Studies on Self-Reported Multiple Chemical Sensitivity

in South Australia
D. James Fitzgerald

Scientific Services, Department of Health, Adelaide

The prevalence of Multiple Chemical Sensitivity (MCS) in South Australia is

unknown and was sought through population-based telephone surveys of
approximately 4000 adults. These surveys revealed a 1% self-reported MCS
prevalence but also a more general hypersensitivity prevalence of about
16%. Symptomology and symptom severity suggest a significant negative
impact of environmental chemicals in the community.
Key words: Multiple Chemical Sensitivity; Hypersensitivity
Multiple Chemical Sensitivity (MCS) is clinical management guidelines for MCS
a chronic condition characterised by in Australia. Yet it is evident that some
fatigue, headaches, fibromyalgia, anxiety, medical practitioners attempt to diagnose
nausea, depression, dizziness and various and treat MCS.
other non-specific symptoms (Graveling In the late 1990s, the South Australian
et al. 1999; Labarge & McCaffrey 2000; Department of Health became increasingly
Pall 2007). Sufferers consider that low aware of cases of MCS. At that time, claims
doses of a wide range of environmental were being made of a high prevalence of
chemicals can trigger these symptoms, MCS in the community. In order to inform
though such causation is scientifically this issue, the South Australian Department
difficult to prove (Bornschein et al. 2007; of Health commissioned two randomised
Staudenmayer 2001). Due to this, MCS population-based surveys the results of
is a controversial condition, and it is which form the basis of this paper.
reported as having aspects of toxicogenic
and psychogenic aetiology. Detractors Methods
from a toxicogenic origin prefer the Self-reporting data were obtained from
descriptor Idiopathic Environmental computer-aided telephone interviewing
Intolerance (Staudenmayer et al. 2003a,b) (CATI), arranged through the South
and have demonstrated psychiatric Australian Department of Health’s
comorbidity among patients with MCS, Population Research & Outcome Studies
including anxiety, panic disorder and Unit in collaboration with Harrison
depression (Bailer et al. 2004; Bornschein Health Research, Adelaide. Responses
et al. 2002; Caccappolo-van Vliet et al. to questions were entered directly into
2002). Supporters of a toxicogenic origin the computer and the CATI system
have given consideration to a range of enforces a range of checks with most
mechanisms, including toxicant-induced questions having a set of predetermined
loss of tolerance (Miller 2000), elevated response categories. Response categories
nitric oxide/peroxynitrile (Pall 2003, 2007), can also be automatically rotated when
immunological dysregulation, neurogenic required, to minimise bias. Open-ended
inflammation, and limbic kindling/neural responses were recorded verbatim by the
sensitisation (Graveling et al. 1999). interviewer.
Currently, there are no biomarkers for The survey methodology is reported in
MCS, and there are no diagnostic or detail elsewhere (Population Research &
Vo l . 8 N o . 3 2 0 0 8 e n v i r o n m e n ta l h e a lt h 33
D. James Fitzgerald

Outcomes Studies Unit 2002); however, of MCS, that is, fibromyalgia. Otherwise,
in brief, all households in South Australia, the two questions are the same. In both
with a number listed in the Electronic phases, the incidence of adult asthma
White Pages were eligible for selection was similar (11.5% and 11.7%), and
in the sample. Telephone numbers were corresponded to the known incidence
selected randomly and approximately of asthma in the community (Wilson et al.
2000 interviews were conducted on 2006), indicating that the sampled cohorts
people aged 18 years and over (n=2007 were representative of the population
in September 2002 [Phase 1] and n=2002 at large.
in June 2004 [Phase 2]). A letter was sent The MCS prevalence data from these
to each selected household introducing initial questions are shown in Table 1.
the survey. Within each household, the These reveal an MCS prevalence of 0.7 to
person who had their birthday last was 1.0% in the adult population. The slightly
selected for interview. There was no lower rate in Phase II could reflect the
replacement for non-respondents. Data use of the present tense in the question,
were weighted by probability of selection and which, though the number of cases is
in the household and by age, sex and small, might suggest a degree of recovery
area of residence to the most recent
from the condition.
Australian Bureau of Statistics Estimated
Resident Population for South Australia Table 1: MCS prevalence and gender data
for 30 June 2001 (Phase I) or for 30 June from Phase I and Phase II
2002 (Phase II). Phase1 Male Female Total % of total
population2
Results and Discussion I 4 17 21 1.0%
II 4 10 14 0.7%
Prevalence, age and gender
distribution Total 8 27 35 0.87%
Notes:
Phase I and Phase II each included
1 See text for specific questions
one key question to determine MCS 2 Based on 2007 people interviewed in Phase I, and 2002 in
prevalence: Phase II

