Content uploaded by Claire King
Author content
All content in this area was uploaded by Claire King on Mar 30, 2022
Content may be subject to copyright.
ORIGINAL PAPER
A Systematic Review of People with Autism Spectrum Disorder
and the Criminal Justice System
Claire King •Glynis H. Murphy
Springer Science+Business Media New York 2014
Abstract This paper provides a systemic review of the
available literature on people with autism spectrum disor-
der (ASD) in the criminal justice system (CJS). The review
considers two main types of study: those that examined the
prevalence of people with ASD in the CJS and those that
examined the prevalence of offending in populations with
ASD. In addition, types of offences in people with ASD,
co-morbid psychiatric diagnoses, and characteristics of
people with ASD who commit offences (including pre-
disposing factors) are considered. A combination of search
terms was used in a variety of databases in order to find all
of the available literature on this topic, and research studies
were included based on specified inclusion and exclusion
criteria. It was found that whilst there is an emerging lit-
erature base on this topic, there are a wide variety of
methodologies used, making direct comparison difficult.
Nevertheless it can be concluded so far that people with
ASD do not seem to be disproportionately over-represented
in the CJS, though they commit a range of crimes and seem
to have a number of predisposing features. There is poor
evidence of the presence of comorbid psychiatric diagnoses
(except in mental health settings) amongst offenders with
ASD, and little evidence of the oft-asserted over-repre-
sentation of certain kinds of crimes. It is recommended that
further research of good quality is required in this area,
rather than studies that examine populations that are not
representative of all those with ASD.
Keywords Autism ASD ASC Asperger syndrome
Crime Offending Criminal justice
Introduction
Autism and Asperger syndrome have been recognised as
clinical entities since the 1940s, but they have only entered
the standardised psychiatric diagnostic classification
schemes more recently. Nowadays, both are recognised as
part of a spectrum of conditions, referred to in the recently
published DSM-5 (APA 2013) as ‘autism spectrum disor-
der’ (ASD), a group of pervasive developmental disorders
characterised by impairments in social communication,
social interaction and social imagination, which were
suggested by Wing (1996) to be the ‘triad of impairments’
(Wing 1996). The prevalence of ASD is now generally
regarded to be approximately 1 in 100 (Baird et al. 2006;
Brugha et al. 2007) in Western countries.
It is not unusual for people with ASD to show chal-
lenging behaviours, and often these may become chronic
(Murphy et al. 2005) and require highly specialist inter-
ventions. At times, these behaviours put people at risk of
entering the criminal justice system (CJS), especially if
they are relatively able individuals (most jurisdictions do
not allow people with very low abilities to enter the CJS,
on the grounds that they are likely to lack mens rea, may be
unfit to plead and may not know right from wrong). In the
research literature, a series of case studies of people with
ASD (describing between one and six individuals in each
case) have appeared that describe the kinds of crimes and
circumstances that may arise for people with ASD in the
CJS (Baron-Cohen 1988; Barry-Walsh and Mullen 2004;
Chen et al. 2003; Cooper et al. 1993; Everall and Lecouteur
1990; Fujikawa et al. 2002; Haskins and Silva 2006;
Mawson et al. 1985; Murrie et al. 2002; Schwartz-Watts
2005; Toichi 2002). There are also some well known cases
of individuals with ASD who engaged in offending
behaviours that have drawn widespread media attention
C. King G. H. Murphy (&)
Tizard Centre, University of Kent, Giles Lane,
Canterbury, Kent CT2 7LR, UK
e-mail: g.h.murphy@kent.ac.uk
123
J Autism Dev Disord
DOI 10.1007/s10803-014-2046-5
(such as the case of Gary McKinnon in the UK who hacked
into US government computers looking for evidence of
UFOs, apparently causing over $800,000 worth of damage;
and see also Kumagami and Matsuura 2009 for three
examples from Japan). In addition, there are reports in
some follow-up studies of people with ASD, of behaviour
leading to the involvement of the police (Cederlund et al.
2008). So the question often asked is: are people with ASD
at raised risk of engaging in illegal behaviours?
