Volume 118, Issue 6 , Pages 1257-1264, December 2006
Multiple chemical sensitivities: A systematic review of provocation studies
Article Outline
A systematic review of provocation studies of persons reporting multiple chemical sensitivities (MCS) was conducted from databases searched from inception to May 2006. Thirty-seven studies were identified, testing 784 persons reporting MCS, 547 control subjects, and 180 individuals of whom a subset were chemically sensitive. Blinding was inadequate in most studies. In 21 studies odors of chemicals were probably apparent; 19 of these reported positive responses to provocations among chemically sensitive individuals, and 1 study demonstrated that negative expectations were significantly associated with increased symptom reporting after provocations. Seven studies used chemicals at or below odor thresholds, and 6 failed to show consistent responses among sensitive individuals after active provocation. Six studies used forced-choice discrimination and demonstrated that chemically sensitive individuals were not better at detecting odor thresholds than nonsensitive participants. Three studies tested individuals by using nose clips/face masks and confirmed response, possibly mediated through eye exposure. Three studies used olfactory masking agents to conceal stimuli, and none of these found associations between provocations and response. We conclude that persons with MCS do react to chemical challenges; however, these responses occur when they can discern differences between active and sham substances, suggesting that the mechanism of action is not specific to the chemical itself and might be related to expectations and prior beliefs.
Key words: Multiple chemical sensitivities, idiopathic environmental intolerance, MCS, IEI, cacosmia, provocation studies, functional somatic syndromes
Abbreviations used: MCS, Multiple chemical sensitivities, MTBE, Methyl terbutyl ether
Multiple chemical sensitivities (MCS) is a controversial diagnosis, first defined in the 1950s by a group describing themselves as clinical ecologists.1 Subjects with the condition have a range of subjective symptoms after exposure to low levels of chemicals. Such chemicals can be as diverse as petrol or diesel, perfume, pesticides, or newspaper print.2 The reported symptoms span multiple organs and are poorly defined. Symptoms can include nausea, muscular pain, difficulties with memory, and fatigue.2
The underlying cause of MCS remains disputed. On the one hand, there are suggestions that MCS is a biologic response to low-level chemicals, having an effect on a variety of organ systems, including the immune, nervous, endocrine, or respiratory systems, with a mechanism distinct from accepted immunologic hypersensitivity mechanisms.3 Conversely, other commentators have suggested that the mechanism underlying MCS is predominantly psychological, suggesting that somatic responses to chemicals might be learned responses.4 Others have argued that MCS is a sociocultural phenomenon related to fears of environmental pollution and concerns over industrialization.5
One way of addressing competing models is through experimental trials incorporating systematic exposure to stimuli claimed to trigger symptoms in subjects with MCS. There have been many such trials, both by those identifying themselves with the clinical ecology movement, as well as by those who suggest psychological mechanisms. To date, however, there have not been any systematic reviews of such trials. The present review aims to examine all such trials.
Two key questions were tested: (1) Are subjects with MCS able to identify sham or active stimuli under double-blind provocation conditions in which an adequate placebo is also used? (2) Does the adequacy of blinding in trials have an effect on outcomes?
Methods
Search strategy
A systematic search of databases was conducted from inception to May 2006. Databases included Allied and Complementary Medicine Database (AMED), Applied Social Sciences Index and Abstracts (ASSIA), Cinahl, Cochrane Collaboration Library, Embase, Index to Theses, Web of Science, Medline, and Psychinfo. Further references were obtained from databases set up by user groups available through the Worldwide Web.6
Search terms used included MeSH term headings or free-text key words relating to MCS, such as “idiopathic environmental intolerance,” “chemical intolerance,” and “multiple chemical sensitivities.”
Studies were included in the review if they met all of the following conditions:
Outcomes analyzed included the ability of subjects with MCS to correctly identify provocative stimuli under test conditions, as well as subjective reports or observer-measured outcomes of discomfort/irritation in keeping with classic MCS complaints after exposure to chemicals.2 Neuropsychological outcomes were included because patients with MCS commonly complain of difficulties with memory and concentration.2 Outcomes peripheral to the self-diagnosis of MCS, such as biomarkers, were excluded from this review.
