exacerbated by reason among smokers, i.e., the fact that smokers u -2- V, i -3-

461 not related to the age at which smoking started, the number of years smoking, or the number of cigarettes smoked per day. These findings were in agreement with those of R.D. Passey some ten years earlier. Dr. Passey wrote that it could be said without hesitation that the age at death from lung cancer is not that which is associated with chemical carginogenesis. Passey's findings, as well as those of Doll and Hill, were almost identical to mine - that the age of diagnosis of lung cancer is not related to the age at which the patient started to smoke, the length of time smoking, or the amount smoked. - Those who make claims against smoking assert that there is a dose response relationship between smoking and lung cancer; that is, the risk of developing lung cancer increases with the number of cigarettes smoked and the earlier one begins to smoke, the greater the risk. My findings, like those of Dr. Passey before me, suggest otherwise. In recent years, there have been reports of a so-called "epidemic" of lung cancer, which some have attributed to increased tobacco consumption. However, scientists have noted that this apparent increase may well reflect changes in the detection rate rather than in the actual incidence of the disease. The diagnostic problem is exacerbated by reason of "detection bias" among smokers, i.e., the fact that smokers undergo a more ?i

462 rigorous examination for lung cancer than non-smokers. Studies reporting that lung cancer occurs more frequently in smokers than in non-smokers all assume that the non-smokers are studied with an equal frequency and intensity as the smokers. However, Professor Alvan Feinstein of the Yale Medical School- found this not to be the case at various institutions, including the Yale- New Haven Hospital. Hence, he concluded in 1974 that "cigarette smoking may contribute more to the diagnosis of lung cancer than it does to producing the disease itself." The various anomalies found my myself and others in the population studies combined with the problems associated with diagnosing lung cancer, lead me to the conclusion that more research is needed in this field before we can accept as proven the theory that smoking causes lung cancer. -4- To the Committee: IN£ This is my statement r Distingu NAR: SnN 415/6 463 UNIVERSITY OF S