|Year : 2021 | Volume
| Issue : 4 | Page : 134-141
Disease comorbidities associated with chemical intolerance
Raymond F Palmer, Tatjana Walker, Roger B Perales, Rodolfo Rincon, Carlos Roberto Jaén, Claudia S Miller
Department of Family and Community Medicine, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA
|Date of Submission||24-Sep-2021|
|Date of Decision||10-Dec-2021|
|Date of Acceptance||11-Dec-2021|
|Date of Web Publication||29-Dec-2021|
Raymond F Palmer
7703 Floyd Curl Drive, San Antonio, TX
Source of Support: None, Conflict of Interest: None
Background: Chemical intolerance (CI) is characterized by multisystem symptoms initiated by a one-time high-dose or a persistent low-dose exposure to environmental toxicants. Prior studies have investigated symptom clusters rather than defined comorbid disease clusters. We use a latent class modeling approach to determine the number and type of comorbid disease clusters associated with CI.
Methods: Two hundred respondents with and without CI were recruited to complete the Quick Environmental Exposure and Sensitivity Inventory (QEESI), and a 17-item comorbid disease checklist. A logistic regression model was used to predict the odds of comorbid disease conditions between groups. A latent class analysis was used to inspect the pattern of dichotomous item responses from the 17 comorbid diseases.
Results: Those with the highest QEESI scores had significantly greater probability of each comorbid disease compared to the lowest scoring individuals (P < 0.0001). Three latent class disease clusters were found. Class 1 (17% of the sample) was characterized by a cluster consisting of irritable bowel syndrome (IBS), arthritis, depression, anxiety, fibromyalgia, and chronic fatigue. The second class (53% of the sample) was characterized by a low probability of any of the co-morbid diseases. The third class (30% of the sample) was characterized only by allergy.
Discussion: We have demonstrated that several salient comorbid diseases form a unique statistical cluster among a subset of individuals with CI. Understanding these disease clusters may help physicians and other health care workers to gain a better understanding of individuals with CI. As such, assessing their patients for CI may help identify the salient initiators and triggers of their CI symptoms—therefore guide potential treatment efforts.
Keywords: Chemical intolerance, comorbid disease, idiopathic environmental intolerance, latent class, multiple chemical sensitivity, Quick Environmental Exposure and Sensitivity Inventory
|How to cite this article:|
Palmer RF, Walker T, Perales RB, Rincon R, Jaén CR, Miller CS. Disease comorbidities associated with chemical intolerance. Environ Dis 2021;6:134-41
| Introduction|| |
Chemical intolerance (CI) is characterized by multi-system symptoms initiated by a one-time high dose or a persistent low-dose exposure to environmental toxicants, with new-onset intolerances often occurring upon subsequent exposures to structurally unrelated chemicals, foods, and drugs. CI symptoms include fatigue, headache, weakness, rash, mood changes, musculoskeletal pain, gastrointestinal problems, difficulties with memory and concentration (often described as “brain fog”), and respiratory problems.,,, Increasing numbers of patients attribute their illness to a well-defined exposure event, such as the Gulf War, disasters like the World Trade Center, indoor air contaminants, exposures to pesticides, new construction or remodeling, or a flood-or water-damaged building resulting in mold and bacterial growth.,,
Precise prevalence estimates for CI are difficult to obtain due in part to the various names used for the disorder vary across studies, and there remains to be no universally accepted case definition. Further, the criteria and diagnostic tools used to assess CI differ across studies. Prevalence estimates for CI differ by whether it is clinically diagnosed (0.5%–6.5%) or self-reported (average ~20%) in different population-based surveys.,,,, Further, there is evidence of increasing 10-year prevalence rates in the US and Japan.,
The published literature most often refers to CI as multiple chemical sensitivity (MCS) or idiopathic environmental intolerance, with various ways to assess the condition. A recent comprehensive epidemiologic and diagnostic review indicates that assessing CI most often involves the use of the Quick Environmental Exposure and Sensitivity Inventory (QEESI), a 50-item validated questionnaire designed to assess intolerances to chemicals, foods, and/or drugs. The QEESI has been used in over a dozen countries around the world,, (see Palmer et al., 2021 for a comprehensive list of 77 studies in 16 countries) and offers high sensitivity and specificity that differentiates CI individuals from the general population.,, The QEESI is now considered the reference standard for assessing CI and is considered a surrogate for case definition. However, in clinical practice it is important to eliminate and treat alternative diagnoses that may explain signs and symptoms once a positive QEESI screen is identified. Many patients that may benefit from interventions to ameliorate CI symptoms are deprived of this opportunity because many clinicians fail to identify CI as a clinical diagnosis.