In some ways, it could be argued that they may be at low
risk, since many people with ASD find rules helpful in
surviving in the social world, and laws are simply social
rules of a particular type. On the other hand, Howlin (2004,
p. 302) proposed four factors that might make people with
autism more likely to carry out aggressive or law-breaking
acts:
•Their increased social naivete
´may leave people with an
ASD open to manipulation by others
•A disruption of routines, or over-rigid adherence to
rules, might lead people with an ASD to becoming
aggressive
•A lack of understanding of social situations (and poor
negotiating skills) might lead to people with an ASD
becoming aggressive
•An obsessional interest might lead someone to com-
mitting an offence in the pursuit of that interest,
perhaps exacerbated by a failure to recognise the
implications of his/her behaviour for him/herself and
others (as could be argued in the case of Gary
McKinnon)
Wing (1981) also proposed that low levels of empathy in
people with an ASD may contribute to the likelihood of an
offence (though this is usually interpreted now as a deficit in
theory of mind), and similarly Lerner et al. (2012) suggested
impaired theory of mind, poor emotional regulation, and
problems with moral reasoning may raise the risk of an
offence. Meanwhile, Newman and Ghaziuddin (2008)in
reviewing a series of single-case, and small-scale studies of
people with ASD and violent offending, concluded that
psychiatric co-morbidity (which they argued was present in
most of these cases), was a major contributory factor to their
offending. Furthermore, as Mayes (2003) and Freckelton
(2012,2013) have argued, some of the characteristics of
people with ASD almost certainly impact on a person’s
fitness to plead, culpability, criminal responsibility, and
ability to survive custodial disposals, yet not all courts are
sympathetic to expert witnesses’ views about their ASD
client’s special difficulties.
The relatively new status of autism/ASD as a diagnostic
entity (not appearing in DSM until the 1980s) is likely to
have impacted on the amount of research specifically
focusing on people with ASD who commit criminal
offences. In contrast, there has been considerable research on
offending and intellectual disabilities (ID) and, given that
many people with ASD also have ID, it may be useful
therefore to explore previous research that has focused on
people with ID and offending behaviour. In any case, ID is
characterised by impairments in cognitive, language and
social abilities and so it could be hypothesised that people
with ID would have some of the same difficulties as people
with ASD. Existing research on people with ID and the CJS
has focused on three main areas: prevalence of offending
behaviour in people with ID, prevalence of ID in offender
populations; and vulnerabilities of people with ID in the CJS.
It is often said that previous research has found higher
rates of offending behaviour within populations of people
with ID, compared to non-disabled populations, but actu-
ally this finding is not robust, and the better the method-
ology of the study, the more the effect disappears. Birth
cohort studies (e.g. Hodgins 1992; Hodgins et al. 1996) and
other cohort studies (e.g. McBrien et al. 2003) reported
high rates of conviction for people with ID, but such
studies have a biased sample of people with ID, in that the
samples were determined administratively. Other research
in ID has examined how many people with ID there are in
various parts of the CJS. Some parts have reported higher
percentages than the 2 % that would be expected (e.g.
about 5–9 % of suspects in police stations have ID
according to Gudjonsson et al. 1993 and Lyall et al. 1995),
while rates are lower than expected in other places when
carefully measured (e.g. in prisons, see Fazel et al. 2008).
Rates also seem to vary across jurisdictions, no doubt at
least partly because the possibilities for diversion out of the
CJS vary across jurisdictions (see Murphy and Mason
2014, for a discussion of this). What does seem to be
overwhelmingly important in offending (in people with and
without ID) is high levels of social deprivation, so that
offenders with and without ID in prisons turn out to be very
similar in terms of social and legal characteristics
(MacEachron 1979), i.e. they are often unemployed and
unmarried, and poorly educated. Dickson et al. (2005) also
found that adolescents with ID were no more likely to have
offended than other adolescents, once poverty and social
deprivation were taken into account. Moreover it appears
that people with ID are more vulnerable in the CJS because
they may not understand their rights, are more suggestible
and acquiescent, are more likely to falsely confess and are
more likely to make poor decisions once in the CJS, if they
do not have good advice, than are the general population
(Clare and Gudjonsson 1993,1995; Gudjonsson 1992;
Perske 2011). Protections built into the CJS for people with
ID are not always effective, as very often it is difficult for
CJS staff (like lawyers or policemen) to know when
someone has an ID (Bean and Nemitz 1994; Medford et al.
