Received: 29 August 2017 | Revised: 19 December 2017 | Accepted: 19 December 2017

DOI: 10.1002/eat.22823

BRIEF REPORT

An examination of the clinical outcomes of adolescents and

young adults with broad autism spectrum traits and autism
spectrum disorder and anorexia nervosa: A multi centre study

Bruno Palazzo Nazar MD, MSc, PhD1 | Vanessa Peynenburg BA2 |

Charlotte Rhind BSc, MSc, PhD3 | Rebecca Hibbs PhD3 | Ulrike Schmidt MD, PhD3 |
Simon Gowers MD, PhD4 | Pamela Macdonald PhD3 | Elizabeth Goddard PhD3 |
Gillian Todd RMN, MSc5 | Nadia Micali MD, PhD6,7 | Janet Treasure OBE, PhD, FRCP, FAED3
1
Institute of Psychiatry—Federal University of Rio de Janeiro (IPUB-UFRJ), Rio de Janeiro, Brazil
2
Department of Psychology, University of Alberta, Edmonton, Alberta, Canada
3
Department of Psychological Medicine, Section of Eating Disorders, King’s College London, Institute of Psychiatry, Psychology & Neuroscience, London, United
Kingdom
4
University of Liverpool, Adolescent Psychiatry, Chester, United Kingdom
5
South London and Maudsley NHS Foundation Trust, London, United Kingdom
6
Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York
7
University College London, Institute of Child Health, Behavioural and Brain Sciences Unit, London, United Kingdom

Correspondence
Bruno Palazzo Nazar, Institute of Psychiatry
Abstract
—Federal University of Rio de Janeiro
Objectives: To compare the clinical outcomes of adolescents and young adults with anorexia
(IPUB-UFRJ), Rua Visconde de Piraja 547,
sala 610, Rio de Janeiro 22415900, Brasil. nervosa (AN) comorbid with broad autism spectrum disorder (ASD) or ASD traits.
Email: [email protected]
Method: The developmental and well-being assessment and social aptitude scale were used to
Funding information categorize adolescents and young adults with AN (N 5 149) into those with ASD traits (N 5 23),
National Institute for Health Research and those who also fulfilled diagnostic criteria for a possible/probable ASD (N 5 6). We compared
(NIHR), Grant Number: PB–PG-0609- both eating disorders specific measures and broader outcome measures at intake and 12 months
19025; National Institute for Health
follow-up.
Research (NIHR) Mental Health Biomedical
Research Centre at South London; Mauds- Results: Those with ASD traits had significantly more inpatient/day-patient service use (p 5 .015),
ley NHS Foundation Trust and King’s Col-
as well as medication use (p < .001) at baseline. Both groups had high social difficulties and poorer
lege London; NIHR Senior Investigator
Award; Psychiatry Research Trust global functioning (strengths and difficulties questionnaire) at baseline, which improved over time
but remained higher at 12 months in the ASD traits group (p 5 .002). However, the improvement
in eating disorder symptoms at 12 months was similar between groups with or without ASD traits.
Treatment completion rates between AN only and ASD traits were similar (80.1 vs. 86.5%).

Discussion: Adolescents with AN and ASD traits show similar reductions in their eating disorder
symptoms. Nevertheless, their social difficulties remain high suggesting that these are life-long
difficulties rather than starvation effects.

KEYWORDS
anorexia nervosa, autism, clinical outcome, comorbidity, development, eating disorders, treatment,
weight trajectory

174 | V
C 2018 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/eat Int J Eat Disord. 2018;51:174–179.
 1098108x, 2018, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/eat.22823 by <Shibboleth>[email protected], Wiley Online Library on [10/11/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NAZAR ET AL. | 175

difficulties, hyperactivity, emotional problems, and conduct prob-

1 | INTRODUCTION
lems), completed both by parent and patients at baseline and 12

Neurodevelopmental disorders have been associated with eating disor- months.

ders (ED), with attention-deficit/hyperactivity disorder linked to bulimia Social aptitude scale (SAS)(Liddle, Batty, & Goodman, 2009). Parents

nervosa (BN) and binge-eating disorder (Nazar et al., 2016), and autism report retrospectively on their child’s social development comparing

spectrum disorders (ASDs) to anorexia nervosa (AN) (Baron-Cohen their child’s abilities to those of their peers, completed at baseline only.

et al., 2013; Tchanturia et al., 2013). There has been interest in the The short evaluation of eating disorders (SEED) (Bauer, Winn,

impact these might have upon treatment and prognosis (Westwood & Schmidt, & Kordy, 2005). A six-item measure of ED symptom severity.