Review process
The literature search, assessment of inclusion, and data extraction were conducted by using a structured table (JD). All main authors were contacted and asked if they had any unpublished data to reduce publication biases (JD). Accuracy of data extraction was cross-checked by a second reviewer (GJR). Differences were resolved by discussion among all 3 authors (JD, GJR, and SW).
Results
Search results
Using the MeSH terms and key words described above, approximately 10,000 titles or abstracts were identified, from which 165 were obtained for examination in full because they appeared relevant to the review. Of the 165 obtained studies, 113 were excluded because they were not relevant to our review. In addition, 15 studies were excluded because they did not meet all of the entry criteria for our review. This included studies using carbon dioxide7 and sodium lactate8 and studies using the chemical found in chili, capsaicin, as a provocation. We excluded these because they are not classic stimuli described by subjects with MCS. Several other studies were excluded because they used outcome measures peripheral to the self-diagnosis of MCS. One further study was excluded because the authors were explicit in stating that the initial sample did not include persons with a prior reported hypersensitivity or intolerance to chemicals.9
Duplicate publications were included as one study.10, 11, 12, 13, 14 However, assessment of certain duplicate publications was not always clear-cut. For example, we have included 3 articles by Rea15, 16, 17 as separate studies, given that they appeared to report slightly different results. However, because the samples included in each article reported almost identical sets of highly idiosyncratic sensitivities, we remain uncertain as to what the true relationship was between the articles.
Several authors confirmed further unpublished studies but were unable to supply these (K. Osterberg, personal communication, May 2006; N. Fiedler, L. Molhave, and M. R. Joffres, personal communication, September 2005). One author reported having done more than 15,000 individual double-blind challenges but had stopped publishing findings (W. Rea, personal communication, September 2005). Abstracts for 2 unpublished studies were also included, and further details were obtained by contacting the authors (S. Bornschein, personal communication, May 2006; T. Hummel, personal communication, June 2006).18, 19 In the case of one unpublished study, no replies were received from the authors, and this was subsequently excluded.20
Studies were categorized by the quality of blinding used: double- or single-blind studies that also included an odorous masking agent to prevent identification of the stimuli (Table I),21, 22, 23, 24, 25, 26 double-blind studies without masking (Table II),11, 12, 15, 16, 17, 18, 19, 27, 28, 29, 30, 31, 32 single-blind studies without masking (Table III),13, 14, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44 and open-label studies or studies with unclear blinding (see Table E1 in the Online Repository at www.jacionline.org).10, 45, 46, 47
Table I. Double- and single-blind provocation studies with olfactory masking
| Reference | N | Exposure | Placebo | Mask | Route | Significant responses (P < .05 unless otherwise indicated) |
|---|---|---|---|---|---|---|
| Double blind | ||||||
| 36 CS | Specific to subject | Distilled water | Vanilla | Inh CC | No differences in neuropsychological testing | |
| 20 CS | Specific to subject | Clean air | Peppermint, cinnamon, anise | Inh CC | No differences in number of symptoms, correct guesses, or objective signs to active and placebo exposures | |
| Single blind | ||||||
| 14 CS/15 H | Acetone, VOCs ± FM | FM | FM | Inh | More irritation from placebo (coffee odor) vs active agent in CS; no differences in eye redness or nasal cavity changes after exposure between CS and H | |
| 10 CS/7 H | Glue, scented body wash, dryer sheet | Unscented shampoo, clean air | Wore nose clips; no olfactory mask | Inh | Changes to skin conductance in CS compared with after placebo noted; however, no differences in EMG, pulse, respiration, temperature, and cognition between active and control groups | |
| 11 CS | Perfume on damp compress | Saline on damp compress | Eye shields for Inh, nose clips for eye provocation; no olfactory mask | Inh/eye | (1) Airway exposure: greater dyspnea (P < .05) after perfume provocations compared with placebo. (2) After eye exposure: significant dyspnea (P < .01), cough (P < .05), and eye irritation (P < .05), compared with placebo. | |