2000; Talbot 2008). The ability of people with ID to
J Autism Dev Disord
123
understand the complex language and terminology used in
the CJS is also thought to be problematic, with some
people not even understanding the difference between the
words ‘guilty’ and ‘not guilty’ or having the meanings of
these words reversed (Smith 1993). Given that people with
ASD have difficulties with social rules and also have dif-
ficulty with communication (though rather different deficits
from people with ID), and given that many people with
ASD also have ID, it seems likely that those with ASD may
well have some similar difficulties to people with ID within
the CJS.
Nevertheless, in comparison to the research on people
with ID in the CJS, there are very few studies of people
with ASD. In recent years in the UK, there have been
several policy-based publications focused specifically on
Autism, such as ‘Fulfilling and Rewarding Lives, the
strategy for adults with autism’ (Department of Health
2010) and ‘Autism, a guide for criminal justice profes-
sionals’ (National Autistic Society 2011) that make refer-
ence to people with autism in the CJS. However, in order to
plan services for people with ASD who offend, and to
prevent people with ASD from offending where possible, it
is important to understand how prevalent offending is in
this population, the types of offences people may commit
and whether there are any co-existing factors that might
relate to offending behaviour. The few published reviews
that have appeared (e.g. Cashin and Newman 2009; Gomez
de la Cuesta 2010; Mourisden 2012), have been selective
descriptive reviews, rather than systematic reviews. They
have reported some evidence of higher rates of autism
within offender populations, but they noted that most of the
available research had been carried out in forensic hospital
settings. They have also reported that, when examining
populations of people with ASD, there was great variability
of evidence regarding rates of offending, with some evi-
dence of the relevance of other mental health needs, and
very little research on the experience of people with ASD
themselves in the CJS The current review planned to use
the areas highlighted in previous research as a guide to the
likely important themes when carrying out a systematic
review of the available literature.
Methods
Search Strategy
A search of professional databases was undertaken using
the following keywords as search terms:
Autism keywords
•Autis*
•ASD
•ASC
•Asperger
•Pervasive developmental disorder
Criminal justice system keywords
•CJS
•Prison
•Probation
•Court
•Secure
•Forensic
•Crim*
•Offen*
Each of the autism keywords was searched alongside each of
the CJS keywords. No date restrictions were placed on the
search and the databases searched were PsycINFO, MED-
LINE, Cochrane Database of Systematic Reviews and
Criminal Justice Abstracts. The initial search was under-
taken in April 2011, repeated in January 2012 and in January
2013, and a hand search of the most recent issue of journals
that printed two or more articles that met the inclusion
criteria was also carried out. The National Autistic Society
internet database of research published about ASD was also
searched using all of the CJS keywords. Finally the reference
lists of the articles selected were scrutinised for further
publications of relevance (see Fig. 1).
Inclusion and Exclusion Criteria
The title and abstract of each of the articles identified
through the database search was reviewed to ensure that
they met the following inclusion criteria:
•English-language
•Peer reviewed journal
•Participants with a diagnosis of an ASD according to
either ICD-10 or DSM-IV-TR criteria
•Participants had some involvement in the CJS, by way
of contact with the police, courts, prison service,
probation or secure hospitals.
Articles were excluded if:
•They only reported ‘autistic symptoms’, using a
questionnaire, with no attempt at diagnosis/interviews
•They were concerned with witnesses with ASD in the
CJS, not suspects or offenders
•They were single case studies
•They were dissertations
•They focused on treatment
•They were reviews
Each of the articles that appeared to meet the inclusion
criteria was reviewed in full to assess the quality of the
study. One paper (Soderstrom 2005) was subsequently
J Autism Dev Disord
123
excluded as it appeared to contain the same data in relation
to the topic of this review, as were published in a previous
study (which has been included) by the same author
(Soderstrom et al. 2004). Another paper was excluded on
the grounds that it measured only symptoms of ASD and
treated these as a continuum, rather than considering those
diagnosed with ASD (Hart-Kerkhoffs et al. 2009), and a
third paper was excluded because, although it examined
vulnerabilities of people with ASD in interrogative inter-
views, the participants were not actually involved in the
CJS (Maras and Bowler 2012). These three excluded
papers do not appear in the tables of resulting papers; they
are considered in the Discussion section where relevant.
