Content uploaded by Erica Salomone
Author content
All content in this area was uploaded by Erica Salomone on Apr 17, 2015
Content may be subject to copyright.
ORIGINAL ARTICLE
Prevalence and correlates of use of complementary
and alternative medicine in children with autism spectrum
disorder in Europe
Erica Salomone
1
&Tony Charman
1
&Helen McConachie
2
&Petra Warreyn
3
&
Working Group 4, COST Action ‘Enhancing the Scientific Study of Early Autism’
Received: 21 November 2014 /Revised: 26 February 2015 /Accepted: 19 March 2015
#Springer-Verlag Berlin Heidelberg 2015
Abstract This study examined the prevalence and correlates
of use of complementary and alternative medicine (CAM)
among a sample of children with autism spectrum dis-
order (ASD)<7 years in 18 European countries (N=1,
680). Forty-seven percent of parents reported having
tried any CAM approach in the past 6 months. Diets
and supplements were used by 25 % of the sample
and mind–body practices by 24 %; other unconventional
approaches were used by 25 % of the families, and a
minority of parents reported having tried any invasive or po-
tentially harmful approach (2 %). Parents in Eastern Europe
reported significantly higher rates of CAM use. In the total
sample, children with lower verbal ability and children
using prescribed medications were more likely to be
receiving diets or supplements. Concurrent use of high
levels of conventional psychosocial intervention was
significantly associated with use of mind–body practices.
Higher parental educational level also increased the likelihood
of both use of diets and supplements and use of mind–body
practices.
Conclusion: The high prevalence of CAM use among a
sample of young children with ASD is an indication that
Communicated by Jaan Toelen
The Working Group 4 also includes: Anett Kaale, anett.kaale@r-bup.no
(Norway); Bernadette Rogé, roge@univ-tlse2.fr and Frederique
Bonnet-Brilhaut, frederique.brilhault@univ-tours.fr (France), Iris
Oosterling, i.oosterling@karakter.com (the Netherlands), Selda
Ozdemir, seldaozdemir@gazi.edu.tr (Turkey), Antonio Narzisi,
antonio.narzisi@inpe.unipi.it and Filippo Muratori f.muratori@inpe.unipi.it,
(Italy), Joaquin Fuentes, fuentes.j@telefonica.net (Spain), Mikael
Heimann mikael.heimann@liu.se, (Sweden), Michele Noterdaeme,
noterdaeme.michele@josefinum.de, Christine Freitag, Christine
Margarete.Freitag@kgu.de, Luise Poustka, Luise.Poustka@zi-mannheim.de
and Judith Sinzig, judith.sinzig@lvr.de (Germany), Sue Fletcher-Watson,
sfwatson@staffmail.ed.ac.uk and Jonathan Green, jonathan.green@
manchester.ac.uk (the UK).
*Erica Salomone
erica.salomone@unito.it
Tony Charman
tony.charman@kcl.ac.uk
Helen McConachie
helen.mcconachie@newcastle.ac.uk
Petra Warreyn
petra.warreyn@ugent.be
1
Department of Psychology, King’s College London, Institute of
Psychiatry, Psychology and Neuroscience, London, UK
2
Instituteof Health and Society, Newcastle University, Newcastle, UK
3
Department of Experimental Clinical and Health Psychology, Ghent
University, Ghent, Belgium
Eur J Pediatr
DOI 10.1007/s00431-015-2531-7
parents need to be supported in the choice of treatments early
on in the assessment process, particularly in some parts of
Europe.
What is Known:
•Use of complementary and alternative medicine (CAM) in children with
autism spectrum disorder is common.
•In non-EU samples, parents with higher educational level and parents of
low functioning children are more likely to use CAM with their children.
What is New:
•This study provides the first data on prevalence and correlates of use of
CAM approaches in a large sample of young children with autism in
Europe (N =1,680).
•Rates of CAM use were particularly high in Eastern Europe and
correlates of use varied by type of CAM across Europe.
Keywords Autism spectrum disorder .Complementary and
alternative medicine .Europe .Diets .Supplements .
Mind–body practices
Abbreviations
ANOVA Analysis of variance
ASD Autism spectrum disorder
CAM Complementary and alternative medicine
CI Confidence intervals
IQR Interquartile range
NCCAM National Center for Complementary and
Alternative Medicine
OR Odds ratio
RCT Randomisedcontrolledtrial
SES Socio-economic status
Introduction
Autism spectrum disorder (ASD) is a behaviourally defined
disorder characterised by impairments in social communication
abilities and the presence of restricted and repetitive behaviours
and atypical sensory responses [3]. Despite the evidence that
behavioural and social communication interventions can ame-
liorate symptoms and improve outcomes [29], it is not a con-
dition for which a ‘cure’is currently available. The uncertainty
concerning the developmental outcomes, the limitations to
existing treatments and the lack of a simple cure have been
indicated as possible reasons for the high prevalence of use of
therapies based outside the domain of conventional medical
and psychological practice by families of children with autism
[21]. Such therapies, generally defined as complementary and
alternative medicine (CAM), comprise a myriad of ‘interven-
tions’that range from unproven and untested treatments to
approaches that have been found to be harmful. The National
Center for Complementary and Alternative Medicine
(NCCAM) distinguishes the following broad areas of CAM:
‘natural products’(often sold as dietary supplements), ‘mind
and body practices’(such as massage or sensory integration
therapy) and a residual category of ‘other’complementary
health approaches that do not fit neatly in the previous ones,
such as homeopathy (http://nccam.nih.gov/). Research on
CAM use broadly refers back to this classification, but
additional meaningful categories of CAM such as ‘invasive or
potentially unsafe approaches’[1] and other unconventional
approaches that are not strictly classifiable as CAM (such as
pet therapy) are also often included in such surveys. This, and
the fact that the NCCAM classification has changed over time,
have led to some inconsistency across studies.
