Fog & Smoke Working Group - Reading List

Letter to Mr. Lowell Fowler from Dr. James Kehrer


August 28, 1995

Mr. Lowell R. Fowler
High End Systems, Inc.
2217 West Braker Lane
Austin, TX 78758

Dear Mr. Fowler:

I have reviewed the copy of the report prepared by Consultech Engineering Company for Actors' Equity Association on the health effects of glycol-based fogs used in theatrical productions. As you know, I have extensively examined the research data base on these glycols, and am familiar with their toxicology. There are numerous flaws with the "scientific" analyses provided in the Consultech report. Thus, my previous conclusions that these fogs are safe when generated by High End Systems machines and used as directed remains unaltered.

The toxicology of the glycols themselves will, of course, be similar for all fog machines. These glycols are hygroscopic and at high concentrations will have a drying effect on the nose, eyes and throat. Physically active performers will likely experience more of these effects. However, these effects are transient, and rapidly reverse once exposure stops. No permanent injury will result, and thus it is not perceived as a health risk.

The Consultech report focuses on two main areas. First, the results of chemical analyses of the fog fluids indicating the presence of degradation products that may be toxic, and second the results of health effects surveys on fog exposed and non exposed performers. I will deal with each of these areas separately, plus a discussion of general points raised in the report.

General Points
(page numbers refer to those in the Consultech report)

Glycol-based fogs are often referred to as smoke, although they do not consist of combustion products. Despite this obvious difference, the Consultech report equates glycol vapors with cigarette smoke (page 3 and page 9). Cigarette smoke is clearly irritating and contains literally thousands of different chemicals. Such a comparison is both misleading and inappropriate.

Concerns are voiced (page 4) about equating effects upon the respiratory system in animals with those in humans since their respiratory systems differ and animals lack many of the sensitivities of humans. This argument suggests that animal studies would not provide valuable toxicologic and other biomedical information prior to studies in humans. This is clearly not true as animals are well established as adequate models. There are respiratory differences between humans and some animals that might limit the conclusions that could be drawn. Fortunately, however, the most important studies on glycol-fogs have been done in monkeys and, because some of these glycols were for a time used as "air sterilizers" in sick rooms, also in humans. The toxicology data has firmly shown no adverse effects other than drying actions. In fact, because of the disinfectant activity, lung diseases were lower in exposed groups.

It is indicated on page 6 that "a large number" of complaints have been received from performers exposed to fogs. This is a very imprecise description. How many? How many relative to the exposed population and to control populations? Were they unsolicited complaints or were they received in response to the concerns voiced by Actors' Equity?

It is indicated on page 6 that Consultech has provided technical assistance for 3 litigations. Thus, this company is would appear to have a conflict of interest.

The issue is raised that there is an absence of information on potential interactions between the combinations of chemicals used in these fogs (page 8). This is, in many respects, a spurious argument. Unexpected interactions are always a possibility; and one that is difficult to dismiss. However, some of the fog fluids contain only one chemical. Most others, only two. The two chemicals mainly used (propylene glycol and triethylene glycol) have almost identical physical properties and no identifiable toxicities except at enormous doses. There is no reason to expect increased toxicity when two non-toxic chemicals from the same chemical grouping are combined.

The toxicology of the glycols is misrepresented on page 10. Triethylene glycol can be toxic when given intravenously, but so can water. This compound is not a "poison". Further, the fact that heating it to decomposition yields "acrid smoke and irritating fumes" is irrelevant since this level of heating is not done in fogs generated by High End Systems' machines. Finally, the statement that "propylene glycol is an experimental teratogen" (page 10) is wrong. All data have shown this compound is not teratogenic.

The Union Carbide study demonstrating acute toxicity of triethylene glycol is discussed on page 49. The implication is that this study documents a serious toxicity of triethylene glycol. This is a misrepresentation of these data that were obtained by using extraordinarily high fog concentrations; levels that have no relevance to the entertainment industry and simply show that any compound will be toxic if enough is given.

Chemical Analyses

The Consultech report states (page 7) that its thrust is on the properties of the chemicals used in fogs, and not on any specific product. However, all companies products are not the same, and the methods (particularly temperature) used vary greatly. Clearly, the lower the temperature the less likely it becomes any decomposition products will be generated. Thus, fog machines such as those made by High End Systems (that heat fog fluids to a maximum of 425F) can not be equated with those used to gather the data presented in this report, where heating to 555-700F was used.