| 9 CS | Perfume on damp compress | Saline on damp compress | Nose clips; mask over mouth with or without carbon filter | Inh | Perfume provoked a stronger response in CS (P < .01); no significant reduction of symptoms with face masks; eye symptoms (P < .05) and dyspnea (P < .05) to perfume |
Table II. Double-blind provocation studies: no olfactory masking
| Reference | N | Exposure | Placebo | Route | Significant responses (P < .05 unless otherwise indicated) |
|---|---|---|---|---|---|
| Bornschein et al18 | 20 CS, 20 H | VOCs less than odor threshold (800 μg/m3) | Clean air | Inh, CC | No difference in proportion of correct ratings between CS and H (true-positive and true-negative ratings/total number of ratings: 46.9% H vs 55.1% CS) |
| Osterberg et al28 | 29CS, 16EA, 39CS/EA, 54 H | N-BuAc (0.37, 1.5, 6 ppm) | Base line state | Inh CC | No differences between groups in smell intensity; CS/EA group showed increased annoyance, sustained fatigue, and membrane irritation compared with H |
| Fiedler et al11 | 12CS/19 H | Gasoline, gasoline + 11% MTBE, and gasoline + 15% MTBE | Clean air | Inh CC | Corrected for multiple comparisons; no differences between CS and H on any symptom subscale, except greater MTBE symptoms after 15% MTBE/gasoline vs 11% MTBE/gasoline provocation |
| Vieregge et al19∗ | 23CS/23 H | 2-prop | Room air | Inh | CS not better at odor identification and discrimination vs H |
| Hummel et al29∗ | 23 CS | 2-prop, PEA | Room air | Inh FC | 20% of CS showed symptoms regardless of challenge; no effect on PEA detection with FC |
| Kjaergard et al27 | 14 CS/21 H | 22 VOCs | Clean air | Inh CC | CS “more likely” to report irritation, tiredness, worse expiratory flow rate, and impaired digit symbol test results compared with H; no statistics |
| Rea et al30 | 19 CS | Variety (eg, FOR, pesticides) | Saline | Inh, oral | Oral: 14/19 responded with usual symptoms; Inh: 10/10 responded to at least one with usual symptoms; no statistics |
| Rea et al31 | 50 CS | VOCs; pilot light (NB) | Saline, water | Inh | 50% reacted to ≥1 active substances; 10 CS: increased pulse (P < .001) |
| Pan et al32 | 49 CS/14 H | 0.74% FOR | Spring water | Inh | 31/49 (63.3%) CS positive, 1 (2.0%) questionable, 17 (34.7%) no reaction |
| Molhave et al12 | 62 CS | VOCs (5 or 25 mg/m3) | Clean air | Inh CC | Higher ratings of odor intensity and reduced air quality after active provocations, impaired digit span; no effect on blinking frequency |
| Rea17 | 10 CS | Variety | Saline | Inh | 9/10 CS reacted to pilot flame (NB); “recognition/repulsion” to perfume, cigarette smoke |
| Rea15 | 10 CS | Variety | Saline | Inh | 6/10 showed vascular “signs and symptoms” after pilot flame exposure; “spontaneous” bruising in 10/10 CS |
| Rea16 | 12 CS | Variety | Saline | Inh | 10/12 CS: response to chemicals; all experienced arrhythmias on 2 occasions |
∗Chemically sensitive subjects were the same individuals in these experiments. |
Table III. Single-blind provocation studies: no olfactory masking
| Reference | N | Exposure | Placebo | Route | Significant responses (P < .05 unless otherwise indicated) |
|---|---|---|---|---|---|
| Papo et al35 | 23 CS, 44 non-CS | n-Butanol, various(eg, gasoline, menthol) | Baseline | FC Inh | No differences in olfactory threshold/intensity; PEA, H2S, and menthol perceived more negatively in CS vs non-CS |
| Osterberg et al42 | 10 CS, 20 H | n-Butanol, n-butyl acetate, toluene | Baseline | FC Inh CC | Normal smell in CS and H; CS vs H: mucous membrane irritation, fatigue, and decreased reaction times; no difference in odor intensity/annoyance in CS vs H |
| Ojima et al36 | 25 CS 50 H | Various (eg, gasoline, leather, soap, menthol) | Baseline | Inh | No differences in odor identification; more odors judged “unpleasant” and less odors judged “pleasant” in CS vs H |
| Kakuta et al37 | 10 CS, 6 H | Ethanol, IA, xylene, toluene, FOR | Air/baseline | Inh | All CS responded to 2-3 active provocations with headaches/coughing; no statistics |
| Caccappolo et al33 | 33 CS/56 non-CS | PEA in glycerol, pyridine in mineral oil; various dilutions | Air/baseline state | Inh FC | No differences in CS vs non-CS in odor detection/identification; PEA: more irritation in CS vs non-CS; PYR: no differences between CS and non-CS |
| Dalton34 | 180 CS/H∗ | Balsam, butanol, methyl salicylate (wintergreen); bias conditions† | Air/baseline | Inh CC | Odor intensity associated with bias and odor reactivity (balsam R2 = 0.68, wintergreen R2 = 0.63, butanol R2 = 0.78) |
