Perchlorate in human milk: separating the science from sensationalism

Citation metadata

Author: M. Jane Heinig
Date: May 2005
From: Journal of Human Lactation(Vol. 21, Issue 2)
Publisher: Sage Publications, Inc.
Document Type: Editorial
Length: 1,743 words

Document controls

Main content

Article Preview :

In February of this year, an article was published on the Web by researchers at Texas Tech University titled "Perchlorate and Iodide in Dairy and Breastmilk." (1) The article received significant press coverage in the United States, resulting in concern among some public health professionals about the safety of breast milk among US women. The headline of one of the Web-based articles read "Scientists Find High levels of the Toxic Rocket Fuel Chemical in Human Breast Milk; Majority of Infants Will Exceed New NAS/EPA Perchlorate Safety Level." (2) Such headlines are nothing new. Articles challenging the benefits and even the safety of breastfeeding are found in the media every few weeks or months. As professionals, we are often called upon to respond to such reports and separate the hype from the facts and findings. Unfortunately, critical evaluation does not translate well into messages that are practical for the general public. Headlines such as "Breast Milk Is Toxic" are more likely to sell papers than a scholarly description of a study's limitations. However, we know that careful examination of research articles is essential to understanding how each study contributes to our growing base of evidence. It is our responsibility to be informed about the reality behind sensationalism of research in the press. The article by Kirk and coworkers provides an excellent example of how the interpretation of an article by the media can be far removed from the original findings. Although the focus of the article was US women, media coverage of similar articles is common worldwide.

In Kirk's study, perchlorate concentration was assessed in 47 samples of cow's milk obtained in 11 states and 36 human milk samples from women in 18 states. Five of the human milk samples came from California, 10 from Texas; many states were represented by only 1 sample. Iodide (a form of iodine) levels were also assessed in a subset of the samples. Perchlorate was detectable in all but 1 of the milk (dairy and human) samples. Mean perchlorate levels were 2.0 and 10.5 [micro]g/ L in the dairy milk and breast milk samples, respectively.

Many of the media reports imply that the findings of the study by Kirk et al are generalizable to the US population because the study included milk samples from 36 healthy lactating women from 18 states who were "recruited at random." Because it is often practically impossible to study each individual within large populations, researchers select a subset of the population to study in hopes that the subset can be used to accurately describe a population as a whole. Obtaining a sufficient and representative sample...

Source Citation

Source Citation   

Gale Document Number: GALE|A140911703