Tchanturia, 2017). The outcomes include a three items severity score for AN (ANTSI) and

In a Swedish cohort of AN adolescents, 32% had ASD traits at three items for BN (BNTSI). The score from each item can range from

some time over the course of 18 years follow-up and this sub-group 0 to 3 (0 5 symptom not present; 3 5 symptom is extreme).

had a poorer long term prognosis (Anckarsäter et al., 2012), with partic- The depression, stress and anxiety scale (DASS-21) (Henry & Craw-

ular problems in mental and social functioning (Nielsen et al., 2015). In ford, 2005). A shortened version of the DASS with good internal con-

a series of AN cases referred to a specialized child and adolescent serv- sistency on each of the sub-scales.

ice, 6.9% had broad ASD traits (using the autism-spectrum quotient)
and this sub-group were more likely to progress to more intensive lev- 4 | STATISTICAL ANALYSES
els of care (day or inpatient) (Stewart, McEwen, Konstantellou, Eisler, &
Simic, 2017). In line with these findings, we have previously reported Statistical analyses were completed using IBM SPSS Statistics Version
that 19% of AN adolescents presenting for treatment had ASD traits 22. The sample was divided into two sub-groups (mutually exclusive)
with 4% receiving a probable/possible ASD diagnosis (Rhind et al., based on scores on the SAS using standard cut off criteria (Liddle,
2014). These findings suggest that people with AN and associated ASD Batty, & Goodman, 2009): (a) AN only (SAS above 16) (n 5 126); (b)
traits may have a less favorable prognosis. broad ASD traits (n 5 23). Among the latter group, six individuals
We therefore aimed to examine the 12 months outcomes of formed the ASD diagnosis subset which was also compared to AN
patients who had ASD traits and those with a probable/possible ASD only.
diagnosis in a sample of individuals with AN participating in a random- As the group with broad ASD traits is small, we report Cohen’s d
ized controlled trial. effect sizes to indicate the magnitude of differences between groups.
Cohen’s effect sizes are understood as negligible (<.15), small

2 | METHOD (.15  d < .40), medium (.40  d < .75), large (.75  d).
Mann–Whitney U tests compared differences between sub-groups
The present report is a secondary data analysis from the Experienced where the data was non-parametrically distributed. Chi-square tests
Carers Helping Others (ECHO) trial. This trial aimed to assess the bene- were performed for comparing proportions and Spearman correlations
fits of a carer intervention added to usual treatment, where they for non-parametric associations.
received materials and telephone coaching to better cope with their
offspring’s ED. 5 | RESULTS
Participants were aged between 13 and 21 and consisted of 149
adolescents with AN or Atypical AN, among which, 15.4% had ASD The baseline sociodemographic characteristics are presented in Table 1
traits and 4% received a possible/probable ASD diagnosis. Some base- and the clinical characteristics are presented in Table 2. Regarding
line features have been described previously (Rhind et al., 2014) as well ECHO treatment allocation, there were no significant differences
as details regarding the original trial (Hibbs et al., 2015). between AN only and broad ASD traits groups (X2 5 .077; p 5 .782).
Participants with broad ASD traits had had more general inpatient and
3 | ASSESSMENT MEASURES day-patient days (U 5 1,077.0, z 5 2.433, p 5 .015) and had been more
frequently admitted to an ED specialist inpatient treatment (18.2% vs.
Details regarding measures can be found in the Supporting Information 5%; (X2 5 6.62; p 5 .02). Furthermore, the ASD group presented with
and are summarized below: more antipsychotic use (X2 5 11.74; p < .001) prior to presentation.
The developmental and well-being assessment (DAWBA) (Goodman, Baseline SAS scores were significantly and inversely correlated with
Ford, Richards, Gatward, & Meltzer, 2000). Both parent and patients baseline antipsychotic use in the total sample (rho 5 2.21; p 5 .017). At
completed the ED and ASD modules for this computerized diagnostic baseline, ED symptoms were similar between groups but the group
instrument and trained psychiatrists confirmed the psychiatric diagnosis with broad ASD traits had higher parental (p < .0001) and self-reported
afterwards (DSM-IV [American Psychiatric Association, 2000] and ICD- (U 5 795.5, z 5 22.16, p 5 .03) total general difficulties on the SDQ.
10 [World Health Organization, 1993]). There was no significant difference (X2 5 .588; p 5 .44) between
Strengths and difficulties questionnaire (SDQ) (Goodman & Scott, the proportion of participants who completed the 12 month follow-up
1999). An instrument with five sub-scales (peer problems, prosocial in the AN only (80.1%) and the broad ASD traits group (86.95%). Both
 1098108x, 2018, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/eat.22823 by <Shibboleth>[email protected], Wiley Online Library on [10/11/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
176 | NAZAR ET AL.