Information Collected from Research/Data Analysis
The research papers identified were analysed with a focus
on the following issues:
•Within offender populations, the prevalence rates of
ASD
•Within populations of people with ASD, the prevalence
rates of offending behaviour
•Types of offence committed by people with ASD
•Psychiatric co-morbidity in people with ASD who offend
•Vulnerabilities and other characteristics of people with
ASD within the CJS
Due to the variety of methodologies used and focus of
research in this area it was not possible to complete a meta-
analysis of the data collected. Instead descriptive data were
produced and tabulated for each of the areas listed.
Figure 1shows the flow chart for articles, including the
numbers found at each stage, and the final number.
Results
Prevalence of ASD in Offender Populations
Ideally, studies of the prevalence of ASD amongst arrestees
or offenders, should entail unbiased samples (either total
PsycINFO, MEDLINE, Cochrane Database, Criminal Justice
Abstracts (April 2011): 1853 records
PsycINFO, MEDLINE, Cochrane Database, Criminal Justice Ab-
stracts (January 2012 & 2013): 4 records
NAS database (May 2012): 41
Hand searching and reference checking: 1 record.
Total: 1899
240 duplicate papers re-
moved
Screened titles and abstracts:
1659
Excluded 1569 papers
Full papers ordered and read:
90
Excluded 68 papers
Total papers included: 22
Remove duplicates
Fig. 1 Flow chart of
methodology
J Autism Dev Disord
123
population samples or random samples) of the section of
the CJS under study (for example, of those arrested as
suspects; of those appearing in court; of those convicted).
Identification of people with ASD in the sample should
entail ASD screening, followed by developmental inter-
view [such as the autism diagnostic interview (ADI)] or a
similar robust procedure. In fact, as Table 1shows, only
two of the studies came close to this level of perfection in
terms of collecting unbiased samples (studies 3 and 4), but
neither had good methodology for ASD diagnosis.
Of the seven studies in Table 1, three were from Sweden,
three from UK and one from Japan. Almost all of them
(studies 1, 2, 5, 6, 7) involved examining rates of ASD
amongst people referred for forensic psychiatric evaluation
(i.e. people in the CJS who were already thought to have
some kind of mental health issues—studies 1, 6, 7) or
amongst people already hospitalised in a forensic mental
health facility (studies 2 and 5). Only two reported preva-
lence of ASD in an unbiased sample (study 3 and 4): one
(study 3) of these involved a series of 335 cases appearing
before the juvenile courts in Japan, while the other (study 4)
involved the total populations of 12 prisons in Scotland. As
regards measures of ASC, four studies employed screening
instruments (studies 2, 4, 5, 7) followed by file audit or
interview, four employed full psychiatric evaluations
(studies 1, 5, 6, and 7, with some of these also including in-
patient stays) and one involved file review and interviews
only (study 3). Three involved at least some developmental
interviews with family members (studies 4, 6 and 7).
It can be seen from Table 1that the prevalence rate of
ASD found in all of these studies was higher than the 1 %
prevalence rate found in the general population, suggesting
that ASD is more prevalent in those people who offend.
However, the prevalence rates reported showed a great deal
of variation (from 3 % or less in study 1, 2 and 5, to up to
27 % for PDD in study 6). Most of this variation is likely to
reflect the methodology used and the type or source of the
sample (for example, forensic psychiatry samples often
produce high prevalence rates, as those referred are highly
likely to have mental health needs of some description). It
is difficult to make direct comparisons between the studies,
as there is a lack of consistency in the measures employed
and in the terminologies used to describe autism spectrum
conditions, with some studies using ‘autism’ and ‘Asper-
ger’ (e.g. study 1, 2 and 5), others using ‘pervasive
developmental disorder’ (study 3) and others classifying all
parts of the spectrum.