Literature on comorbidities associated with chemical intolerance
It has been reported that individuals with CI have an average of 24 clinical visits a year related to their symptoms and other comorbidities. Various other reports reveal that symptoms of CI are often shared with those of chronic fatigue syndrome (CFS) and fibromyalgia (FM), yet demographic and other clinical features do not clearly distinguish patients with CFS, FM, or CI.
Dantoft et al. report several multisystem symptoms associated with CI: Ocular and respiratory (e.g., sinus, lungs, throat, etc.), central nervous system (headache, difficulties concentrating, dizziness, exhaustion/fatigue, panic/anxiety), and symptoms from other organs (skin, heart, gastrointestinal tract, muscles, urinary tract). In a clinical sample of 400 primary care patients, Katerndahl et al. report that 20% of the sample met criteria for CI and had significantly higher rates of allergies and more often met screening criteria for possible major depressive disorder, panic disorder, and generalized anxiety disorder compared to those without CI.
Other reported comorbidities of CI include rhinitis, sinusitis, bronchitis, migraine headache, irritable bowel, multiple food intolerances, arthritis, anxiety, and affective disorders. In a population-based survey of over 1000 respondents with asthma/allergy, “building intolerance,” and CI, a high degree of similar comorbid conditions existed between these groups, suggesting that there may be similar underlying mechanisms between these conditions.
In a large Japanese population-based cross-sectional survey, Azuma et al. report that atopic dermatitis, allergic rhinitis, food allergy, depression, fatigue, and somatic symptoms were positively correlated with CI. In a Canadian population survey of 22,000 adults, it was reported that individuals with CI had 2.37 times greater odds of major depressive disorder, and 3.09 times greater odds of both major depressive disorder and generalized anxiety disorder together.
In a population-based survey of 1,137 individuals, Steinemann reported that of those with CI, 71% had asthma and 86.2% experienced health problems such as migraine headaches when exposed to fragranced consumer products including air fresheners, scented laundry or cleaning products.
Statistical clustering of symptoms and chemical intolerance
The aforementioned studies make it clear that there are various reports of multi-system symptoms and other medical comorbidities associated with CI. However, there has been little systematic statistical analysis of how these varied symptoms or conditions empirically cluster together in individuals with CI. Del Casale et al. used a principal components factor analysis of a symptom checklist among CI patients and found hyperosmia (sensitivity to smell), asthenia (weakness, low energy), and dyspnoea (labored breathing) to be the most common symptoms, along with coughs and headaches.
Hierarchical cluster analysis was used by Eis et al. to detect groups of CI patient symptoms that show maximum similarity to each other and, simultaneously, maximum differences. They found that two-thirds of patient respondents report non-unspecific general symptoms. Other complaints were related to musculoskeletal or other somatic issues. However, there were no differences between those with and without CI regarding symptom clusters. In that study, they point out as a limitation that latent class analysis (LCA) was not used to investigate their hypothesis, and they acknowledge that it would have been a better approach.
Eliasen et al. used a latent class approach to identify somatic symptom profiles in a large general adult population and found that three-fourths of the population did not have somatic complaints. Only 3.9% had profiles defined by multiple somatic symptom complaints. However, this study did not assess for CI. In their follow-up study, a latent class approach was used to classify 31 self-reported somatic symptoms reported by individuals with “Functional somatic syndromes” and “bodily distress syndrome” (e.g., nomenclature counterparts of CI). Eight symptom profiles were identified, which were characterized by combinations of muscle and joint pain, gastrointestinal symptoms, musculoskeletal, cardiopulmonary, and some mixtures of each with general symptoms and one with a high probability of all symptoms. Other profiles were characterized by FM, CFS, and irritable bowel syndrome (IBS) – health conditions known to be associated with CI.