T AB LE 1 Sociodemographic characteristics

AN only versus
AN 1 broad AN 1 broad
AN only (SAS > 16) ASD (SAS< 16) ASD differences ASD subset Total sample
N 5 126 N 5 23 (Cohen’s d; p value)* N56 N 5 149

Gender (male:female; %female) 11:115 1:22 16; .47 0:6 12:137

(91.26%) (95.65%) (100%) (91.94%)

Age at baseline (years) mean (SD) 16.9 (62.2) 17.0 (62.0) .04; .84 16.97 (61.76) 16.9 (62.1)

Education in years 11.75 (62.05) 11.7 (1.67) .02; .95 11.83 (61.72) 11.7 (62)

Age of onset in years (median, range) 14.0 (11.0–20.0) 14.3 (7.0–19.0) .12; .39 15 (7–16) 14.0 (7.0–20.0)

Age of diagnosis in years (median, range) 15.0 (12.0–21.0) 15.5 (13.0–20.0) .11; .95 15.0 (13.0–20.0) 15.0 (12.0–21.0)

Lowest BMI 15.4 (62.11) 15.7 (62.6) 2.13; .67 14.68 (61.03) 15.5 (62.21)

Illness duration in months 21.92 (622.19) 24.69 (623.7) 2.12; .58 17.83 (615.8) 22.34 (622.37)

Employment p value
Full- or part-time 10 (7.9%) 2 (8.6%) .85 0% 12 (8.1%)
Students 105 (83.3%) 20 (87%) .36 6 (100%) 125 (83.9%)
Other 11 (8.7%) 1 (4.3%) .47 0% 12 (8.1%)

Living arrangements at intake p value

With parents 116 (92.1%) 21 (95.5%) .64 6 (100%) 137 (91.9%)
Alone 0 1 (4.5%) .01** 0% 1 (0.6%)
Friends/flatmates 4 (3.2%) 0 .39 0% 5 (3.4%)
Uni residence 4 (4.0%) 0 .39 0% 5 (3.4%)
Other 1 (0.8%) 0 .67 0% 1 (0.6%)

Note. *p values of Mann–Whitney tests for medians or v2s for frequencies; **significant at the <.05 level; ***significant at the <.001 level.