The Consultech report makes a major issue of the tests that showed the presence of various decomposition products. The exact conditions of these tests are not given in the report. However, based on the information provided, the results of these tests would not appear to be highly relevant to the exposures humans would receive. For example, fog fluid samples were obtained and analyzed both before and after heating (page 11). No data are given on the results of analyses before heating. After heating at 555-700F various volatile organics were reported. However, there is no information on the length of time they were heated (and how this relates to heating in the fog machines) nor on the levels of the volatile compounds found. The question of levels of these compounds also applies to the analysis performed on fogs produced by a machine (page 12). Finding these compounds in "measurable amounts" is not particularly meaningful since modern analytical instrumentation can detect extraordinarily low levels. It also would be of interest to determine whether similar levels are found in fog produced by machines from other companies, or even in the ambient air of the theaters where they might arise from plastics, cleaning fluids, glues, etc.

The table following page 13 that lists toxicities associated with the degradation products appears to be designed to highlight the dangers of these chemicals. However, levels are again critical since virtually all of these compounds can be found in other situations not considered a toxic hazard.


The design of the Consultech Engineering Actors' Equity Survey is not acceptable for the development of solid scientific evidence since it introduces massive amounts of response bias. In essence, a survey of this type will provide the results wanted; in this case an indictment of glycol-based fog fluids. As a result, none of the data provided is useful. Interestingly, the report acknowledges the presence of response bias on page 32, yet they proceed to dismiss this major flaw and conclude that the results of the survey support their position. In fact, the results could be construed to invalidate their position. There were only 231 responses (of which 98% were exposed to fogs) out of 14,133 Actors' Equity members, with an estimated fog-exposed population of 3000. It seems clear that very few people had any concerns.

Despite these major flaws, the Consultech report proceeds to discuss the data in detail including "statistically significant" results. Since there was no control population (as admitted in the report on page 16), it is not clear how statistics could be run in a valid fashion. Examination of the questionnaire itself revealed leading questions that would clearly bias the results even further. For example, question 7 was "Do you believe your health has been affected by glycol fog?" It is virtually impossible in a survey of this sort not to get a positive response to such a question.

Every aspect of the Consultech Engineering Actors' Equity Survey can be challenged. Thus, even as a descriptive survey (page 31) it does not provide a valid indication of the health risks associated with glycol-based fogs. The fact that this survey was "much more detailed" than the NIOSH study (page 31) does not make it any better. A thorough epidemiologic study is required to draw the conclusions this report attempts to make. The available survey results do not even begin to approach the level of sophistication needed.

The other surveys described in this report at first appear to have more utility. However, there is again a serious lack of controls (or at least information on the controls). This makes it impossible to assess whether the results have any significance. For example, glycol fog and non-fog groups were selected. How did the performers in these two groups compare in terms of performance characteristics (singing, dancing, etc.) age, sex, and existing health problems. Also, were the productions in the fog and non-fog groups comparable in terms of theater size, theater air flow, number of performances per week, other air-borne chemicals (i.e. smoke from gunpowder), etc. If the sole selection criteria was the presence or absence of fog, it is highly unlikely the two groups are comparable. It would also be important to have a group of age/sex matched controls who were not performers. Finally, the treatments selected for analysis in this survey are not all reasonable. For example, acne is not associated in anyway with the compounds of concern. Also, why is intestinal malabsorption included? The drying effects may well be valid, but as stated above, this is a readily reversible effect.

I would be pleased to discuss any aspect of this report with you at your convenience.


James P. Kehrer, Ph.D.
Head, Division of Pharmacology and Toxicology
Gustavus and Louise Pfeiffer Professor of Toxicology

Dr. Kehrer is a Professor at The University of Texas at Austin







Chairs & Publications
Meetings & Minutes
Reading List













© Copyright , ESTA. All Rights Reserved. | Terms of Use | Privacy Statement | Contact Us | ESTA
ESTA, P.O. Box 23200, Brooklyn, NY 11202-3200 • Tel: +1 212 244 1505