| Fernandez et al38 | 12 CS, 21 non-CS | Vanilla in PG, peppermint in PG | Air or PG | Inh FC | No differences in odor identification in CS vs non-CS; no difference between groups on pleasantness/intensity |
| Orbaek et al43 | 24 CS, 12 H | Toluene, n-butyl acetate | Air/baseline | Inh CC | CS vs H: more annoyance and fatigue, especially after toluene; no effect on cognitive tests |
| Bell et al13 | 33 CS, 33 H | Butanol, galaxolide, PG | Air, water, PG | Inh CC | No differences in proportion of CS vs H correctly detecting sham or chemical; CS rated provocations less pleasant than H (P < .04) |
| Rudell et al39 | 3 CS, 5 H | Floor material- benzyl alcohol | Air | Inh CC | Authors unable to “reproduce symptoms and signs experienced by the subjects” |
| Doty et al41 | 18 CS, 18 H | PEA in PG, MEK in air | PG and air | Inh FC | No differences between CS and H in odor identification threshold |
| Iregren44 | 14 CS, 12 H | Toluene (0.2 and 3.2 mmol/m3, 300 mg/m3) | Baseline | Inh CC | No difference on neuropsychological tests in CS vs H; increased irritation to provocations in CS vs H (P < .01); increased headache, fatigue, dizziness to increasing toluene in both (P < .0001) |
| Camp and Morgan14 | 30 CS‡ | Carbonless copy paper | Bond paper | Inh CC | Absence of consistent complaints to active or control provocations |
| Frigas et al40 | 13 CS | FOR DB in 3/13 CS, SB in 10/13 | Air | Inh CC | 11/13 CS: no change in FEV1 after provocation |
∗Subjects given “odor reactivity” scores. |
†See text for bias condition details. |
‡Nine of 30 participants: no previous symptoms/unsure of symptoms. |
Studies using olfactory masking agents or nose clips
Three studies testing 70 chemically sensitive participants incorporated blinding and an olfactory mask and failed to show any significant effects on subjects with MCS after provocation with chemicals. One of these studies demonstrated that subjects with MCS were more likely to react adversely to placebo over active substances.23 Three further studies incorporated the use of nose clips but no olfactory mask and tested 30 subjects with MCS, showing some significant effects, including changes to skin conductance, irritation, and dyspnea, after exposure with chemicals. These effects might have been mediated through the eyes (Table I).
Studies not using olfactory masking agents
In a total of 8 of 13 studies described as double blind but not incorporating any olfactory masking agent, positive responses to provocations among subjects with MCS were reported (Table II). One of these studies used a forced-choice method of provocation and demonstrated that 20% of chemically sensitive participants showed positive responses whether provoked with active or sham substances.29 One other study reported greater methyl terbutyl ether (MTBE) symptoms after 15% MTBE/gasoline provocations versus 11% MTBE/gasoline provocations.11 However, the forced-choice stages of this trial failed to demonstrate that chemically sensitive individuals were better at differentiating between gasoline-only/gasoline-MTBE exposures with or without olfactory masking.11
Fourteen studies incorporated single-blind methodology without olfactory masking agents (Table III). Five of these studies used forced-choice methods of discrimination, and all of these studies failed to demonstrate that chemically sensitive participants were better at detecting odors than healthy control subjects, whereas 1 of the forced-choice studies showed that participants with MCS were more likely to report negative responses to phenyl ethyl alcohol, usually described as a pleasant odor and commonly found in perfume.33 One study testing 180 persons demonstrated that reporting of somatic symptoms and irritation from odors was related to bias conditions before exposure and was more likely in persons scoring highly on chemical odor intolerance indices.34 Three of the single-blind studies, testing 46 individuals, failed to consistently reproduce symptoms after exposure. The majority of the single-blind studies confirmed that although active provocations did not affect neuropsychological functioning in healthy or chemically sensitive subjects (with the exception of one study42), chemically sensitive subjects were more likely to report most odors as “unpleasant” compared with healthy subjects or report greater fatigue, somatic symptoms, or annoyance after active provocations.
