As a physician disabled with multiple chemical sensitivity, I read with trepidation the article by Simon and colleagues [1] and the accompanying editorial [2].

Although I applaud the authors for attempting to examine the role of immunologic, psychological, and neuropsychological factors in multiple chemical sensitivity, a possible conflict of interest and flaws in the study design make their lack of positive immunologic and neuropsychological findings suspect. First, the study was sponsored in part by the Boeing Company, which has a huge vested interest in disproving the existence of multiple chemical sensitivity as a "real" illness. Many former employees in Boeing's Washington State plants have pending workers' compensation lawsuits alleging disability from exposures to workplace chemicals.

Another problem is defining patients on the basis of a diagnosis of chemical sensitivity in their chart. No criteria are listed for how the diagnosis was made. The wide spectrum of disease among chemically sensitive patients means that even selecting persons with symptoms in at least three organ systems and who report sensitivity to four or more substances (as Simon and colleagues did) will still mix patients who are mildly, moderately, and severely ill. This may have diluted the findings because only the most severely ill might have shown test abnormalities.

Mixing the laboratory values also presents a problem in interpreting the cellular immune studies because values above and below the normal range cancel each other and make it appear that no abnormalities are present. Other studies have shown that chemically sensitive persons often have immune abnormalities that differ from those of other persons with multiple chemical sensitivity [3-5].

Another source of ambiguity involves the analysis of symptoms that are said to have occurred before or after a finite event called "the onset of chemical sensitivity". Can the exact onset of chemical sensitivity ever be known? Although some people with multiple chemical sensitivity report being totally healthy one day and extremely ill the next, most chemically sensitive persons experience declining health (which can last from several hours to years) punctuated by significant worsenings.

Further, the term "somatization disorder" is a useless and misleading label given to persons with many physical symptoms not explainable by a known medical condition. It is a psychiatric diagnosis only by default. Still, the authors found that almost 25% of case patients showed no evidence of clinically significant psychological distress and that the presence of psychological symptoms in other patients did not predate the onset of chemical sensitivity.

My final complaint about this paper is that the discussion of the use of strict avoidance regimens for the treatment of multiple chemical sensitivity is inappropriate, given that the authors admit that their study did not address this issue. They still go on to state that the "benefits of severe avoidance do not justify the accompanying disability and isolation". This unsubstantiated speculation goes far beyond the scope of their study.