groups had similar increases in BMI and weight for height from baseline treatment; one had persistent AN and low weight (BMI 5 15.7 kg/m2);
to 12 months, as well as similar reductions in ED symptoms. After 12 another transitioned to BN and eventually reached a BMI of 25 kg/m2.
months, there were no differences between the groups in medication One patient experienced continuing social and mood problems and
use (X2 5 2.44; p 5 .111) (Table 1). another severe OCD. The remaining two were not in treatment and
Both groups had decreases in general difficulties (SDQ) over time. were presumed to be well.
However, those in the ASD group continued to have a higher level of
total general difficulties (SDQ) at 12 month (U 5 49.0, z 5 23.10,
p 5 .002). Parents, in general, reported higher levels of problems than
6 | DISCUSSION
the patients themselves but there was convergence between parent
The aim of this study was to examine the clinical course of young peo-
(U 5 485.0, z 5 23.23, p 5 .0001) and patient reports (U 5 280.0,
ple with AN who also had ASD traits (broad and narrow/diagnostic).
z 5 23.68, p < .0001). Significantly higher general difficulties (SDQ)
The ASD traits group had a greater use of intensive treatment (in/day-
among those in the ASD traits group were found for both parent
patient) and medication use before presentation to the specialist clinic
(U 5 497.0, z 5 4.24, p < .0001) and self-report (U 5 795.5, z 5 22.16,
and more general difficulties (SDQ), both at baseline and after receiving
p 5 .03), and for the subscale peer difficulties (SDQ), both on parent
specialist treatment. However, eating disorder symptoms (BMI, weight
(U 5 422.0, z 5 24.56, p < .0001) and self-report (U 5 629.5,
for height and SEED) in the broad and narrow/diagnostic groups did
z 5 23.25, p < .0001). SDQ results are presented in Table 3 for AN
not differ from the comparison group both at baseline and at 12
only versus Broad ASD traits group, and SDQ data from the ASD sub-
months. The majority of the narrow/diagnostic group continued to
set is presented in a different version of this table on the Supporting
have contact with mental health services 2 years after presentation,
Information.
two of whom had a persistent eating disorder. A poorer prognosis in
the ASD group appears to relate to the more severe general comorbid-
5.1 | The “ASD probable/possible diagnosis” subgroup
ity than to the eating disorder outcomes alone.
The social difficulties of the six participants (five possible, one definite) The 15.4% (23/149) prevalence of SAS scores below cut-off (ASD
with an ASD diagnosis improved over time (d 5 2.02; p 5 .03) but, as traits group) in this sample is the same as that found in a clinical cohort
expected, remained higher than in the other group of patients. Other of AN and EDNOS cases (Stewart et al., 2017). However, the computer
domains of difficulties also improved (Table 2). generated probability of ASD diagnosis from the DAWBA in the clinical
We obtained an update on the clinical functioning of these six pos- cohort (Stewart et al., 2017) did not differ from community norms,
sible cases at 2-years post intake. Four of them had continued in while in our sample, a trained psychiatrist reviewed the DAWBA and
 1098108x, 2018, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/eat.22823 by <Shibboleth>[email protected], Wiley Online Library on [10/11/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NAZAR ET AL. | 177

T AB LE 2 Comparisons of clinical characteristics from baseline to 12 month follow up

Effect size -Cohen’s d Effect size between groups

Groups Baseline 12 Month follow up (p value) (Cohen’s d)

BMI AN only 17.0 (62.2) 18.7 (62.4) 2.73 (p < .001)

N 5 126 N 5 101
.08
AN 1 Broad ASD 17.3 (62.6) 19.2 (63.1) 2.94 (p 5 .001)
N 5 23 N 5 20
ASD subset 16.74 (62.53) 18.58 (62.3) 21.01 (p 5 .04) N.A.
N56 N56

Weight for height AN only 82.4 (611.0) 88.6 (611.8) 2.52 (p < .001)
N 5 114 N 5 98
.07
AN 1 broad ASD 82.9 (613.3) 90.9 (615.8) 2.82 (p 5 .003)
N 5 22 N 5 20
ASD subset 77.09 (610.33) 87.61 (611.58) 21.05 (p 5 .13) N.A.
N56 N56

ANTSI AN only 1.9 (6.6) 1.4 (6.7) .76 (p < .001)

N 5 121 N 5 83
0
AN 1 broad ASD 1.8 (6.6) 1.3 (6.7) .68 (p 5 .005)
N523 N 5 16
ASD subset 1.91 (6 .75) 1.33 (6 .76) .75 (p 5 .11) N.A.
N56 N56

