This Article  • References  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Similar articles in JAMA  Social Bookmarking   What's this?

James H. Stebbings, ScD; Henry F. Lucas; Richard E. Toohey, PhD
Argonne National Laboratory Argonne. III

JAMA. 1986;255(7):902.

	Since this article does not have an abstract, we have provided the first 150 words of the full text PDF and any section headings.

To the Editor.—

The article by Lyman et al¹ associating radium in groundwater with leukemia has major flaws. Ecological correlation studies are considered useful explorations of disease causality only where data from cohort (or case-control) studies are lacking. Consistency of correlations over time and space and the effects of socioeconomic and other variables as confounders, and also as "control" variables, are investigated whenever possible. Statistical significance levels cannot be taken literally when spatial autocorrelations exist. Lyman et al¹ appear unaware of this.

Two further weaknesses appear in their article. The difficulties, inherent in ecological correlation studies, in linking environmental agents to actual population exposures will not be pursued herein. But the authors' failure to discuss the bone sarcomas and carcinomas of nasal sinuses and mastoid air cells, long known to be induced by radium and present in far larger numbers than are leukemias in the radium-exposed cohorts they . . . [Full Text PDF of this Article]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?