Prevalence of Offending Behaviour in People
with ASD
Much as unbiased samples were required from parts of the
CJS for the previous section (on prevalence of ASD in the
CJS), this section requires unbiased samples of people with
ASD, preferably all diagnosed using robust developmental
interviews (such as the ADI), in order to consider preva-
lence of offending behaviour. For example, a good study
would have a consecutive series of people within a defined
geographical area, where the clinic in question did all the
diagnostic work for the local area and kept a complete
register of all those diagnosed. Good studies would also
have an unbiased comparison sample so as to be able to
conclude whether or not people with ASD were at more or
less risk than the rest of the population for offending. The
six studies shown in Table 2came from UK (2), USA (2),
Denmark (1) and Austria (1), and only two used geo-
graphically based total registers of people (studies 9 and 10
from USA). The others attempted to obtain full samples
from particular areas and periods of time (e.g. part of
Wales in study 8; Hans Asperger’s clinic sample in study
11; in-patient referrals in study 12; community sample in
study 13). Most studies (9, 10, 11, 12, 13) had some kind of
comparison group, though this was not always well-mat-
ched to the ASD group.
It can be seen from Table 2that the prevalence figures
for offending behaviour in people with ASD, reported by
the six studies varied a great deal, with rates for people
with Asperger syndrome varying from 2.74 to 26 %, and
even up to 48 % for self-reported crime (in study 12). All
of the studies that did use a comparison group of some kind
(studies 9, 10, 11, 12, 13) found rates of offending
behaviour in the ASD group that were the same or lower
than in the comparison groups without ASD. One study
also examined prevalence rates according to the classifi-
cations of ASD (Mouridsen et al. 2008) and found higher
rates of offending in participants with Asperger syndrome,
than those with childhood autism and atypical autism. A
complicating factor in interpreting the results of these
studies is the varying way in which studies counted
‘offending’. Some studies logged only convictions and
tended to get lower rates (e.g. study 11), whereas others
counted contacts with justice departments (e.g. studies 9
and 10), getting middling rates, and yet others counted self-
report of criminal activities (e.g. study 13) and these
obtained the highest ‘offending’ rates.
Types of Offence Committed by People with ASD
In relation to types of offence, in order to be able to conclude
with confidence that people with ASD are more or less likely
to commit particular types of offences, studies need to have
unbiased samples of people with ASD and of people without
ASD. Yet, as can be seen in Table 3, out of the seven studies,
only two studies (studies 3 and 10) achieved this standard.
Cheely et al. (2012) (study 10) found young people with
ASD, when they did commit offences (as judged by contact
J Autism Dev Disord
123
with the justice department), significantly more often com-
mitted crimes against people and significantly less often
committed property offences, than young people without
ASD. They also were significantly more often involved in
school disturbances and significantly less often in probation
violations than young people without ASD. The two groups
did not differ significantly on other offence types (e.g. public
order offences, drug crimes). Kumagami and Matsuura
(2009) (study 3) found offenders with PDD engaged in
mostly similar types of crime to non-ASD offenders, though
property crime rates were lower, and sexual crime rates were
higher, than for non-ASD offenders.
Table 1 Prevalence rates of autism spectrum disorder in offender populations
Author,
country
Study population Number of
participants
Methods Data on prevalence rates (* =significant
differences)
1. Enayati
et al.
(2008),
Sweden
All offenders convicted in Sweden
referred for forensic psychiatric
investigation between 1997 and
2001
Total 2,609: 214
arsonists (155
men; 59
women); 2,395
other offenders
Mental state examinations, diagnosis
(DSM-IV), psychological testing,
personality assessments, life history and
ward observations during a 4 week
inpatient assessment by a multi-
disciplinary team
Autism: Male arsonists—1.3 %; male
other offenders—0.3 %; female
arsonists—0 %; female other
offenders—1 %
Asperger’s syndrome: Male arsonists—
7.1 %*; male other offenders—2.5 %*;