The efficacy of CAM treatment is controversial, but, for
most of these approaches, there simply is not enough evidence
to evaluate them [23]. For example, while gluten- and casein-
free diets are widely used and reported to be efficacious by
parents [40], to date, only two RCTs have tested their efficacy,
yielding mixed results that prevent any recommendation of
these exclusion diets as standard treatments [8,19]. Omega-
3 fatty acids are increasingly used in ASD despite lack of
understanding on which might be the optimal dosage and
insufficient evidence of efficacy [18]. Moreover, while CAM
is often used in combination with medication, little is known
about potential aversive effects of the interaction between
drugs and supplements, which requires careful monitoring
[20]. There is some positive evidence for some CAM ap-
proaches, such as horse-riding [16]andmassage[34]. A
Cochrane review of auditory integration training, a costly
and theoretically ill-specified treatment, did not find sufficient
evidence to support its use [35].
In US-based samples, there is some evidence that CAM use
in children or young people with ASD is associated with great-
er functional difficulty [15,32,38], but this has not always
been replicated [1]. A higher parental educational level and
high levels of use of conventional therapy (more than 20 hours
per week) have also been found to be associated with CAM use
in children with ASD [1]. Cultural and systemic factors (such
as families’own recognition and beliefs around aetiology and
course of symptoms as well as the actual availability of con-
ventional therapy) might also play a role in the decision to use
CAM [25]. Professionals’opinions vary widely on the topic
[31] and might be another source of influence on family
choice. These aspects are likely to differ in different parts of
the world [4]; however, with the exception of a non-systematic
review based on professionals’opinions [41], no studies
to date report on the use of CAM in Europe. Moreover,
different factors might play a different role in use of
specific types of CAM, but this is only beginning to
be addressed [32]. The present study aimed at describing
the prevalence of use of CAM in Europe, as well as identify-
ing the correlates of use of the two main classes of CAM: diets
and supplements and mind–body practices.
Eur J Pediatr
Methods
Ethical approval was given by the Research Ethics Committee
of the Faculty of Children and Learning, Institute of
Education, London, UK. Parents provided informed consent
before completing the survey (IOE/FPS 385).
Survey
The present study focuses on a set of questions on use of CAM
that was part of a wider-scope survey on use of interventions in
Europe [33]. The survey was open for completion for 45 days.
A total of 1,680 families with a child with ASD aged 7 years or
younger in 18 countries completed the online survey: Belgium,
Czech Republic, Denmark, Finland, France, Germany,
Hungary, Iceland, Ireland, Italy, the Netherlands, Norway,
Poland, Portugal, Romania, Spain, the former Yugoslav
Republic of Macedonia and the United Kingdom. Participants
were recruited via national parents’associations who advertised
the link on their websites, with the exception of parents in the
former Yugoslav Republic of Macedonia who were recruited
through the Paediatric Clinic of Skopje in absence of a national
parents’association and completed a paper version of the sur-
vey. Before launching the survey, the questionnaire was piloted
with parents from the UK (N=8) and Italy (N=2); as a result of
the pilot, the possibility to select a generic conventional inter-
vention if the nature of the approach was not known to the
parent was further highlighted in the initial instructions.
Participant characteristics are summarised in Table 1.
Parent characteristics General background information on
respondents was gathered: relationship to child (mother/fa-
ther/other) and educational level (below high school diploma,
high school diploma, bachelor/degree, postgraduate). The ed-
ucational level was collapsed for analysis into the following
two categories: low educational level (up to high school di-
ploma, 37 %) and high educational level (degree and post-
graduate, 63 %). To comply with the relevant legislation on
cross-national sharing of sensitive personal data in some of the
participating countries, parents were asked to report on the
country of residency, but data on nationality and ethnicity
were not collected.
Child characteristics Information on the age of the child at
survey completion was collected and dichotomised to reflect
the age at which typically children start school in Europe
(below age 5 years, 52 % and 5 years and above, 48 %).
Child verbal ability was rated by parents selecting one of five
options (does not talk; uses single words; uses two- or three-
word phrases; uses sentences with four or more words; uses
complex sentences). The options were collapsed into two cat-
egories for the purposes of analysis: low verbal ability (non-
verbal or single words speech, 37 %) and use of at least phrase
speech (63 %).
Use of conventional therapies and prescription
medication Parents were asked to report on current use of
conventional behavioural, developmental and psychosocial
intervention (such as applied behavioural analysis, occupa-
tional therapy, speech and language therapy) and medication.
A total of 1,529 parents (91 %) reported using at least one
conventional intervention. The number of conventional inter-
ventions used ranged from 1 to 7 (M =2.39, SD=1.43; IQR,
1–3); more detailed results are reported in Salomone et al.
[33]. For the purposes of this analysis, we classified the sam-
ple for level of use of conventional treatments. Three levels of
use were defined based on the distribution of number of inter-
ventions used: no use (zero therapies used; 9 %), medium
level of use (use of one to three therapies; 70.5 %) and high
level of use (use of four or more therapies; 20.5 %). Parents
reported using at least one medication in 19.7 % of cases in the
total sample. Use of medication was dichotomised for this
analysis into a ‘use of any medication’binary variable.