The aforementioned symptom profiles are consistent with the literature on CI.,,, However, with varied results, prior studies have investigated symptom clusters rather than distinct comorbid disease clusters. As such, we use a latent class modeling approach to determine the number and type of comorbid disease clusters associated with QEESI-defined CI. While it is true that symptom and disease clusters may be highly correlated, diagnostic criteria for specific diseases do not always include all possible symptoms clusters. Identifying disease clusters may provide prompts for clinicians to probe for evidence of CI and thus include mitigation strategies to improve symptoms.
| Methods|| |
Two hundred respondents were recruited as part of the Hoffman Toxicant-Induced Loss of Tolerance (TILT) program (www.TILTresearch.org), an environmental health research project designed to improve health outcomes of individuals with CI by identifying environmental triggers in the home and providing best practices for prevention and intervention. Potential respondents were randomly recruited from the waiting room of a busy family practice clinic and from online solicitation. Participants needed to be at least 18 years old. Potential recruits were first screened for CI using the Brief Environmental Exposure and Sensitivity Inventory (BREESI), a brief 3-item questionnaire asking about intolerance to chemicals, foods, and drugs. The BREESI has demonstrated excellent positive and predictive values for CI., Answering yes to any one of the items determined whether participants should complete the QEESI., While the QEESI is comprised of four 10-item scales with a 1–10 Likert response (total scores having a potential range from 0 to 100 for each scale), only the Chemical and Symptom scales are used to classify individuals into severity groups., The cut-off criteria for “very suggestive” of CI is a score ≥40 on both the QEESI CI and Symptom Scales. The criteria for “not suggestive” of CI are scores ≤19 on each of those scales. Scores in the midrange were not eligible for the study. The first 50 clinical respondents who met the criteria for “very suggestive” of CI and consented to be in the study were retained as CI cases. Similarly, the first 50 respondents who met the criteria for “not suggestive” of CI and consented were retained as the comparison group. To obtain equal numbers of cases and controls from each source, the same strategy was used for the online recruits. There was not an attempt to match on age or gender, as potential differences were addressed by including those variables in a multivariate regression model. After IRB-approved consent (University of Texas IRB protocol #HSC20150821H), qualifying participants completed a demographic and health survey that included a list of physician-diagnosed comorbid diseases.
After descriptive statistics were assessed, a logistic regression model was used to predict the odds of disease comparing the very suggestive group (recorded as1) to the not suggestive group (recorded as 0). Covariates included gender, age, ethnicity, and recruitment source (clinic or online).
A LCA was used to inspect the pattern of dichotomous item responses from the comorbid disease section of the health survey. The LCA model is analogous to factor analysis. Both posit an underlying latent variable measured by observed variables. The difference lies in the fact that LCA uses categorical items and response probabilities to determine the likelihood of a particular class membership. This approach is especially useful when assessing qualitative differences between people, in this case, how comorbid diseases cluster together. LCA will be used to identify subtypes of individual response patterns in the data and assign a class membership probability score for each person through item-response probabilities. PROC LTA in SAS software will be used for the LCA analysis. To substantiate the empirical classifications of the comorbid symptoms, we used the CI and symptom severity scores from the QEESI as predictors of class.
| Results|| |
[Table 1] shows that there are no statistical differences between cases and controls on education or marital status. There is, however, a greater percentage of females among the cases compared to controls (61% vs. 39%, P <.0001) and cases tend to be older (54 vs. 40, P <.001). There tended to be fewer Hispanics among cases (P <.001) as well as lower income levels. These demographic variables were used as covariates in the logistic model where case status was used to predict the odds of comorbid disease.
[Table 2] shows the distribution of the 17 comorbid physician-diagnosed disease categories comparing the two groups. Within each of the 17 comorbid diseases, those very suggestive of CI demonstrated between 75% and 100% endorsement of each disease compared to between 0% and 25% among those in the not suggestive group. By Chi-square analysis, 16 of the diseases show statistically significant differences between the groups. Odds ratio and 95% confidence intervals are also shown where CI status is predicting each comorbid disease. Many of the Ns in the not suggestive group were too low to yield appropriate reliable parameter estimates.
|Table 2: Percentage of comorbid diseases compared between those with and without chemical intolerance|
Click here to view
The number of latent classes are determined by evaluating the g2 statistic, the-2LL (-2 times the log likelihood) statistic, the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). With the addition of each class, these values should decrease. In the case of g2, a difference in values after adding a class can be statistically evaluated using a Chi-square difference test and the DF (Collins and Landa, 2010). [Table 3] shows that the 2-class model fits statistically better than the one class model, and the 3-class model fits better than the 2-class model, as evidenced by significant decreases in g2 and the other measures of fit. The data do not support a 4-class model. Entropy with values approaching 1 indicates clear delineation of classes with values over 0.7 acceptable.