There were 4 studies, testing 64 individuals under nonblind conditions (see Table E1 in the Online Repository at www.jacionline.org). Among these, all studies showed positive responses to provocations, with one study demonstrating violent “nocebo” responses to nebulized saline in the one single-blind exposure performed in the study.45
Discussion
The findings demonstrate that chemical provocation studies that actively mask stimuli show that persons with MCS are unable to differentiate between active and control conditions, symptomatically or otherwise. Although subjects with MCS do react to masked provocations, this is at equal levels in both (active vs sham) conditions. However, under conditions in which participants are better able to discern differences between provocation stimuli because of insufficient masking of active and placebo conditions, individuals react more often to active over sham provocations. This pattern of response might be due to beliefs or expectations regarding the exposure.23, 34
The importance of prior expectations was reinforced in the study by Dalton.34 Prior to exposure in this study, participants were informed that they were either to be exposed to “natural extracts with relaxing effects” (positive bias), industrial solvents (negative bias), or odorants “approved for olfactory research” (neutral). In this study participants were more likely to report irritation and characteristic symptoms when they believed they were being exposed to industrial solvents (harmful bias) over the positive or neutral-bias groups (P < .05). Using regression analysis, the author demonstrated that individuals rating high on “odor reactivity” scores were more likely to rate odors as irritating if given a negative bias condition than those rating lower on odor reactivity scores. Individuals with higher odor reactivity scores were also more likely to report somatic symptoms after exposures believed to be harmful.
When is a “double-blind” study not truly double blind?
A difficulty with the analysis was the variable quality of many of the trials. Many studies did not use adequate masking agents and thus compromised study blindness because participants might have been able to tell to what they were exposed. For example, in some studies the placebo was relatively odorless and colorless, such as clean air or water, whereas the active agent included typically noxious/pungent substances, such as formaldehyde, volatile organic chemicals, or tobacco smoke. Most participants, whether reporting MCS or not, should be able to discern between such placebo and active substances. In one study described as double blind, the authors themselves noted that “the exposed subjects” were able to “smell the substances, as they obviously can.”27 Although creating an adequate mask for test substances can be difficult, some investigators managed this by using olfactory maskers, such as coffee, peppermint, or vanilla.21, 22, 23
It could be argued that adding an olfactory mask to provocations is simply adding a new chemical to which participants might also react. For example, in one study23 participants were more likely to respond adversely to the coffee odor mask than to unmasked acetone or volatile organic chemicals, although the authors postulated that this could have been because the painters in this study were less accustomed to coffee odor placebo than acetone (which was present in their day-to-day environments) and therefore interpreted the coffee odor as threatening.
The problem of odorant masks was surmounted in one study by first exposing participants to clean air/olfactory mask–only provocations.22 Only participants not reacting to the initial olfactory mask exposures were eligible for entry into the actual double-blind study, which involved exposure to sham versus active provocations with olfactory masking agents.22 In other studies investigators exposed participants to chemicals at or below odor thresholds.10, 29, 32, 40, 44 Most of these failed to demonstrate that subjects with MCS were better at detecting active provocations compared with healthy participants; however, they did tend to confirm increased states of autonomic arousal or MCS symptoms, independent of type of provocation, when compared with control subjects.