BNTSI AN only .8 (6.6) .5 (6 .6) .48 (p < .001)

N5113 N5107
.52
AN 1 broad ASD .6 (6.4) .6 (6.6) 0 (p 5 .95)
N 5 22 N 5 21
ASD subset .37 (6.34) .41 (6.46) 2.09 (p 5 .67) N.A.
N56 N56

DASS-21 AN only 65.7 (631.1) 49 (631.2) .54 (p < .001)

N 5 125 N 5 91
.41
AN 1 broad ASD 63.1 (629.4) 59.3 (634.9) .01
N 5 23 N 5 19 (p 5 .77)
ASD subset 61.3 (634.14) 55.33 (628.61) .18 (p 5 .50) N.A.
N56 N56

Using psychotropic medication AN only 27 (21.4%) 41 (32.5%) .53 (p < .001)

N (%)a N 5 125 N 5 89
.28** (small effect)
AN 1 broad ASD 8 (36.4%) 12 (54.5%) .63 (p 5 .056)
N 5 22 N 5 12
ASD subset 2 (33.3%) 3 (50%) N.A.
N56 N56

Antidepressants AN only 26 (20.6%) 37 (29.4%) .45 (p < .001)

N 5 126 N 5 90
.20** (small effect)
AN 1 broad ASD 5 (22.7%) 9 (40.9%) .59 (p 5 .07)
N 5 22 N 5 18
ASD subset 2 (33.3%) 3 (50%) N.A.
N56 N56

Antipsychotics AN only 1 (0.8%) 8 (6.3%) .40 (p 5 .003)

N 5 126 N 5 91
.08** (no effect)
AN 1 broad ASD 3 (13.6%) 1 (4.5%) .27 (p 5 .39)
N 5 22 N 5 18
ASD subset 0 (0%) 1 (16.7%) N.A.
N56 N56

a
From total percentage including missing data; Cohen’s d 5 calculated by the correlation difference as suggested by Cohen (1988).

we found one probable and five possible cases (Rhind et al., 2014). in adult samples (Westwood et al., 2016). Also, the number of cases
There were differences in the clinical specifiers between these studies. who met diagnostic criteria for ASD was higher in inpatients with a
The clinical cohort (Stewart et al., 2017) included people with an age more severe form of illness (Wentz et al., 2005).
range of 9–18 (mean 5 14.6), whereas the current study included peo- In this study, we found that although ED related symptoms and
ple from age 13–21 years (mean 5 17). Previous studies suggest that physical recovery did not differ between groups, there was a greater
the prevalence of ASD traits varies according to the age/duration of ill- severity of residual general comorbidity in the group with ASD. The
ness of the sample included. For example, in a recent meta-analysis group with ASD traits had had more intensive treatment at baseline
lower levels of ASD traits were reported in children/adolescents than and a large proportion of those with probable/possible diagnosis
 1098108x, 2018, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/eat.22823 by <Shibboleth>[email protected], Wiley Online Library on [10/11/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
178 | NAZAR ET AL.

remained in treatment. In the Swedish cohort followed up over 30 Research Title: Expert Carers Helping Others (ECHO) (IRAS Code:
years, participants with ASD traits were found to have a poorer global 55754 CSP: 55754). The views expressed in this publication are
outcome (using Morgan and Russell scales) (Nielsen et al., 2015). Inpa- those of the authors and not necessarily those of the NHS, the
tients with ASD traits and/or diagnosis also didn’t improve as much as NIHR, or the Department of Health. SL, JT, and US receive salary
their non-ASD counterparts from group Cognitive Remediation Ther- support from the National Institute for Health Research (NIHR)
apy (Tchanturia, Larsson, & Adamson, 2016). Interestingly, this study Mental Health Biomedical Research Centre at South London and
found that although people with ASD engaged with treatment as much Maudsley NHS Foundation Trust and King’s College London. US is
as non-ASD, the same was found in another adult inpatient sample supported by an NIHR Senior Investigator Award. CR was supported
(Huke, Turk, Saeidi, Kent, & Morgan, 2014) by the Psychiatry Research Trust.