Within the confines of self-reporting, it

Phase I has been shown that false negatives might
“Have you ever been told by a doctor that comprise about 50% of the true number of
you have any of the following conditions? cases (Baker et al. 2004). It is thus possible
- asthma, other respiratory problems, that an MCS prevalence of up to 2% or
chronic fatigue syndrome, heart disease, down to 0.5% actually exists in the South
multiple chemical sensitivity?” Australian community.
This, therefore, is the first attempt to
Phase II gain an understanding of MCS prevalence
“Have you been told by a medical doctor in South Australia. Similar prevalence has
that you currently have any of the following been reported in Denmark (Danish Ministry
conditions? - asthma, other respiratory of the Environment 2006), while several
problems, chronic fatigue syndrome, heart other surveys report a prevalence of 2%-6%
disease, multiple chemical sensitivity, (Caress & Steinemann 2003; Gibson 2006
fibromyalgia (muscle pain)?” Kreutzer et al. 1999; Meggs et al. 1996)
The Phase II question thus specified although some sought a diagnosis of ‘MCS
the type of doctor, a current diagnosis, or environmental illness’. In a NSW survey,
and added one of the common symptoms with the question “Have you ever been
34 e n v i r o n m e n ta l h e a lt h Vo l . 8 N o . 3 2 0 0 8
 Studies on Self-Reported Multiple Chemical Sensitivity in South Australia

diagnosed with a chemical sensitivity?”, the These data indicate a greater proportion
prevalence was 2.9% (NSW Public Health of female MCS cases, being 4.25-fold in
2003). It must be borne in mind that the Phase I and 2.5-fold in Phase II (average
survey data depend on the question that 3.4-fold overall; total / = 1.05). This
is asked, on how the medical profession could be explained if more females than
view and diagnose chemical sensitivity, males with the condition visited doctors;
and on how patients interpret and relate a nevertheless this predilection toward
medical diagnosis. female cases is in accordance with other
Regarding age distribution, the MCS surveys (Caress & Steinemann 2003;
cases from both phases were summed, Joffres et al. 2001; Kreutzer et al. 1999;
were age-stratified and the prevalence NSW Public Health 2003). If this is a real
calculated per population size in seven phenomenon, it suggests an underlying
age groupings. Figure 1 shows no cases gender-specific mechanism.
in the 18-24 year group, then a Gaussian-
type distribution with peak prevalences of City versus country
1.5%, 1.12% and 1.53% in the 45-54 year It has been suggested that compared to
group, 55-64 year group, and the 65-74 city environments, country environments
year group, respectively. These data may are less polluted and would, therefore,
suggest a late onset of MCS, though this be less likely to impact on chemically-
survey did not include adolescents or sensitive individuals. In reality, many
children. This contrasts with the significant country environments are subject to regular
number of hypersensitive cases in these agricultural spraying. Notwithstanding,
early age groups (data not shown; Caress the present survey examined this issue,
& Steinemann 2003). revealing an MCS prevalence of 0.8%

Figure 1: Age distribution of combined MCS prevalence data. Total number of cases in each
age group was divided by the total number of individuals in the corresponding age group