Others attempted to surmount the problem of blinding by making participants wear nose clips or face masks.24, 25, 26 None of these studies used olfactory masking agents. Participants in these studies showed some response to active provocations, and the authors suggested that participants might have perceived provocations through the eyes. Such methods of blinding therefore appear less satisfactory than studies that use subolfactory threshold provocations or olfactory masking agents.
Strengths and weaknesses of this review
The inclusion criteria for this review were kept broad to increase the number of studies that could be analyzed. However, this meant that included studies were heterogeneous. Some studies recruited their sample from the community by using questionnaires13 or included active workers with a history of exposure44 rather than outpatients with MCS per se (A. Iregren, personal communication, June 2006). The characteristics of such a sample could differ significantly from self-professed subjects recruited from clinical ecology clinics. It could be argued that such studies should therefore be reviewed separately. Other studies introduced further subcategorizations of sensitivity (eg, “MTBE sensitive”11 and “toxic encephalopathy types 2a/2b”43). We still included these studies under the general rubric of chemical sensitivity if they met our inclusion criteria. This could be a potential disadvantage if there are distinctions between subgroups, which would be lost by combining all groups; however, MCS is a self-reported diagnosis that remains poorly defined, and in some respects grouping these studies together was therefore justifiable. Despite these shortcomings, the overall finding of this review, that the quality of blinding in provocation trials is of paramount importance in helping to establish possible mechanisms of response in participants, remains a valid and important one.
The heterogeneity of the reviewed studies limited the possibility of meta-analysis. Studies used a range of different concentrations, as well as types of exposure, over differing time periods. In a review of this type, location and publication biases are also important. Location bias was addressed by including databases set up by patient and user groups within our searches. We also ensured that articles that were not in English but appeared to relate to the aims of our study were retrieved and translated. Publication biases were reduced by contacting all main authors.
To our knowledge, this is the first systematic review of provocation trials in persons with MCS and builds on the findings of previous reviews.19, 48, 49, 50 Unlike previous reviews,19, 48, 49, 50 we endeavored to include all studies, whether conducted by groups associated with the clinical ecology movement or groups associated with psychology or medicine. Within this area, there are widely acknowledged polarities between groups,3, 5, 51 and we aimed to overcome this by contacting all persons involved in this field of research, irrespective of their background.
Future research and recommendations
A striking finding was that the ability of participants with MCS to accurately detect or respond to active provocations was less likely when strict blinding was observed and when a mask of the experimental stimuli was incorporated. In addition, subjects with MCS did not show differences with respect to odor detection thresholds compared with healthy subjects nor, in general, were they better able to distinguish between active and control stimuli when exposed at levels that were subthreshold for odors. This therefore suggests that there could be salient psychological mechanisms operating that lead sufferers to respond, occasionally with marked physiologic responses, when they are aware of the exposure. Many of the responses recorded in the studies after unmasked provocations included outcomes consistent with autonomic arousal and anxiety. Therefore the findings suggest that a behavioral mechanism of action rather than a purely biologic or physiologic mechanism related to the characteristics of the substance might explain this condition. This is consistent with previous research that has suggested that behavioral conditioning could be an important mechanism in the pathogenesis of MCS. Participants with MCS exposed to carbon dioxide provocations can experience typical MCS symptoms similar to those experienced by patients with panic disorder.7, 8 Exposure to foul-smelling odors coupled with carbon dioxide can lead healthy participants to acquire typical somatic symptoms, including hyperventilation, chest tightness, and feelings of choking.52 Learned somatic symptoms can be re-elicited a week later after provocation with noxious odors alone and might generalize to new odors not used in initial provocations.52 Furthermore, symptom learning has also been shown to occur more readily in healthy participants given worrying a priori environmental information about exposures, and under such conditions, pleasant odors can become triggers for symptoms as well,53 with participants' prior expectations rather than actual experimental conditions governing symptom learning.54 Therefore the results of this study do not support recent policy initiatives to ban the use of fragrances in public places.55 Our results suggest that imposing such measures could aggravate the problem further.
If behavioral conditioning is a mechanism of action, treatment for MCS might be possible by using cognitive behavioral therapy. This could be used in persons who do not necessarily have a comorbid psychiatric disorder.
Future work should focus on behavioral approaches in the treatment of MCS.