7 | CLINICAL IMPLICATIONS ORC ID

Bruno Palazzo Nazar http://orcid.org/0000-0002-8325-1088
Patients with ASD comorbidity have a poorer global outcome (with
residual problems that relate to their ASD traits), although eating symp-
RE FE RE NC ES
toms do appear for the most part to respond to standard care. It is pos-
American Psychiatric Association (2000). Diagnostic and statistical manual
sible that this group need a form of psychosocial intervention which
of mental disorders, fourth edition, text revision (DSM-IV-TR) (4th ed.,
focuses more on social identity and functioning (McNamara & Parsons, Vol. 1). Arlington, VA: American Psychiatric Association.
2016). For example, the addition of the New Maudsley method of col- Anckarsäter, H., Hofvander, B., Billstedt, E., Gillberg, I. C., Gillberg, C.,
laborative care, which equips families with skills training to improve Wentz, E., & Råstam, M. (2012). The sociocommunicative deficit sub-
their understanding and support for the individual, was found to group in anorexia nervosa: Autism spectrum disorders and neurocog-
nition in a community-based, longitudinal study. Psychological
improve peer functioning and prosocial behavior in AN patients (Hibbs
Medicine, 42(9), 1957–1967.
et al., 2015).
Baron-Cohen, S., Jaffa, T., Davies, S., Auyeung, B., Allison, C., & Wheel-
wright, S. (2013). Do girls with anorexia nervosa have elevated autis-
8 | STRENGTHS AND LIMITATIONS tic traits? Molecular Autism, 4(1), 24.
Bauer, S., Winn, S., Schmidt, U., & Kordy, H. (2005). Construction, scor-
A major strength of this study was the use of multimodal and multi- ing and validation of the short evaluation of eating disorders (SEED).
European Eating Disorders Review, 13(3), 191–200.
informant assessments measures (DAWBA; SAS; SDQ) and the setting
Goodman, R., & Scott, S. (1999). Comparing the Strengths and Difficul-
within a RCT with repeated assessment measures (e.g., monthly weight
ties Questionnaire and the Child Behaviour Checklist: is small beauti-
measures). Nevertheless, given the low prevalence of ASD traits, the ful? J Abnorm Child Psychol, 27(1), 17–24.
study may not have had sufficient power to detect different outcomes. Goodman, R., Ford, T., Richards, H., Gatward, R., & Meltzer, H. (2000).
Also, a full cost effectiveness analysis has not been completed. The development and well-being assessment: Description and initial
Although this study was representative of adolescents with AN pre- validation of an integrated assessment of child and adolescent psy-
chopathology. Journal of Child Psychology and Psychiatry, 41(5), 645–
senting across the United Kingdom, there may have been a degree of
655.
selection, as carers had to agree to participate in order for adolescents
Henry, J. D., & Crawford, J. R. (2005). The short-form version of the
to be included in the study. Also, we didn’t use a structured clinical depression anxiety stress scales (DASS-21): Construct validity and
interview with patients to diagnose ASD and we didn’t control for the normative data in a large non-clinical sample. British Journal of Clinical
presence of other comorbidities. Finally, the use of a six-item assess- Psychology, 44(2), 227–239.

ment (SEED) might have impaired our analysis as it didn’t cover all the Hibbs, R., Magill, N., Goddard, E., Rhind, C., Raenker, S., Macdonald, P., &
Treasure, J. (2015). Clinical effectiveness of a skills training interven-
aspects of ED symptomatology.
tion for caregivers in improving patient and caregiver health following
in-patient treatment for severe anorexia nervosa : Pragmagmatic
9 | CONCLUSION randomised controlled trial. British Journal of Psychiatry Open, 1(Sep-
tember), 56–66.