CAM A list of CAM approaches was drawn from the litera-
ture. Parents were asked to endorse all the approaches that
they had used with their child in the previous 6 months. The
CAM approaches, listed alphabetically in the form, were suc-
cessively classified into four categories for the purpose of
statistical analysis: the three categories proposed by the
NCCAM (diets and supplements; mind and body practices;
Tabl e 1 Participants
Europe (N=1,680)
Gender
male n(%) 1,389 (82.7 %)
Age
months M (SD) 58.18 (14.04)
≤5years n(%) 880 (52.4 %)
>5 years n(%) 800 (47.6 %)
Verbal ability
non-verbal/single words n(%) 620 (36.9 %)
at least phrase speech n(%) 1,060 (63.1 %)
Respondent educational level
up to high school diploma n(%) 615 (36.6 %)
graduate and postgraduate n(%) 1,065 (63.4 %)
Use of medication
at least one n(%) 331 (19.7 %)
Use of conventional therapies
number of therapies M (SD) 2.39 (1.43)
no use of therapies n(%) 151 (9 %)
using 1 to 3 therapies n(%) 1,184 (70.5 %)
using 4+ therapies n(%) 345 (20.5 %)
Eur J Pediatr
other unconventional approaches) and a fourth category of
‘invasive, disproven or potentially unsafe CAM’(including
chelation, hyperbaric oxygen therapy and packing) which
was added [following 1].
Data analysis
Descriptive statistics were used to report on use of each CAM
approach, grouped in four over-arching categories. Prevalence
of use of these categories was examined in the total sample
andbyEuropeanregion[37]: Western Europe (Belgium,
France, Germany and the Netherlands), Northern Europe
(Denmark, Finland, Iceland, Ireland, Norway, the United
Kingdom), Eastern Europe (Czech Republic, Hungary,
Poland, Romania) and Southern Europe (Italy, the former
Yugoslav Republic of Macedonia, Portugal, Spain). To com-
pare the effect of European region of residence on the amount
of CAM used, a one-way between subjects analysis of vari-
ance (ANOVA) was conducted with the mean number of
CAM approaches used as dependent variable. Post hoc,
pairwise, Bonferroni-corrected comparisons were conducted.
To investigate the association of child and parental charac-
teristics with use of CAM, we conducted logistic regressions
on the total sample for two primary outcomes: use of any diets
or supplements and use of any mind-body practices. These
categories were selected for the analysis on the basis of the
following criteria: conceptual relevance, homogeneity of ap-
proaches included and frequency of use. In each model, the
predictors were: child’s gender, verbal ability and age, paren-
tal educational level, use of any prescription medication, use
of conventional therapies categorised into three dummy vari-
ables (no use of therapy, medium level of use and high level of
use; the first category was used as the reference).
Results
Frequency of CAM
Frequency of use of individual CAM approaches is reported in
Tab le 2. A total of 789 respondents (47 %) reported using at
least one type of CAM. The rate of use of any CAM was
significantly higher in Eastern (66 %) than in Western (41 %,
p<0.001), Northern (46 %, p<0.001) and Southern (40 %,
p<0.001) Europe. The prevalence of use in Northern Europe
was also significantly higher than in Southern Europe (p=
0.038). In the total sample, the most commonly used CAM
approaches were diets and supplements (24 % reported using
any); use of vitamins was reported by 259 parents (15 %), and
gluten- or casein-free diets were reported by 227 (13 %). The
proportion of parents reporting using diets and supplements
was significantly higher in Eastern Europe (38 %) than in
Western (17 %, p<0.001), Northern (28 %, p=0.003) and
Southern Europe (20 %, p< 0.001). Reported use in Northern
Europe was also significantly higher than use in Western
(p<0.001) and Southern Europe (p=0.007). Mind and body
practices were reported by 395 respondents in the total sample
(23 %); among these, sensory integration therapy (14 %) and
massage (7 %) were the most commonly used treatments.
Parents in Eastern Europe also reported the highest rate of use
of any mind–body practices (34 %); this proportion was signif-
icantly higher than rates in Western (20 %, p<0.001), Northern
(28 %, p=0.043) and Southern Europe (16 %, p<0.001).
Reported use of mind–body practices in Northern Europe was
also significantly higher than in Southern (p<0.001) and
Western Europe (<0.006). A number of other unconventional
approaches not included in the previously mentioned classes of
CAM were reported in 24 % of the total sample (n=411):
Among these, pet therapy (n=233, 14 %) and homeopathy
(n=161, 10 %) were the most widely used. The proportion of
parents reporting using such approaches was significantly
higher in Eastern Europe (43 %) than in Western (24 %,
p<0.001), Northern (12 %, p< 0.001) and Southern Europe
(20 %, p<0.001). Reported use in Western and Southern
Europe was also significantly higher than in Northern Europe
(p<0.001 and p= 0.006). A small minority of parents (n=40,
2.4 %) reported using any invasive, disproven or potentially
unsafe CAM (chelation, hyperbaric chamber and packing).
Rate of use of such approaches was significantly higher in
Eastern Europe (5.1 %) than in Western (0.8 %, p<0.001),
Northern (1.5 %, p=0.006) and Southern Europe (2.5 %, p=
0.037). The rate in Southern Europe was also significantly
higher than the rate in Western Europe (p= 0.034). The total
number of different CAM approaches used for those parents
who used any CAM approaches ranged from 1 to 12 with a
mean of 2.15 (SD= 1.55, IQR, 1–3) in the total sample. A
significant effect of European region was found on number of
CAM approaches used, F(3, 785)= 9.72, p<0.001, ω=0.18.
Post hoc comparisons indicated that the mean number of ap-
proaches used with children living in Eastern Europe (M=2.60,
SD=1.87, IQR, 1–3) was significantly higher than the mean
number of approaches used with children living in Western
Europe (M = 1.92, SD = 1.40, IQR, 1–2; p<0.001), Northern
Europe (M=2.06, SD =1.22, IQR: 1–3; p=0.004) and
Southern Europe (M=1.93, SD =1.42, IQR, 1–2; p<0.001).
Predictors of CAM use
Logistic regression models were performed on the total sam-
ple with use of any diets/supplements and use of any mind–
body practice as outcome variables. Table 3reports the odds
ratios and 95 % CIs for the predictors of each logistic regres-
sion model. For all models, the χ
2
statistics were significant
(all p<0.001) and the Hosmer and Lemeshow’sgoodness-of-
fit tests [17] were not significant (hence indicating well-fitting
models). The Nagelskerke’sR
2
values [27] were low (range,
Eur J Pediatr
0.03–0.08), which is an indication that several other relevant
variables had not been included in the model. For each
predictor, the effects reported below are intended to be
over and above the effect of all other variables included
in the model.