[Figure 1] graphically shows the probability of a “yes” response as a function of class and comorbid disease item. Probabilities close to 70% and above may be considered as belonging to a specific class. Distinct from the other classes, the first class (Class 1, 17% of the sample) is characterized by a cluster consisting of IBS, arthritis, depression, anxiety, FM, and chronic fatigue. The second class (Class 2, 53% of the sample) shows a low probability of any of the co-morbid diseases. The third class (Class 3, 30% of the sample) is characterized only by allergy, which it shares with Class 1, but otherwise has disease probabilities lower than class 1, but greater than class 3.
|Figure 1: Probability of having a comorbidity as a function of class and item|
Click here to view
[Table 4] and [Figure 2] show that those in Class 1 had the highest QEESI scores on the Chemical and Symptom scales, followed by Class 3. Both classes are considered highly suggestive of CI, with Class 1 being the most severe. Class 2 had the lowest scores, and they were not suggestive of CI. An ANOVA analysis of means using Tukey multiple comparisons indicates significant differences on the QEESI scales scores between groups (P < 0.0001).
|Table 4: Descriptive statistics of Quick Environmental Exposure and Sensitivity Inventory Scales by latent class group|
Click here to view
|Figure 2: Quick Environmental Exposure and Sensitivity Inventory Scales by Latent Class Group|
Click here to view
| Discussion|| |
To our knowledge, this study is the first to report comorbidities that empirically form a unique statistical cluster particular to CI severity. As [Table 4] shows, Class 3 represents a less-severe CI status than class 1, although the scores in this class fit within the very suggestive category. Notwithstanding, the comorbid disease profiles are distinct. For example, Class 1, associated with the highest QEESI scores, carries most of the disease burden distinct from the other two classes, and includes affective syndromes (e.g., depression and anxiety, consistent with Simon et al.) While physicians may see these comorbidities together in practice, assessing for CI may not be apparent. This suggests that when affective and physical co-morbidities are seen together, it may be an indication of assessing for CI using the QEESI.
Gender, age, and chemical intolerance
Our results are consistent with several other studies reporting a higher prevalence of CI among females compared to males.,,,,,,,,,, Plausible explanations for this difference may be biological in nature or may stem from differences in exposures to key triggers (e.g., women may be repeatedly exposed to cleaning solvents in poorly ventilated spaces, fragrances in beauty products, etc.). Further, it is well-established that males and females differ in their immune response to foreign and self-antigens. Elevated humoral immunity (immunoglobulins) in females compared to males is physiologically well-conserved. Other physiological explanations may involve the interaction between estrogen, inflammation, redox biology, mitochondria, and autoimmunity. However, the nature of these interactions as explanatory factors for gender difference in CI is unclear and requires further study.
We also found that the age of the very suggestive group was about 10 years older than the not suggestive group. Given that prior research indicates that the percentage of middle-aged female MCS patients is proportionally higher than males,,,,,, it is plausible that females may be driving the age difference results in our study.
Implications for treatment and prevention
These comorbid conditions and the multi-system symptoms exhibited by those with CI have been a clinical challenge, often leaving physicians and patients alike frustrated because of the complex interactions and limited available treatment options., The identified disease clusters (class 1) reported here, may provide potential clinical guidance when clinicians are faced with these diagnoses. It is important to use a multi-prong approach that treats the clustered conditions concurrently with CI when present. The adoption of the BREESI and QEESI screening tests would potentially provide additional options to mitigate symptoms related to environmental exposures that may aggravate the co-morbidities.
Physicians and other health care workers, who may be unaware of the potential initiators and triggers of CI, may lack the understanding that can help them look past disease symptoms alone. Understanding the potential initiators and triggers of those with CI may help physicians gain a better understanding of the potential underlying causes and may therefore assist in treatment efforts.