For the most part, the eating disorder symptoms in adolescents/young Huke, V., Turk, J., Saeidi, S., Kent, A., & Morgan, J. F. (2014). The clinical
implications of high levels of autism spectrum disorder features in
adults with AN and ASD traits resolve in the same time frame, and to
anorexia nervosa: A pilot study. European Eating Disorders Review, 22
the same degree, as those of people with AN without such traits. Con- (2), 116–121.
tinuing problems in social and emotional functioning occurred in those Liddle, E. B., Batty, M. J., & Goodman, R. (2009). The social aptitudes
with broad ASD traits and those with an ASD diagnosis. scale: An initial validation. Social Psychiatry and Psychiatric Epidemiol-
ogy, 44(6), 508–513.

AC KNOW LE DGME NT S McNamara, N., & Parsons, H. (2016). “Everyone here wants everyone
else to get better”: The role of social identity in eating disorder
This report/article presents independent research commissioned by recovery. British Journal of Social Psychology, 55(4), 662–680.
the National Institute for Health Research (NIHR) under its Research Nazar, B. P., Bernardes, C., Peachey, G., Sergeant, J., Mattos, P., & Treas-
for patient benefit (RfPB) programme (PB–PG-0609–19025). ure, J. (2016). The risk of eating disorders comorbid with attention-
 1098108x, 2018, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/eat.22823 by <Shibboleth>[email protected], Wiley Online Library on [10/11/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
NAZAR ET AL. | 179

deficit/hyperactivity disorder: A systematic review and meta-analysis. autistic traits in anorexia nervosa: A systematic review and meta-
International Journal of Eating Disorders, 49(12), 1045–1057. analysis. Journal of Autism and Developmental Disorders, 46(3), 964–
Nielsen, S., Anckarsäter, H., Gillberg, C., Gillberg, C., Råstam, M., & 977.
Wentz, E. (2015). Effects of autism spectrum disorders on outcome Westwood, H., & Tchanturia, K. (2017). Autism Spectrum Disorder in
in teenage-onset anorexia nervosa evaluated by the Morgan-Russell Anorexia Nervosa: An updated literature review. Curr Psychiatry Rep,
outcome assessment schedule: A controlled community-based study. 19(7), 41.
Molecular Autism, 6, 14. World Health Organization (1993). The ICD-10 classification of mental
Rhind, C., Bonfioli, E., Hibbs, R., Goddard, E., Macdonald, P., Gowers, S., & and behavioral disorders: Diagnostic criteria for research. Geneva, Swit-
Treasure, J. (2014). An examination of autism spectrum traits in adoles- zerland: World Health Organization.
cents with anorexia nervosa and their parents. Molecular Autism, 5(1), 56.
Stewart, C. S., McEwen, F. S., Konstantellou, A., Eisler, I., & Simic, M.
(2017). Impact of ASD traits on treatment outcomes of eating disor- SUP POR TI NG INFOR MATION
ders in girls. European Eating Disorders Review, 25(2), 123–128. Additional Supporting Information may be found online in the sup-
Tchanturia, K., Larsson, E., & Adamson, J. (2016). How anorexia nervosa porting information tab for this article.
patients with high and low autistic traits respond to group cognitive
remediation therapy. BMC Psychiatry, 16(1), 334.
Tchanturia, K., Smith, E., Weineck, F., Fidanboylu, E., Kern, N., Treasure,
J., & Baron Cohen, S. (2013). Exploring autistic traits in anorexia: A How to cite this article: Nazar BP, Peynenburg V, Rhind C, et al.
clinical study. Molecular Autism, 4(1), 44. An examination of the clinical outcomes of adolescents and
Wentz, E., Lacey, J. H., Waller, G., Råstam, M., Turk, J., & Gillberg, C. young adults with broad autism spectrum traits and autism
(2005). Childhood onset neuropsychiatric disorders in adult eating spectrum disorder and anorexia nervosa: A multi centre study.
disorder patients. European Child & Adolescent Psychiatry, 14(8),
Int J Eat Disord. 2018;51:174–179. https://doi.org/10.1002/
431–437.
eat.22823
Westwood, H., Eisler, I., Mandy, W., Leppanen, J., Treasure, J., & Tchan-
turia, K. (2016). Using the autism-spectrum quotient to measure