Use of any diets/supplements
Higher parental educational level, low verbal ability in the
children, and use of prescription medication increased the
likelihood of using diets or supplements. Child’sgenderand
age and use of conventional therapy were not significantly
associated with use of diets or supplements.
Use of any mind–body practices
Mind and body practices were less likely to being used with
boys than with girls. A higher parental educational level and
high levels of use of conventional psychosocial interventions
were associated with concurrent use of mind-and-body
Tabl e 2 UseofCAMinEurope
CAM and unconventional
approaches
Europe
(N=1,680)
Western Europe
(n=473)
Northern Europe
(n=341)
Eastern Europe
(n=354)
Southern Europe
(n=512)
N(%) n(%) n(%) n(%) n(%)
Diets and supplements
vitamins 259 (15.4 %) 33 (7 %) 61 (17.9 %) 101 (28.5 %) 64 (12.5 %)
gluten-/casein-free diet 227 (13.5 %) 43 (9.1 %) 40 (11.7 %) 71 (20.1 %) 73 (14.3 %)
yeast-free diet 30 (1.8 %) 7 (1.5 %) 7 (2.1 %) 9 (2.5 %) 7 (1.4 %)
other diet 77 (4.6 %) 25 (5.3 %) 28 (8.2 %) 18 (5.1 %) 6 (1.2 %)
Any diet or supplements 410 (24.4 %) 80 (16.9 %) c 94 (27.6 %) b 133 (37.6 %) a 103 (20.1 %) c
Mind and body practices
acupressure 6 (0.4 %) 1 (0.2 %) 2 (0.6 %) 3 (0.8 %) 0 (0 %)
acupunture 6 (0.4 %) 1 (0.2 %) 0 (0 %) 1 (0.3 %) 4 (0.8 %)
auditory integration training 40 (2.4 %) 4 (0.8 %) 0 (0 %) 19 (5.4 %) 10 (2 %)
biofeedback 15 (0.9 %) 0 (0 %) 0 (0 %) 10 (2.8 %) 4 (0.8 %)
craniosacral therapy 40 (2.4 %) 8 (1.7 %) 13 (3.8 %) 8 (2.3 %) 11 (2.1 %)
deep pressure therapy 48 (2.9 %) 9 (1.9 %) 31 (9.1 %) 4 (1.1 %) 4 (0.8 %)
massage 119 (7.1 %) 33 (7 %) 33 (9.7 %) 40 (11.3 %) 13 (2.5 %)
osteopathy 45 (2.7 %) 34 (7.2 %) 2 (0.6 %) 2 (0.6 %) 7 (1.4 %)
sensory integration therapy 228 (13.6 %) 35 (7.4 %) 46 (13.5 %) 93 (26.3 %) 54 (10.5 %)
Any mind and body practice 395 (23.5 %) 95 (20.1 %) c 95 (27.9 %) b 121 (34.2 %) a 84 (16.4 %) c
Invasive, disproven or potentially unsafe CAM
chelation 25 (1.5 %) 3 (0.6 %) 2 (0.6 %) 13 (3.7 %) 7 (1.4 %)
hyperbaric oxygen therapy 13 (0.8 %) 1 (0.2 %) 0 (0 %) 4 (1.1 %) 8 (1.6 %)
packing 4 (0.2 %) 0 (0 %) 3 (0.9 %) 1 (0.3 %) 0 (0 %)
Any invasive, disproven or unsafe 40 (2.4 %) 4 (0.8 %) c 5 (1.5 %) b, c 18 (5.1 %) a 13 (2.5 %) b
Other unconventional approaches
aromatherapy 26 (1.5 %) 10 (2.1 %) 6 (1.8 %) 9 (2.5 %) 1 (0.2 %)
doman Delacato patterning 15 (0.9 %) 1 (0.2 %) 1 (0.3 %) 7 (2 %) 6 (1.2 %)
facilitated communication 47 (2.8 %) 21 (4.4 %) 1 (0.3 %) 15 (4.2 %) 10 (2 %)
holding therapy 27 (1.6 %) 4 (0.8 %) 5 (1.5 %) 15 (4.2 %) 3 (0.6 %)
homeopathy 161 (9.6 %) 61 (12.9 %) 8 (2.3 %) 56 (15.8 %) 36 (7 %)
oxytocin 5 (0.3 %) 2 (0.4 %) 1 (0.3 %) (%) 2 (0.4 %)
pet therapy 233 (13.9 %) 40 (8.5 %) 25 (7.3 %) 106 (29.9 %) 62 (12.1 %)
Any other unconventional approach 411 (24.5) 114 (24.1 %) b 41 (12 %) c 154 (43.5 %) a 102 (19.9 %) b
Any CAM/unconventional approaches 789 (47 %) 196 (41.4 %) b, c 157 (46 %) b 233 (65.8 %) a 203 (39.6 %) c
Frequencies in a row without a common lower-case letter are significantly different at the 0.01 level, according to χ
2
tests (pranged from <0.001 to
0.006)
Eur J Pediatr
practices. Child’s age and verbal ability, use of medication and
medium levels of use of conventional treatments were
not significantly associated with use of this category of CAM.
Discussion
This study is the first to report on use of CAM in young
children with autism in Europe. We found that, overall,
47 % of parents reported using at least one type of CAM or
other unconventional treatment in the previous 6 months.
Rates of use were homogeneous across Europe with the ex-
ception of significantly higher rates in Eastern Europe (66 %).