Miller,,, and Ashford and Miller first proposed a 2-step mechanism called TILT, that involves understanding the initiators and triggers of CI. TILT captures the wide variety of multi-system symptoms and intolerances associated with CI symptoms. TILT develops in two stages: Initiation by a major exposure event, or a series of exposures (Stage I, Initiation), followed by triggering of multisystem symptoms in response to everyday chemical inhalants, foods/food additives, and/or medications/drugs (Stage II, Triggering). Initiating exposures include chemical spills, pesticides, cleaning agents, solvents, mold, combustion products, medications, and medical devices (such as implants), and indoor air contaminants associated with materials used in construction or remodeling. For greater detail about the TILT mechanism see Ashford and Miller and Masri et al.
The number and type of comorbid disease clusters associated with CI/TILT may be important in two ways: First, clinicians may use our finding of clustered co-morbidities as prompts to explore the potential co-existence of CI and thus activate mitigating maneuvers, and second, the clusters may point to common etiological pathways such as Mast Cell Activation Syndrome (MCAS), which has recently been shown to be associated with CI. Potential therapeutic drug options might then be explored.,, The “comorbidity clusters” among those very suggestive of CI suggest that even patients who fit time-honored consensus diagnostic criteria merit testing with the QEESI and the taking of careful exposure histories to help determine whether TILT and xenobiotic sensitization of mast cells may have occurred.
Avoidance of TILT initiators and triggers is important for the prevention and treatment of CI. Initiators and triggers that remain unaddressed can perpetuate symptoms, leading to “unexplained syndromes” or “idiopathic illnesses.” When people who are depressed, irritable, or anxious attribute their conditions to chemical exposures, they are often referred to mental health practitioners and are apt to receive psychiatric or psychological diagnoses such as Somatic Symptom Disorder” (DSM-5) rather than medical intervention., As such, all potential etiologies including TILT should be considered.
Choosing which comorbidities to include in this study was based on the most common ones associated with inflammatory or other immune function processes. Clearly, we did not include an exhaustive list. For example, we included asthma, but not more refined variations like “airway inflammation.” One concern was the response burden of the participant. As such we may have missed some important distinctions. Further, we did not assess the length of time the participant had the co-morbidities. The length of time would have been a plausible covariate in the logistic regression models to adjust for potential confounding of disease length of time. Further, the generalizability of these results should be taken cautiously as our sample size was relatively small and not drawn from a population sample. Additional study with a larger population-based sample is warranted.
| Conclusion|| |
Various medical specialties are involved with the CI comorbidities described in this paper, these include pulmonologists (asthma), neurologists (migraine headaches), rheumatologists (FM), immunologists (CFS), psychiatrists/psychologists (depression, anxiety), and all of these comorbidities are treated or referred by primary care physicians. With increasing prevalence of CI, it would be prudent to suggest that clinicians and researchers screen for chemical intolerance (using the BREESI and confirming with the QEESI), take detailed exposure histories, and ask about potential initiating exposure events for any patients to whom they are tempted to assign descriptive, nonetiologic labels. This process may provide an avenue for effective treatments in some who suffer from CI.
Institutional review board statement
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the University of Texas Health Science Center San Antonio Institutional Review Board (approval number HSC20200718N).
Informed consent statement
Written informed consent was waived due to completely anonymous volunteer participation.
Data availability statement
The dataset analyzed during the current study is available from the corresponding author on reasonable request.
We appreciate the Marilyn Brachman Hoffman Foundation, Fort Worth, Texas (TX), for their generous support of this research.
Financial support and sponsorship
This research was funded by Marilyn Brachman Hoffman Foundation, Fort Worth, Texas.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Ashford N, Miller C. Chemical Exposures: Low Levels and High Stakes. New York: Von Nostrand Reinhold; 1998.
Miller CS. Toxicant-induced loss of tolerance – An emerging theory of disease? Environ Health Perspect 1997;105 Suppl 2:445-53.
Miller CS. The compelling anomaly of chemical intolerance. Ann N Y Acad Sci 2001;933:1-23.
Genuis SJ. Sensitivity-related illness: The escalating pandemic of allergy, food intolerance and chemical sensitivity. Sci Total Environ 2010;408:6047-61.
Masri S, Miller CS, Palmer RF, et al
. Toxicant-induced loss of tolerance for chemicals, foods, and drugs: assessing patterns of exposure behind a global phenomenon. Environ Sci Eur 33, 65 (2021). https://doi.org/10.1186/s12302-021-00504-z
[Last accessed on 2021 Dec 16].