female arsonists—3.4 %; female other
offenders—2.6 %
2. Hare et al.
(1999), UK
Population of the three English
Special Hospitals (high secure
psychiatric hospitals)
Stage 1: 1,305
screened; 240
reached cut off
Stage 2: 215 of
240 files
reviewed
Stage 1: Total population of hospitals
screened with ward staff with specially
developed ASD questionnaire
Stage 2: Files reviewed for all over cut-
off on screening questionnaire, using
criteria for ASD from ICD-10
N=1,305
31 cases of autism identified
Prevalence rate of 2.4 %
3. Kumagami
and
Matsuura
(2009),
Japan
All juvenile cases tried in four
family courts in Japan in 1 year,
excluding traffic violations and
car accidents
Total cases
examined =428
Semi-structured interview by child
psychiatrist (using DSM-IV), school
records and court records
Adverse childhood experiences
questionnaire (see Table 5for details)
Number of participants in courts A, B, C
with pervasive developmental
disorder—11 (3.2 %)
Higher rate of PDD in court D which was
a specialist court—17 (18.2 %)
4. Robinson
et al.
(2012), UK
12 prisons in Scotland, including 2
with young offenders.
40 % of Scottish
prison
population:
2,458 prisoners
(127 women)
Stage 1: Screening of all 2,458 prisoners
on new 20-item instrument, based on
ASDI. For sub-sample of 126, relative
interviewed on ASDI and person
assessed on AQ, Quick test (for IQ),
reading and facial emotion recognition
test
97 (4 %) of all prisoners scored positive
(5 or more) on screening
Sub-sample: 90 of 95 negative on both
screening and AQ (5 were positive on
AQ). 29 of the 32 positive on screening,
and negative on AQ (2 positive on
both). 1 refusal. No ASDI interviews
positive
ROC curve (screening vs AQ) AUC only
59.6 %
5. Scragg and
Shah
(1994), UK
Entire male population at one high
secure psychiatric hospital
screened for Asperger
Syndrome.
392 screened at
stage 1
17 at stage 2 and 3
Stage 1: Screening of all patients’ case
notes for ‘autistic-type behaviours’. If 3
or more symptoms, went on to stage 2
Stage 2: Screening schedule for autistic
behaviour with key nurses
Stage 3: Patient interview
N=392
17 reached stage 2: 6 diagnosed with
Asperger syndrome; 3 more equivocal
Prevalence rate of 1.5 % (2.3 % including
equivocal cases)
6. Siponmaa
et al.
(2001),
Sweden
Young people (15–22 years):
consecutive referrals to Forensic
Psychiatric service, over 5 years,
after committing serious
offences
126 (originally
130, but records
for 4 participants
not available)
Interviews with all patients by social
worker; assessments by psychologists;
psychiatric interview and psychiatric
state examination; some relatives
interviewed; IQ tests also completed
Pervasive developmental disorder—34
(27 %)
Pervasive developmental disorder—not
otherwise specified—21 (17 %)
Asperger syndrome—13 (10 %)
7. Soderstrom
et al.
(2004),
Sweden
People referred for forensic
psychiatric investigation by
court, following severe violent
or sexual crimes (all later
convicted in court).
100 cases
92 men
8 women
Axis 1 disorders: SCID-I; Yale–Brown
Obsessive–Compulsive Scale (Y-
BOCS); AS Screening Questionnaire
(ASSQ) and AS diagnostic interview
(ASDI). Personality disorder: SCID-II;
PCL-R. Life History of Aggression
Scale
Interviews; file reviews; some interviews
with relatives
N=100
Autism—5 (5 %)
Asperger syndrome—3 (3 %)
Autistic spectrum disorder (NOS)—10
(10 %)
Studies are listed alphabetically. Studies may appear in several tables, if they reported several aspects. They retain the number allocated on their first appearance
ASD Autistic spectrum disorder, PDD pervasive developmental disorder, AS Asperger syndrome
J Autism Dev Disord
123
Table 2 Prevalence of offending behaviour in people with ASD
Author,
country
Study population Number of participants Methods Data on prevalence rates
8. Allen et al.
(2008), UK
Adults in South Wales with AS
and in contact with
community services, forensic
services, autism services,
probation services or prisons
126 people with AS, 33 with
offending histories
Subsamples: Staff
questionnaires on n =16;
interview with person n =6
All diagnosed with AS.
Informant questionnaires, on
history and behaviour, plus
ASDI, completed with staff;
semi-structured interviews
with person with AS (all
male)
126 people with ASD identified
in area
33 had engaged in offending
behaviour
Prevalence rate of 26 %
9. Brookman-
Frazee et al.