Prevalence data from US samples obtained from reviews of
patients charts vary from 30–50 % [1,21,32]to70–90 % [14,
15]. In the total sample, parents reported using diets or sup-
plements in 24 % of cases. Previous reports of use of diets
ranged 27–42 % [11,13,15], but comparisons are made dif-
ficult by the different level of detail across studies. A similar
proportion of parents in our total sample reported using mind–
body practices (24 %). Rates from previous studies ranged
20–30 % [14,15], but comparison should be made with cau-
tion as different definitions were used or CAM approaches
were considered individually rather than as a class.
We also enquired about some invasive or potentially harm-
ful treatments: chelation, hyperbaric chamber and packing.
Chelation is medical procedure involving administering vari-
ous chemical substances for the purpose of binding and then
withdrawing specific metals from the person’sbody;itspo-
tential serious side effects (including death) and the lack of
sound scientific rationale argue against its therapeutic use [6].
Hyperbaric oxygen therapy involves breathing oxygen in a
pressurised chamber for the purpose of increasing the amount
of oxygen in the blood; it is both ineffective [12] and unsafe
(potential side effects include paralysis and air embolism).
Packing involves wrapping the individual in towels previous-
ly wet in cold water to supposedly reinforce the individuals’
consciousness of their bodily limits; this practice, which
appears to be a clear violation of human rights, has not been
evaluated systematically [7]. In our sample, 40 parents (2 %)
reported having used any of these treatments in the previous 6
months with their children.
Among the other unconventional treatments included in the
survey, it is relevant to note the high prevalence of reported
use of pet therapy (14 %). Pet therapy is a generic term that
encompasses both the use of ‘assistance’pets (i.e. placement
of a pet in the family) and the use of ‘therapy’pets by a
therapist at home or in other settings; it is not possible to know
whether in our sample parents were endorsing the former or
the latter. Higher figures (24 %) have been reported before [5].
There were significant regional differences in the rates of
use of the four over-arching categories of CAM, with consis-
tently higher rates in Eastern Europe than in the rest of Europe.
This might be due to lack of access to evidence-based infor-
mation in those countries, possibly as a residual consequence
of the historical divide on health policies in Europe [24]orto
cultural differences in attitudes of professionals and commu-
nity members that are only beginning to be explored [41].
We were interested in identifying correlates of use of CAM
in Europe. As CAM is comprised of a plethora of different
types of ‘treatments’, we investigated correlates of use of the
two main classes of approaches, selected for their conceptual
relevance and relative homogeneity: diets and supplements
and mind–body practices. No gender differences were found
for use of diets and supplements. There was a tendency for
more mind–body practices to be used with girls than with
boys, but this finding should be interpreted with caution as
the females in the sample were only a minority (n=291,
18 %). In our sample, non-verbal children and children with
single-words speech were more likely to being treated with
diets (30 % increase in the probability of use), suggesting that
parents of lower functioning children may tend to look to a
range of interventions to respond to more severe difficulties.
Additionally, over and above the effect of verbal ability, chil-
dren using prescription medication were also more likely to be
treated with diets than children not using medication (62 %
Tabl e 3 Predictors of CAM use
in Europe Any diets/supplements Any mind–body practice
OR (p) 95 % CI OR (p)95%CI
child’s age (>5 years) 1.03 (.818) 0.81–1.30 1.17 (.207) 0.92–1.49
child’s gender (male) 0.97 (.812) 0.72–1.30 0.68 (.010) 0.51–0.91
child’s verbal ability (non-verbal) 1.30 (.034) 1.02–1.65 1.18 (.201) 0.92–1.51
parental educational level (high) 1.35 (.013) 1.07–1.72 1.64 (<.001) 1.27–2.11
use of prescription medication 1.62 (<.001) 1.24–2.12 1.26 (.110) 0.95–1.67
conventional therapy: use of 1–3 therapies 1.13 (.585) 0.74–1.71 1.60 (.069) 0.96–2.65
conventional therapy: use of 4+ therapies 1.41 (.151) 0.88–2.24 4.36 (<.001) 2.56–7.42
χ
2
(7)= 28.055, p<0.001;
H-L χ
2
ns; R
2
=0.03
χ
2
(7)= 95.548, p< 0.001;
H-L χ
2
ns; R
2
=0.08
Eur J Pediatr
increase). Interestingly though, neither of these associations
was found for use of mind–body practices. This suggests that
previous evidence of higher use of CAM in low-functioning
children [14,32] might be specific to some CAM types. The
association of use of medication with use of diets but not
mind–body practices might be due to the use of supplements
or alterations in the diet as an attempt to counter-balance po-
tential side effects of medications or to ‘boost’their efficacy
[15]. Alternatively, the association could reflect parental atti-
tudes or beliefs (e.g., a generic belief in chemical/biological
mechanisms) or the willingness of the child to orally intake
pills or tablets. Increased diet use in children concurrently
taking medications may also reflect an attempt to counteract
the weight-gain associated with many psychotropic medica-
tions, although we did not ask parents to report why their child
was on a diet, which should be done in future studies. In
addition, we do not have information on whether diets or
supplements were medically prescribed as a treatment for spe-
cific conditions (such as iron deficiency).
Parents with a high educational level have been consistent-
ly reported in previous studies to be more likely to use CAM
than parents with a lower educational level [1,14], and in our
study, more educated parents were more likely to choose diets
or supplements for their child as well as using mind-and-body
practices. Notably, the increase in the likelihood was higher
for the mind–body practices (64 %) than diets and supple-
ments (35 %). Mind-and-body practices are practitioner-deliv-
ered, and their cost is on average almost double the cost of
self-care therapies such as supplements [28], and this might
explain why in our sample mind-and-body practices were sig-
nificantly less used by parents with a lower socio-economic
status (SES; indexed by their educational level).
When the correlation of CAM use and use of conventional
treatments has been explored, it appears that availability and
use of conventional treatments does not lessen use of CAM.
Indeed, CAM use has been found to be associated with receiv-
ing 20 or more hours per week of conventional treatment [1].