Miller CS, Mitzel HC. Chemical sensitivity attributed to pesticide exposure versus remodeling. Arch Environ Health 1995;50:119-29.
Proctor SP. Chemical sensitivity and gulf war veterans' illnesses. Occup Med 2000;15:587-99.
Martini A, Iavicoli S, Corso L. Multiple chemical sensitivity and the workplace: Current position and need for an occupational health surveillance protocol. Oxid Med Cell Longev 2013;2013:351457.
Azuma K, Uchiyama I, Katoh T, Ogata H, Arashidani K, Kunugita N. Prevalence and characteristics of chemical intolerance: A Japanese population-based study. Arch Environ Occup Health 2015;70:341-53.
Caress SM, Steinemann AC. A national population study of the prevalence of multiple chemical sensitivity. Arch Environ Health 2004;59:300-5.
Palmer RF, Walker T, Kattari D, Rincon R, Perales RB, Jaén CR, et al.
Validation of a brief screening instrument for chemical intolerance in a large U.S. national sample. Int J Environ Res Public Health 2021;18:8714.
Dantoft TM, Nordin S, Andersson L, Petersen MW, Skovbjerg S, Jørgensen T. Multiple chemical sensitivity described in the Danish general population: Cohort characteristics and the importance of screening for functional somatic syndrome comorbidity – The DanFunD study. PLoS One 2021;16:e0246461.
Pigatto PD, Guzzi G. Prevalence and risk factors for multiple chemical sensitivity in Australia. Prev Med Rep 2019;14:100856.
Hojo S, Mizukoshi A, Azuma K, Okumura J, Ishikawa S, Miyata M, et al.
Survey on changes in subjective symptoms, onset/trigger factors, allergic diseases, and chemical exposures in the past decade of Japanese patients with multiple chemical sensitivity. Int J Hyg Environ Health 2018;221:1085-96.
Steinemann A. National prevalence and effects of multiple chemical sensitivities. J Occup Environ Med 2018;60:e152-6.
Rossi S, Pitidis A. Multiple chemical sensitivity: Review of the state of the art in epidemiology, diagnosis, and future perspectives. J Occup Environ Med 2018;60:138-46.
Hojo S, Kumano H, Yoshino H, Kakuta K, Ishikawa S. Application of quick environment exposure sensitivity inventory (QEESI) for Japanese population: Study of reliability and validity of the questionnaire. Toxicol Ind Health 2003;19:41-9.
Jeon BH, Lee SH, Kim HA. A validation of the Korean version of QEESI© (The quick environmental exposure and sensitivity inventory). Korean J Occup Environ Med 2012;24:96-114.
Nordin S, Andersson L. Evaluation of a Swedish version of the quick environmental exposure and sensitivity inventory. Int Arch Occup Environ Health 2010;83:95-104.
Palmer RF, Jaén CR, Perales RB, Rincon R, Forster JN, Miller CS. Three questions for identifying chemically intolerant individuals in clinical and epidemiological populations: The brief environmental exposure and sensitivity inventory (BREESI). PLoS One 2020;15:e0238296.
Miller CS, Prihoda TJ. The environmental exposure and sensitivity inventory (EESI): A standardized approach for measuring chemical intolerances for research and clinical applications. Toxicol Ind Health 1999;15:370-85.
Miller CS, Prihoda TJ. A controlled comparison of symptoms and chemical intolerances reported by Gulf War veterans, implant recipients and persons with multiple chemical sensitivity. Toxicol Ind Health 1999;15:386-97.
Buchwald D, Garrity D. Comparison of patients with chronic fatigue syndrome, fibromyalgia, and multiple chemical sensitivities. Arch Intern Med 1994;154:2049-53.
Katerndahl DA, Bell IR, Palmer RF, Miller CS. Chemical intolerance in primary care settings: Prevalence, comorbidity, and outcomes. Ann Fam Med 2012;10:357-65.
Lind N, Söderholm A, Palmquist E, Andersson L, Millqvist E, Nordin S. Comorbidity and multimorbidity of asthma and allergy and intolerance to chemicals and certain buildings. J Occup Environ Med 2017;59:80-4.
Johnson D, Colman I. The association between multiple chemical sensitivity and mental illness: Evidence from a nationally representative sample of Canadians. J Psychosom Res 2017;99:40-4.
Del Casale A, Ferracuti S, Mosca A, Pomes LM, Fiaschè F, Bonanni L, et al.