Vo l . 8 N o . 3 2 0 0 8 e n v i r o n m e n ta l h e a lt h 35
D. James Fitzgerald

in metropolitan Adelaide and 1.1% in among hypersensitive individuals of

country SA. A lack of significant difference decreased prevalence with increasing
between city and country MCS prevalence household income was also reported by
was also found in a NSW survey (NSW others (Caress & Steinemann 2003; Joffres
Public Health 2003). et al. 2001), with a bi-modal relationship
Anecdotally, some MCS sufferers move reported by Kreutzer and colleagues
to the country to seek ‘cleaner’ air, thus (Kreutzer et al. 1999).
these data might reflect this demographic.
Alternatively, the data could simply Other MCS-specific questions
suggest that country environments might Though the number of cases was small,
not be healthier for those with MCS. a range of other questions was posed to
those identifying with MCS in Phase II
Household income (Table 2). The responses tend to confirm
The surveys included a question to all what is generally known, namely that
respondents regarding household income, stress might be a major aetiological
and Figure 2 shows MCS prevalence factor in MCS onset and that the family
as a function of this parameter. The or social life of MCS sufferers is often
data indicate a significant trend towards significantly affected. Regarding those
decreased MCS prevalence with increasing cases who could identify the origin
household income. Unfortunately the of their sensitivity, it has been observed
data did not lend itself to determining elsewhere that such cases were more
the gender stratification by household likely to report severe symptoms than
income, but this would be interesting to those who did not know the original
ascertain in future. This same relationship cause (Caress & Steinemann 2003).

Figure 2: MCS prevalence as a function of household income (Phase I and II combined)

36 e n v i r o n m e n ta l h e a lt h Vo l . 8 N o . 3 2 0 0 8
 Studies on Self-Reported Multiple Chemical Sensitivity in South Australia

Table 2: MCS-specific questions in Phase II males and 21.7% of females responded in

the affirmative, with an average of 15.9%.
Question Response This gender bias and average are similar to
other reports (Caress & Steinemann 2003;
Were you under any 3/4 males Yes
particular stress Kreutzer et al. 1999).
2/10 females Yes
at the time when Considering hypersensitivity in
you first developed country versus metropolitan areas,
symptoms of MCS? the prevalence was 15.3% and 16.1%,
Do you have any 11/14 Yes respectively. Considering symptoms
idea what initially reported by hypersensitive individuals,
caused your chemical 40% experienced headaches, 37% had
sensitivity? asthma or other breathing problems,
To what extent does 5/14 To a great 31% had burning eyes, nose or throat,
your condition affect extent 18% had nausea or stomach problems,
your family or social 2/14 To some extent 17% had eczema, 9% had fatigue and
life?
9% experienced dizziness or fever as
a result of chemical exposure. These
are, therefore, not insignificant reactions.
General chemical hypersensitivity
Further, 8.4% of hypersensitive males
Both phases of this investigation included and 15.7% hypersensitive females
questions to all respondents about general considered that their symptoms were
sensitivity to environmental chemicals. moderate to severe.
The Phase I question and data are given In response to the question, “Have you
in Table 3, revealing prevalence of 3-10% received any medical treatment for your
across the genders for perfume, traffic chemical sensitivity?”, 15.3% of males and
pollution, household and workplace 31.9% of females within the hypersensitive
chemicals. subset answered in the affirmative.
In Phase II, the question posed to non- Together, this represents an overall 4.3%
MCS cases was different: “Do you consider of the total population seeking medical
yourself especially sensitive to everyday treatment due to chemical sensitivity. This
chemicals found in household cleaning is similar to the 6.7% reported by others
products, perfumes, insect sprays, new (Caress & Steinemann 2003).
To ascertain more specifically the
carpets, fresh paints, etc?” To this, 9.9% of
chemical triggers involved, respondents
Table 3: Phase I data on general chemical who identified with hypersensitivity were
hypersensitivity1 asked about specific chemical classes.
Is your health % % % Data in Table 4 indicate, first, that many
seriously affected by males females total individuals were affected by more than
exposure to any of one chemical (and, therefore, could be
the following? undiagnosed MCS cases), and second, that
Perfume 4.5 9.6 7.1 a wide range of common environmental
Traffic pollution 5.3 6.4 5.9 agents must be avoided to reduce risk of
Household 2.8 8.2 5.6 adverse reaction.
chemicals Given that MCS is a controversial area of
Workplace 7.2 5.2 6.2
environmental medicine, a final question
chemicals was put to Phase II non-MCS respondents:
Notes:
“Do you agree or disagree with the
1 Based on 2007 people interviewed following statement? “Chemical sensitivity
Vo l . 8 N o . 3 2 0 0 8 e n v i r o n m e n ta l h e a lt h 37
D. James Fitzgerald