Here, we explored the association between use of convention-
al treatments and use of two specific classes of CAM. There
was no association between use of conventional treatment and
use of diets: the use of such approaches might in fact be more
related to the use of medications, as suggested above. We
found instead a large dose–response effect of use of conven-
tional treatments on use of mind–body practices with a four-
fold increase in the likelihood of concurrent CAM use for
parents reporting already using more than four conventional
treatments for their child, but not for medium levels of treat-
ment (up to three interventions).
This finding suggests that use of mind–body practices is
most strongly related with a tendency to try a wide number of
approaches; this might indicate that some parents, over and
above the effect of their child’s level of functioning (measured
as verbal ability) and of their own educational level (which
can be constructed as a proxy of their SES), tend to look for as
many therapies as possible, whether these be conventional
treatments or CAM approaches. Use of diets appeared to be
most strongly associated with lower functioning of the child
and concurrent use of prescription medication.
There is concern that desperate parents may resort to unsafe
or disproven CAM approaches and public agencies have been
actively campaigning against them [10], but such approaches
were not in wide use in our sample. However, animal-assisted
therapy, whose efficacy is not yet established, is attracting
increasing interest [30] and a considerable number of parents
reported using such approaches in our sample. These findings
have implications for clinicians and professionals involved in
the care of children with ASD, in that they should engage
parents in frank discussions about CAM approaches, the avail-
able evidence and any potential for adverse effects.
Strengths and limitations
There are a number of strengths to the present study, including
the large sample size and the wide scope of the survey, which
enquired about the use of a range of both CAM approaches
and conventional treatments for young children with autism in
Europe. Moreover, while previous research has looked at pre-
dictive factors for use of CAM considering child and parent
characteristics as individual factors or only adjusting for pa-
rental education level, in our study, we used multiple logistic
regression to estimate the contribution of each predictor hav-
ing taken into account the influence of the other factors. These
findings can help to identify families potentially more likely to
adopt CAM approaches, and this information may be benefi-
cial both to primary care providers in their role as clinical
advisors, and to researchers, for example, when designing
trials of CAM approaches.
Nevertheless, the findings should be seen in the context of
some limitations. Firstly, we employed a recruitment method
(online survey advertised via parents’associations) that might
have been prone to selection bias since parents involved in
associations are more likely to have a relatively high income
and educational level [26] and Internet access is still a function
of socio-demographic characteristics in Europe [39]. Our sam-
ple had in fact a higher than average education level [9].
However, while the recruitment strategy used necessarily pre-
vents any claims of generalisability of our results to the
European population, it has enabled us to reach a large number
of families across Europe. Moreover, it has been argued that,
given the controversy around use of CAM, an anonymous
online survey might actually better protect against the poten-
tial risks of selection and reporting bias [36]thanwhenparents
are directly asked by clinicians (as done in most other studies
on the topic). Furthermore, we found that reported levels of
CAM use in the present study were similar to previous studies
in non-EU samples.
Eur J Pediatr
The factors examined in the present study are only some of
the many that might affect the decision-making process un-
derlying the choice of using CAM alongside (or alternative to)
conventional healthcare, which is still largely unexplained.
For example, there is preliminary evidence from a small sam-
ple of French parents (N= 89) that personality characteristics
such as personal control and attribution of cause of autism
affect the decision to use CAM or not [2]. Parents’own use
of CAM is likely to be a relevant factor but has never been
examined in the association with use of CAM in children with
autism. Further research on the topic should include these and
other factors, such as beliefs on ASD aetiology, to better un-
derstand the phenomenon of use of CAM. Finally, reliance on
parent report in absence of direct assessments places a limita-
tion on these findings in relation to severity of child symptoms
and behavioural characteristics.
Conclusions
This was the first study to report on factors associated with use
of CAM in a large sample of young children with autism in
Europe. While little is known on the efficacy (and conversely,
on the potential harm) of CAM approaches, a vast amount of
uncontrolled information is available online, putting parents at
risk of embarking in sometimes costly and often non-
efficacious treatments. Rates of CAM use, including use of
disproven or unsafe approaches, were particularly high in
Eastern Europe. The present study contributed to the under-
standing of the factors associated to use of CAM and provided
some evidence that families that tend to use a wide range of
conventional treatments might also be more likely to be trying
some CAM approaches. The reasons behind this are not fully
understood and may reflect factors that were not captured by
the present study. Nonetheless, these findings, taken together
with the evidence of socio-economic barriers in access to treat-
ment for autism [22,33], provide some insight into the lengths
to which families may go in pursuit of ways to help their child
progress. The findings highlight the need to further advance
research funding and policy development for evidence-based
early interventions for children with ASD across Europe.
Acknowledgments We are grateful to all the parents who participated
in the study and to the parents associations that were involved in
recruiting the participants. This research was supported by the COST
Action BM1004 funded by the European Science Foundation. TC also
received support from the Innovative Medicines Initiative Joint Under-
taking under grant agreement n° 115,300, resources of which are com-
posed of financial contribution from the European Union’s Seventh
Framework Programme (FP7/2007 –2013) and EFPIA companies’in
kind contribution. It was made possible by the clinicians and researchers
who are members of the COST ESSEA (http://www.cost-essea.com/)and
EU-AIMS (http://www.eu-aims.eu/) networks. The COST ESSEA work
group 4 also includes: Anett Kaale (Norway), Bernadette Rogé and
Frederique Bonnet-Brilhaut (France), Iris Oosterling (The Netherlands),
Selda Ozdemir (Turkey), Antonio Narzisi and Filippo Muratori (Italy),
Joaquin Fuentes (Spain), Mikael Heimann (Sweden), Michele
Noterdaeme, Christine Freitag, Luise Poustka and Judith Sinzig
(Germany) and Jonathan Green (UK).
Conflict of Interest All authors report no biomedical financial interests
or potential conflict of interest.