Multiple chemical sensitivity syndrome: A principal component analysis of symptoms. Int J Environ Res Public Health 2020;17:6551.
Eis D, Helm D, Mühlinghaus T, Birkner N, Dietel A, Eikmann T, et al.
The German multicentre study on multiple chemical sensitivity (MCS). Int J Hyg Environ Health 2008;211:658-81.
Eliasen M, Jørgensen T, Schröder A, Dantoft TM, Fink P, Poulsen CH, et al.
Somatic symptom profiles in the general population: A latent class analysis in a Danish population-based health survey. Clin Epidemiol 2017;9:421-33.
Eliasen M, Schröder A, Fink P, Kreiner S, Dantoft TM, Poulsen CH, et al.
step towards a new delimitation of functional somatic syndromes: A latent class analysis of symptoms in a population-based cohort study. J Psychosom Res 2018;108:102-17.
Collins LM, Lanza ST. Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. New York, NY: Wiley; 2010.
Lanza ST, Collins LM, Lemmon DR, Schafer JL. PROC LCA: A SAS procedure for latent class analysis. Struct Equ Modeling 2007;14:671-94.
Celeux G, Soromenho G. An entropy criterion for assessing the number of clusters in a mixture model. J Classif 1996;13:195-212.
Simon GE, Daniell W, Stockbridge H, Claypoole K, Rosenstock L. Immunologic, psychological, and neuropsychological factors in multiple chemical sensitivity. A controlled study. Ann Intern Med 1993;119:97-103.
Dupas D, Dagorne MA. Multiple chemical sensitivity: A diagnosis not to be missed. Rev Mal Respir 2013;30:99-104.
Fitzgerald DJ. Studies on self-reported multiple chemical sensitivity in South Australia. Environ Health 2008;8:33-9.
Carlsen KH, Topp AM, Skovbjerg S. (2012) Living with a Chemically Sensitive Wife: A “We” Situation”, International Scholarly Research Notices. vol. 2012, Article ID 285623, 6 pages. https://doi.org/10.5402/2012/285623
[Last accessed on 2021 Dec 16].
Kreutzer R, Neutra RR, Lashuay N. Prevalence of people reporting sensitivities to chemicals in a population-based survey. Am J Epidemiol 1999;150:1-12.
Fink AL, Klein SL. The evolution of greater humoral immunity in females than males: Implications for vaccine efficacy. Curr Opin Physiol 2018;6:16-20.
Di Florio DN, Sin J, Coronado MJ, Atwal PS, Fairweather D. Sex differences in inflammation, redox biology, mitochondria and autoimmunity. Redox Biol 2020;31:101482.
Bell IR, Miller CS, Schwartz GE. An olfactory-limbic model of multiple chemical sensitivity syndrome: Possible relationships to kindling and affective spectrum disorders. Biol Psychiatry 1992;32:218-42.
Hausteiner C, Bornschein S, Hansen J, Zilker T, Förstl H. Self-reported chemical sensitivity in Germany: A population-based survey. Int J Hyg Environ Health 2005;208:271-8.
Fiedler N, Kipen H. Chemical sensitivity: The scientific literature. Environ Health Perspect 1997;105 Suppl 2:409-15.
Hojo S, Ishikawa S, Kumano H, Miyata M, Sakabe K. Clinical characteristics of physician-diagnosed patients with multiple chemical sensitivity in Japan. Int J Hyg Environ Health 2008;211:682-9.
Miller CS, Palmer RF, Dempsey TT, Ashford, N. Mast cell activation may explain many cases of chemical intolerance. Environ Sci Eur 33, 129 (2021). https://doi.org/10.1186/s12302-021-00570-3
[Last accessed 2021 Dec 16].
Molderings GJ, Haenisch B, Brettner S, Homann J, Menzen M, Dumoulin FL, et al
. Pharmacological treatment options for mast cell activation disease. Naunyn Schmiedebergs Arch Pharmacol 2016;389:671-94.
Molderings GJ, Brettner S, Homann J, Afrin LB. Mast cell activation disease: A concise practical guide for diagnostic workup and therapeutic options. J Hematol Oncol 2011;4:10.
Wirz S, Molderings GJ. A practical guide for treatment of pain in patients with systemic mast cell activation disease. Pain Physician 2017;20:E849-61.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]