Study specific criticisms include:
- The study is a report of a "genetic mutations" in babies born to ETS-exposed mothers. It is not a report of actual cancers occurring in such children.
- The study is small, only 24 newborns.
- The study looked for mutations in a gene called HPRT -- a gene that has nothing to do with cancer.
- The differences in reported mutation rates between babies born to exposed mothers and babies born to non-exposed mothers are on the order of 75 percent to 100 percent. But in genetic mutation studies, scientists usually look for orders of magnitude changes -- i.e., 1000 percent or more.
Putting aside the specifics of this study, existing epidemiology does not bear out the theory that maternal smoking (primary or secondhand) is associated with childhood cancer. A quick Medline search revealed:
- A 1986 study reported weak associations (relative risk = 1.5 for all cancers; and about 2.0 for non-Hodgkin's lymphoma, acute lymphoblastic leukemia and Wilms' tumor) between the heaviest maternal smoking and cancer risk in offspring. [Stjernfeldt M, Berglund K, Lindsten J, and Ludvigsson J; Lancet 1986 Jun 14;1(8494):1350-2]
- A 1992 Swedish study of 497,051 Swedish children born during 1982-1987 reported no association between maternal smoking and childhood cancer. The researchers characterized previous research as "inconclusive." [Pershagen G, Ericson A, and Otterblad-Olausson P; International Journal of Cancer 1992 Feb;21(1):1-5]
- A 1994 IARC study reported "The associations between maternal smoking during pregnancy and childhood cancer have been studied intensively, but there is no clear association overall, or for specific sites." [Tredaniel J, Bofetta P, Little J, Saracci R, and Hirsch A; Paedriatic Perinatal Epidemiology 1994 Jul;8(3):233-55]
- A 1996 NIH study studied 54,795 liveborn children born during 1959-1966. The researchers reported relative risks of: 0.67 (95 percent C.I. 0.38-1.17) for maternal smoking and childhood cancer; 0.82 (95 percent C.I. 0.31-2.11) for maternal smoking and childhood leukemia; and 0.60 (95 percent C.I. 0.30-1.20) for maternal smoking and childhood cancers other than leukemia. [Klebanoff MA, Clemens JD, and Read JS; American Journal of Epidemiology 1996 Dec;144(11):1028-33]
What's the lesson?
Weak study plus negative epidemiology = junk science
Comments on this posting?
Click here to post a public comment on the Trash Talk Bulletin Board.
Click here to send a private comment to the Junkman.
Copyright © 1998 Steven J. Milloy. All rights reserved on original material. Material copyrighted by others is used either with permission or under a claim of "fair use." Site developed and hosted by WestLake Solutions, Inc.
Material presented on this home page constitutes opinion of Steven J. Milloy.