Authors’contribution ES, TC, HMC and PW made substantial con-
tributions to conception and design of the study and to theanalysis and
interpretation of data. ES, TC, HCM, PW and all members of the Working
Group 4, COST Action‘Enhancing the Scientific Study of Early Autism’
made substanstial contributions to the acquisition of data. ES drafted the
article and all authors revised it critically and gave final approval of the
submitted version.
References
1. Akins RS, Krakowiak P, Angkustsiri K, Hertz-Picciotto I, Hansen
RL (2014) Utilization patterns of conventional and complementary/
alternative treatments in children with autism spectrum disorders
and developmental disabilities in a population-based study. J Dev
Behav Pediatr 35(1):1–10. doi:10.1097/DBP.0000000000000013
2. Al Anbar NN, Dardennes RM, Prado-Netto A, Kaye K, Contejean
Y (2010) Treatment choices in autism spectrum disorder: the role of
parental illness perceptions. Res Dev Disabil 31(3):817–828. doi:
10.1016/j.ridd.2010.02.007
3. APA (2013) Diagnostic and statistical manual of mental disorders,
5th edn. American Psychiatric Publishing, Arlington
4. Bernier R, Mao A, Yen J (2010) Psychopathology, families, and
culture: autism. Child Adolesc Psychiatr Clin N Am 19(4):855–
867. doi:10.1016/j.chc.2010.07.005
5. Christon LM, Mackintosh VH, Myers BJ (2010) Use of comple-
mentary and alternative medicine (CAM) treatments by parents of
children with autism spectrum disorders. Res AutismSpectr Disord
4(2):249–259. doi:10.1016/j.rasd.2009.09.013
6. Davis TN, O’Reilly M, Kang S, Lang R, Rispoli M, Sigafoos J,
Lancioni G, Copeland D, Attai S, Mulloy A (2013) Chelation treat-
ment for autism spectrum disorders: a systematic review. Res
Autism Spectr Disord 7(1):49–55. doi:10.1016/j.rasd.2012.06.005
7. DeJong H, Bunton P, Hare D (2014) A systematic review of inter-
ventions used to treat catatonic symptoms in people with autistic
spectrum disorders. J Autism Dev Disord 44(9):2127–2136. doi:10.
1007/s10803-014-2085-y
8. Elder J, Shankar M, Shuster J, Theriaque D, Burns S, Sherrill L
(2006) The gluten-free, casein-free diet in autism: results of a pre-
liminary double blind clinical trial. J Autism Dev Disord 36(3):
413–420. doi:10.1007/s10803-006-0079-0
9. Eurostat tSOotEU (2014) Tertiary educational attainment by sex,
age group 30–34 (t2020_41). doi:http://epp.eurostat.ec.europa.eu/
tgm/table.do?tab=table&init=1&plugin=1&language=en&pcode=
t2020_41
10. FDA (2014) U.S. Food and Drug Administration website: Beware
of false or misleading claims for treating autism http://www.fda.
gov/ForConsumers/ConsumerUpdates/ucm394757.htm.Accessed
16/10/2014
11. Goin-Kochel RP, Myers BJ, Mackintosh VH (2007) Parental re-
ports on the use of treatments and therapies for children with autism
spectrum disorders. Res Autism Spectr Disord 1(3):195–209. doi:
10.1016/j.rasd.2006.08.006
12. Granpeesheh D, Tarbox J, Dixon DR, Wilke AE, Allen MS,
Bradstreet JJ (2010) Randomized trial of hyperbaric oxygen
Eur J Pediatr
therapy for children with autism. Res Autism Spectr Disord 4(2):
268–275. doi:10.1016/j.rasd.2009.09.014
13. Green VA, Pituch KA, Itchon J, Choi A, O’Reilly M, Sigafoos J
(2006) Internet survey of treatments used by parents of children
with autism. Res Dev Disabil 27(1):70–84. doi:10.1016/j.ridd.
2004.12.002
14. Hall SE, Riccio CA (2012) Complementary and alternative treat-
ment use for autism spectrum disorders. Complement Ther Clin
Pract 18(3):159–163. doi:10.1016/j.ctcp.2012.03.004
15. Hanson E, KalishL, Bunce E, Curtis C, McDaniel S, Ware J, Petry J
(2007) Use of complementary and alternative medicine among chil-
dren diagnosed with autism spectrum disorder. J Autism Dev
Disord 37(4):628–636. doi:10.1007/s10803-006-0192-0
16. Holm M, Baird J, Kim Y, Rajora K, D’Silva D, Podolinsky L,
Mazefsky C, Minshew N (2014) Therapeutic horseback riding out-
comes of parent-identified goals for children with autism spectrum
disorder: an ABA’multiple case design examining dosing and gen-
eralization to the home and community. J Autism Dev Disord 44(4):
937–947. doi:10.1007/s10803-013-1949-x
17. Hosmer DW, Lemeshow S (2000) Applied logistic regression.
Wiley, New York
18. James S, Montgomery P, Williams K (2011) Omega-3 fatty acids
supplementation for autism spectrum disorders (ASD). Cochrane
Database Syst Rev 11
19. Knivsberg AM, Reichelt KL, HØien T, NØdland M (2002) A
randomised, controlled study of dietary intervention in autistic syn-
dromes. Nutr Neurosci 5(4):251–261. doi:10.1080/
10284150290028945
20. Levy SE, Hyman SL (2008) Complementary and alternative med-
icine treatments for children with autism spectrum disorders. Child
Adolesc Psychiatr Clin N Am 17(4):803–820. doi:10.1016/j.chc.