Table 4: Phase II data on chemical triggers of hypersensitivity

Which of these chemicals currently trigger any of % % %
your symptoms? males females hypersens
population
Perfumes, etc. 66 90 82.5
Tobacco smoke 29 48 42.2
New building or renovation 34 43 40.4
Pesticides or herbicides 24 36 32.7
Petrochemicals 22 36 32.0
Vehicle smoke 17 32 27.1
Other chemicals 19 19 19.0

is a valid health condition with valid indicate that about 16% of the adult
symptoms.” An overwhelming 86% agreed population identifies as having some
or strongly agreed with this statement. chemical hypersensitivity. Since there are
This may augur well for achieving no diagnostic or clinical guidelines for
success when those with MCS or chemical MCS in Australia, it is possible that the
hypersensitivity seek understanding from
1% MCS prevalence is an under-reporting,
the wider community and when plans are
and that some chemically hypersensitive
implemented to reduce specific chemical
exposures in the community. individuals have symptomology more
aligned with that of MCS cases. The
Conclusion prevalence of hypersensitivity and the
These two population-based surveys severity of symptoms suggest an adverse
reveal a self-reported MCS prevalence effect of common environmental chemicals
in South Australia of about 1% and also in a significant portion of the population.

Acknowledgments
The author acknowledges Sam Mangas for technical assistance; Peter Evans and Cathie
Powell who assisted with question design in Phase II; and staff of the Department of
Health’s Population Research & Outcome Studies Unit who aided study execution and
statistical analysis.

References
Bailer, J., Rist, F., Witthöft, M., Paul, C. & Bayerl, C. 2004, ‘Symptom patterns, and perceptual
and cognitive styles in subjects with multiple chemical sensitivity (MCS)’, Journal of
Environmental Psychology, vol. 24, pp. 517-25.
Baker, M., Stabile, M. & Deri, C. 2004, ‘What do self-reported, objective, measures of health
measure?’, Journal of Human Resources, vol. 30, pp. 1067-93.
Bornschein, S., Hausteiner, C., Zilker T. & Förstl, H. 2002, ‘Psychiatric and somatic disorders
and multiple chemical sensitivity (MCS) in 264 ‘environmental patients’, Psychological
Medicine, vol. 32, pp. 1387-94.
Bornschein, S., Hausteiner, C., Drzezga, A.,Theml T., Heldmann, B. & Grimmer, T. et al.
2007, ‘Neuropsychological and positron emission tomography correlates in idiopathic
environmental intolerances’, Scandinavian Journal of Work and Environmental Health,
vol. 33, pp. 447-53.
Caccappolo-van Vliet, E., Kelly-McNeil, K., Natelson, B., Kipen, H. & Fielder, N. 2002, ‘Anxiety
sensitivity and depression in multiple chemical sensitivities and asthma’, Journal of
Occupational and Environmental Medicine, vol. 44, pp. 890-901.