We thank Gary Hahn for his assistance in the retrieval of many of the articles used in this review, as well as all of the authors who responded to queries sent to them. We also thank Toshiko Mori for help in translating Japanese articles.
Appendix. Supplementary data
References
- . Human ecology and susceptibility to the chemical environment. Springfield (IL): Charles Thomas; 1962;
- . The worker with multiple chemical sensitivities: an overview. In: Cullen M editors. Occupational medicine: state of the art reviews. Vol. 2:Philadelphia: Hanley and Belfus; 1987;
- . Chemical sensitivity and the environment. Immunol Allergy Pract. 1982;4:21–31
- . Learning subjective health complaints. Scand J Psychol. 2002;43:147–152
- . Multiple chemical sensitivity: pseudodisease in historical perspective. Scand J Work Environ Health. 1997;23(suppl 3):35–42
- . Coping with multiple chemical sensitivity: Internet resources. Health Care Internet. 1998;2:57–73
- Carbon dioxide inhalation challenges in idiopathic environmental intolerance. J Allergy Clin Immunol. 2000;105:358–363
- . Panic response to sodium lactate infusion in patients with multiple chemical sensitivity syndrome. J Allergy Clin Immunol. 1997;99:570–574
- . Evaluation of odor and sensory irritation thresholds for methyl isobutyl ketone in humans. AIHA J (Fairfax, Va). 2000;61:340–350
- . Time courses of sensory irritations due to 2-butanone and ethyl benzene exposure: influences of self-reported multiple chemical sensitivity (sMCS). Int J Hyg Environ Health. 2002;204:367–369
- Controlled human exposure to methyl tertiary butyl ether in gasoline symptoms: psychophysiologic and neurobehavioral responses of self reported sensitive persons. Environ Health Perspect. 2000;108:753–763
- . Human reactions to low concentrations of volatile organic compounds. Environ Int. 1986;12:167–175
- . Quantitative EEG patterns during nose versus mouth inhalation of filtered room air in young adults with and without self-reported chemical odor intolerances. Int J Psychophysiol. 1998;28:23–35
- Camp JE, Morgan MS. Upper respiratory irritation from carbonless copy paper exposure. In: Swedish Council for Building Research. 3rd International Conference on Indoor Air Quality and Climate. Vol. 3. Stockholm: Swedish Council for Building Research; 1984. p. 393-6.
- . Environmentally triggered small vessel vasculitis. Clin Ecol. 1977;38:245–251
- . Environmentally triggered cardiac disease. Clin Ecol. 1978;40:243–251
- . Environmentally triggered thrombophlebitis. Clin Ecol. 1976;37:101–109
- . Double-blind chemical exposure in patients with MCS. [abstract] J Toxicol Clin Toxicol. 2005;43:436
- . Intranasal chemoreception in patients with multiple chemical sensitivities and healthy controls. Chemical Senses. 2000;25:801
- . Perceived odor intensity and hedonics in subjects with multiple chemical sensitivity. Chem Senses. 1993;18:584
- . Examining the influence of biological and psychological factors on cognitive performance in chronic fatigue syndrome: a randomized, double-blind, placebo-controlled, crossover study. Int J Behav Med. 2003;10:162–173
- . Double-blind provocation chamber challenges in 20 patients presenting with “multiple chemical sensitivity.”. Regul Toxicol Pharmacol. 1993;18:44–53
- . Multiple chemical sensitivity in male painters: a controlled provocation study. Int J Hyg Environ Health. 2003;206:531–538
- . Physiologic and symptomatic responses to low-level substances in individuals with and without chemical sensitivities: a randomized controlled blinded pilot booth study. Environ Health Perspect. 2005;113:1178–1183
- . Provocations with perfume in the eyes induce airway symptoms in patients with sensory hyperreactivity. Allergy. 1999;54:495–499
- . Placebo-controlled challenges with perfume in patients with asthma-like symptoms. Allergy. 1996;51:434–439
- . Human reactions to a mixture of organic compounds. Atmos Environ. 1991;25A:1417–1426
- . Annoyance and performance of three environmentally intolerant groups during experiment challenge with chemical odors. Scand J Work Environ Health. 2004;30:486–496
- . Intranasal chemoreception in patients with multiple chemical sensitivities: a double-blind investigation. Regul Toxicol Pharmacol. 1996;24:s79–s86
- . Chemical sensitivity in physicians. Bol Asoc Med P R. 1991;83:383–388
- Confirmation of chemical sensitivity by means of double blind inhalant challenge of toxic volatile chemicals. Clin Ecol. 1989;VI:389–393
- Pan Y, Rea WJ, Johnson AR, Fenyves EJ. Formaldehyde sensitivity. Clin Ecol 1989;VI:79-84.