2008.06.004
21. Levy SE, Mandell DS, Merhar S, Ittenbach RF, Pinto-Martin JA
(2003) Use of complementary and alternative medicine among chil-
dren recently diagnosed with autistic spectrum disorder. J Dev
Behav Pediatr 24(6):418–423
22. Liptak GS, Benzoni LB, Mruzek DW, Nolan KW, Thingvoll MA,
Wade CM, Fryer GE (2008) Disparities in diagnosis and access to
health services for children with autism: data from the national
survey of children’s health. J Dev Behav Pediatr 29(3):152–160
23. Lofthouse N, Hendren R, Hurt E, Arnold LE, Butter E (2012) A
review of complementary and alternative treatments for autism
spectrum disorders. Autism Res Treat 2012:21. doi:10.1155/2012/
870391
24. Mackenbach JP, Karanikolos M, McKee M (2013) The unequal
health of Europeans: successes and failures of policies. Lancet
381(9872):1125–1134. doi:10.1016/S0140-6736(12)62082-0
25. Mandell DS, Novak M (2005) The role of culture in families’treat-
ment decisions for children with autism spectrum disorders. Ment
Retard Dev Disabil Res Rev 11(2):110–115. doi:10.1002/mrdd.
20061
26. Mandell DS, Salzer MS (2007) Who joins support groups among
parents of children with autism? Autism 11(2):111–122. doi:10.
1177/1362361307077506
27. Nagelkerke NJD (1991) A note on a general definition of the coef-
ficient of determination. Biometrika 78:691–69228
28. Nahin R, Barnes P, Stussman B, Bloom B (2009) Costs of comple-
mentary and alternative medicine (CAM) and frequency of visits to
CAM practitioners: United States, 2007, vol 18, National Health
Statistics Reports. National Center for Health Statistics, Hyattsville
29. NICE (2013) Autism. The management and support of children and
young people on the autism spectrum [CG170]. National Institute
for Health and Care Excellence, London
30. O’Haire M (2013) Animal-assisted intervention for autism spec-
trum disorder: a systematic literature review. J Autism Dev Disord
43(7):1606–1622. doi:10.1007/s10803-012-1707-5
31. O’Reilly M, Cook L, Karim K (2012) Complementary or controver-
sial care? The opinions of professionals on complementary and alter-
native interventions for autistic spectrum disorder. Clin Child Psychol
Psychiatry 17(4):602–615. doi:10.1177/1359104511435340
32. Perrin JM, Coury DL, Hyman SL, Cole L, Reynolds AM, Clemons
T (2012) Complementary and alternative medicine use in a large
pediatric autism sample. Pediatrics 130(Supplement 2):S77–S82.
doi:10.1542/peds.2012-0900E
33. Salomone E, Beranová Š, Bonnet-Brilhault F, Briciet Lauritsen M,
Budisteanu M, Buitelaar JK, Canal-Bedia R, Felhosi G, Fletcher-
Watson S, Fuentes J, Gallagher L, Garcia Primo P, Gliga F, Gomot
M, Green J, Heimann M, Loa Jónsdóttir S, Kaale A, Kawa R,
Kylliainen A, Lemcke S, Markovska-Simoska S, Marschik P,
McConachie H, Moilanen I, Muratori F, Narzisi A, Noterdaeme
M, Oliveira G, Oosterling I, Pijl M, Pop-Jordanova N, Poustka F,
Roeyers H, Rogé B, Sinzig J, Vicente A, Warreyn P, Charman T
(2015) Use of early intervention for young children with autism
spectrum disorder across Europe. Autism: Int J Res Pract, in press
34. Silva LM, Schalock M, Gabrielsen K (2011) Early intervention for
autism with a parent-delivered Qigong massage program: a ran-
domized controlled trial. Am J Occup Ther 65(5):550–559
35. Sinha Y, Silove N, Hayen A, Williams K (2011) Auditory integra-
tion training and other sound therapies for autism spectrum disor-
ders (ASD). Cochrane Database Syst Rev 12
36. Surette S, Vanderjagt L, Vohra S (2013) Surveys of complementary
and alternative medicine usage: a scoping study of the paediatric
literature. Complement Ther Med 21(Supplement 1 (0)):S48–S53.
doi:10.1016/j.ctim.2011.08.006
37. UN (2006) International Standard ISO 3166–1, Codes for the rep-
resentation of names of countries and their subdivisions—part 1:
Country codes, ISO 3166–1: 2006 (E/F). Geneva
38. Valicenti-McDermott M, Burrows B, Bernstein L, Hottinger K,
Lawson K, Seijo R, Schechtman M, Shulman L, Shinnar S (2014)
Use of complementary and alternative medicine in children with
autism and other developmental disabilities: associations with eth-
nicity, child comorbid symptoms, and parental stress. J Child
Neurol 29(3):360–367. doi:10.1177/0883073812474489
39. Vicente MR, López AJ (2011) Assessing the regional digital divide
across the European Union-27. Telecommun Policy 35(3):220–
237. doi:10.1016/j.telpol.2010.12.013
40. Winburn E, Charlton J, McConachie H, McColl E, Parr J, O’Hare
A, Baird G, Gringras P, Wilson D, Adamson A, Adams S, Le
Couteur A (2014) Parents’and child health professionals’attitudes
towards dietary interventions for children with autism spectrum
disorders. J Autism Dev Disord 44(4):747–757. doi:10.1007/
s10803-013-1922-8
41. Zuzak TJ, Boňková J, Careddu D, Garami M, Hadjipanayis A,
Jazbec J, Merrick J, Miller J, Ozturk C, Persson IAL, Petrova G,
Saz Peiró P, Schraub S, Simões-Wüst AP, Steinsbekk A, Stockert
K, Stoimenova A, Styczynski J, Tzenova-Savova A, Ventegodt S,
Vlieger AM, Längler A (2013) Use of complementary and alterna-
tive medicine by children in Europe: published data and expert
perspectives. Complement Ther Med 21(Supplement 1 (0)):S34–
S47. doi:10.1016/j.ctim.2012.01.001
Eur J Pediatr