38 e n v i r o n m e n ta l h e a lt h Vo l . 8 N o . 3 2 0 0 8
 Studies on Self-Reported Multiple Chemical Sensitivity in South Australia

Caress, S.M. & Steinemann, A.C. 2003, ‘A review of a two-phase population study of multiple
chemical sensitivities’, Environmental Health Perspectives, vol. 111, pp. 1490-1497.
Danish Ministry of the Environment. 2006, ‘News’, <www.mim.dk/eng/News/170504_Rentolin.
htm> 3 October 2008.
Gibson, P.R. 2006, ‘Understanding and accommodating people with Multiple Chemical Sensitivity
in independent living’, <www.ilru.org/html/publications/bookshelf/MCS.html> 3 October
2008.
Graveling, R.A., Pilkington, A., George, J.P.K., Butler, M.P. & Tannahill, S.N. 1999, ‘A review of
multiple chemical sensitivity’, Occupational & Environmental Medicine, vol. 56, pp. 73-
85.
Joffres, M.R., Williams, T., Sabo, B. & Fox, R.A. 2001, ‘Environmental sensitivities: Prevalence of
major symptoms in a referral center, The Nova Scotia Environmental Sensitivities Research
Center study’, Environmental Health Perspectives, vol. 109, pp. 161-165.
Kreutzer, R., Neutra, R.R. & Lashuay, N. 1999, ‘Prevalence of people reporting sensitivities to
chemicals in a population-based survey’, American Journal of Epidemiology, vol. 150,
pp. 1-12.
Labarge, A.S. & McCaffrey, R.J. 2000, ‘Multiple Chemical Sensitivity: A review of the theoretical
and research literature’, Neuropsychology Review, vol. 10, pp. 183-211.
Meggs, W.J., Dunn, K.A., Bloch, R.M., Goodman, P.E. & Davidoff, A.L. 1996, ‘Prevalence and nature
of allergy and chemical sensitivity in a general population’, Archives of Environmental
Health, vol. 51, pp. 275-82.
Miller, C.S. 2000, ‘Mechanisms of action of addictive stimuli: Toxicant-induced loss of tolerance’,
Addiction, vol. 96, pp. 115-139.
NSW Public Health 2003, ‘The New South Wales Adult Health Survey 2002’, NSW Public Health
Bulletin Supplement, vol. 14, no. S-4, pp. 81-2.
Pall, M.L. 2003, ‘Elevated nitric oxide/peroxynitrate theory of multiple chemical sensitivity: Central
role of N-methyl-D-aspartate receptors in the sensitivity mechanism’, Environmental
Health Perspectives, vol. 111, pp. 1461-1464.
Pall, M.L. 2007, ‘Explaining “Unexplained Illnesses”’, Haworth Press, New York.
Population Research & Outcomes Studies Unit 2002, The Health Monitor Survey Methodology,
Department of Health, Adelaide 2002, <http://www.health.sa.gov.au/pros/portals/0/
BR%202002-12%20Health%20Monitor%20Methodology.pdf>
Staudenmayer, H. 2001, ‘Idiopathic environmental intolerances (IEI): Myth and reality’, Toxicology
Letters, vol. 120, pp. 333-42.
Staudenmayer, H., Binkley, K.E., Leznoff, A. & Phillips, S. 2003a, ‘Idiopathic Environmental
Intolerance, Part 1: A causation analysis applying Bradford Hill’s criteria to the toxicogenic
theory’, Toxicology Reviews, vol. 22, pp. 235-46.
Staudenmayer, H., Binkley, K.E., Leznoff, A. & Phillips, S. 2003b, ‘Idiopathic Environmental
Intolerance, Part 2: A causation analysis applying Bradford Hill’s criteria to the psychogenic
theory’, Toxicology Reviews, vol. 22, pp. 247-61.
Wilson, D.H., Adams, R.J., Tucker, G., Appleton, S., Taylor, A.W. & Ruffin, R.E. 2006, ‘Trends
in asthma prevalence and population changes in South Australia, 1990-2003’, Medical
Journal of Australia, vol. 184, pp. 226-9.

Correspondence to:
Jim Fitzgerald
Scientific Services
Department of Health
PO Box 6, Rundle Mall
Adelaide, South Australia, 5000
Email [email protected]

Back to TOC

Vo l . 8 N o . 3 2 0 0 8 e n v i r o n m e n ta l h e a lt h 39