- Odor perception: multiple chemical sensitivities, chronic fatigue, and asthma. J Occup Environ Med. 2000;42:629–638
- . Cognitive influences on health symptoms from acute chemical exposure. Health Psychol. 1999;18:579–590
- Chemosensory function and psychological profile in patients with multiple chemical sensitivity: comparison with odor sensitive and asymptomatic controls. J Psychosom Res. 2006;60:199–209
- . Odor perception in patients with multiple chemical sensitivity. Tohoku J Exp Med. 2002;198:163–173
- . Influence of chemical materials on adolescent children with allergic diseases. Neuro-ophthalmol Japan. 2002;19:176–187
- . Subjective ratings of odorants by women with chemical sensitivity. Toxicol Ind Health. 1999;15:577–581
- . A mobile exposure chamber for challenge tests to the human respiratory tract. Appl Occup Environ Hyg. 1997;12:465–469
- . Bronchial challenge with formaldehyde gas: lack of bronchoconstriction in 13 patients suspected of having formaldehyde-induced asthma. Mayo Clin Proc. 1984;59:295–299
- . Olfactory sensitivity, nasal resistance, and autonomic function in patients with multiple chemical sensitivities. Arch Otolaryngol Head Neck Surg. 1988;114:1422–1427
- . Annoyance and performance during the experimental chemical challenge of subjects with multiple chemical sensitivity. Scand J Work Environ Health. 2003;29:40–50
- . Suprathreshold intensity and annoyance reactions in experimental challenge to toluene and n-butyl acetate among subjects with long-term solvent exposure. Scand J Work Environ Health. 1998;24:432–438
- . Subjective and objective signs of organic solvent toxicity among occupationally exposed workers. Scand J Work Environ Health. 1986;12:469–475
- . Provocative challenges in patients with multiple chemical sensitivity. J Allergy Clin Immunol. 1997;99:438–442
- Responses to diesel vapor exposure among chemically sensitive Gulf War veterans. Psychosom Med. 2004;66:588–598
- . Effect of exposure to volatile organic compounds on plasma levels of neuropeptides, nerve growth factor, and histamine in patients with self-reported multiple chemical sensitivity. Int J Hyg Environ Health. 2004;207:159–163
- . Chemosensory function and response in idiopathic environmental intolerance. Occup Med. 2000;15:539–556
- . A review of multiple chemical sensitivities. Occup Environ Med. 1999;56:73–85
- . Idiopathic environmental intolerances: results of challenge studies. Occup Med. 2000;15:529–537
- Position statement on clinical ecology. J Allergy Clin Immunol. 1986;78:269–271
- Acquiring symptoms in response to odors: a learning perspective on multiple chemical sensitivity. Ann N Y Acad Sci. 2001;933:278–290
- Media warnings about environmental pollution facilitate the acquisition of symptoms in response to chemical substances. Psychosom Med. 2003;65:332–338
- Perceived relation between odors and a negative event determines learning of symptoms in response to chemicals. Int Arch Occup Environ Health. 2004;77:200–204
- Puddicombe D. Catching whiff of a scent ban: city looking to curtail fake fragrances. Ottawa Sun. May 27, 2006; News: Ottawa & Region:5.
Disclosure of potential conflict of interest: The authors have declared that they have no conflict of interest.
PII: S0091-6749(06)01696-4
doi:10.1016/j.jaci.2006.07.046
© 2006 American Academy of Allergy, Asthma and Immunology. Published by Elsevier Inc. All rights reserved.
Volume 118, Issue 6 , Pages 1257-1264, December 2006