(2009), USA
Stratified random sample of all
12,662 children aged
6–17 years receiving services
in one/more system (mental
health, special ed, child
welfare, alcohol/drug,
juvenile justice
3,402 randomly selected but
23 % not located, 25 %
refused. Of the 2,609
contacted, 1,603 provided
data
Parents interviewed re child’s
mental health, special needs
and other factors
Child Health Questionnaire
(CHQ-PF28) and Child
behaviour Checklist (CBCL)
completed and psychiatric
diagnosis by DISC-IV
Of the 1,603 youths enrolled in
at least one service system, 42
had ASD by parent report
(and 178 had ID)
Fewer children with ASD/ID
were involved in the juvenile
justice system than children
without ASD/ID (11 vs 31 %)
10. Cheely
et al. (2012),
USA
Young people (12–18 years of
age) registered on the South
Carolina Autism and
Developmental Disabilities
Monitoring Project
609 identified with ASD File review by 2 clinicians for
diagnoses
ASD monitoring database
linked to records at Dept of
Juvenile Justice to review
charges
Of 609 young people with ASD,
32 had contact with Dept for
Juvenile Justice. So
prevalence rate =5.24 %
Mean number of charges per
person for these 32 people
with ASC was 3.3, compared
to mean of 5.7 for a
comparison group of non-
ASC youths (i.e. signif fewer
charges for ASC group)
11. Hippler
et al. (2010),
Austria
Data from archives of Vienna
University. All individuals
seen by Hans Asperger or
colleague in Vienna,
1951–1986.
73 in AP group (autism/
Asperger), of whom 12
women
104 in AF group (autistic
traits, but no diagnosis)
File review for diagnosis
Criminal records search
(Austrian Penal Register)
Results from AP group only—
N=73; only two convictions
Prevalence rate of 2.74 %
Prevalence rate for offending no
different from general
population rates
12. Mouridsen
et al. (2008),
Denmark
Follow-up of consecutive series
of children with PDD, seen as
inpatients, at Univ. Clinics of
Child Psychiatry, Copenhagen
and Aarhus 1960–1984. All
now adult
341 with PDD (113 childhood
autism, 86 atypical autism,
114 AS). Matched control
group: 933 children without
PDD, from general
population
File review for diagnosis
All participants (PDD and non-
PDD) screened through the
nationwide Danish Criminal
Register to ascertain
convictions
Childhood autism group:
(n =113) 0.9 % had criminal
record versus 18.9 % in
control group (n =339)
Atypical autism group:
(n =86) 8.1 % had criminal
record versus 14.7 % in
control group (n =252)
Asperger group: (n =114)
18.4 % had criminal record
versus 19.6 % in control
group (n =342)
13. Woodbury-
Smith et al.
(2006), UK
Adults with ASD living in one
Health District in England
(diagnosed by ADI-R and IQ
[70)
102 adults with ASD identified
but some declined etc.
Final sample: 25 adults with
ASD
Comparison group of 20
volunteers without ASD
Self-reported law breaking:
Using the Self-reported
Offending Questionnaire
Official statistics of offending
behaviour (for ASD group
only): Home Office Offenders
Index (contains only serious
crimes data)
Groups not significantly
different for age or gender
12 of the 25 ASD group self-
reported crime (48 %)
compared with 16 of the 20
comparison group (80 %)—
significant at p\0.05
Only two people with ASD
(8 %) were listed on the
Offender Index
J Autism Dev Disord
123
Table 3 Types of offences committed by people with ASD
Author,
country
Study population Number of participants Methods Types of offence
(* =statistically significant
difference)
8. Allen et al.
(2008), UK
Adults in South Wales with AS
in contact with services (see
Table 2), and had CJS
involvement
See Table 2for details
of sample
For details of type of
offences: staff
interview data used
n=16
Informant questionnaires
completed with staff
Violent conduct—13 (81 %);
threatening behaviour—12
(75 %); property
destruction—8 (50 %); drug
offences—4 (25 %); theft—4
(25 %); sexual offending—3
(19 %); fraud—1 (6 %);
motoring offences—1 (6 %);
murder—1 (6 %)
10. Cheely
et al. (2012),
USA
Young people (12-18 years) on
Autism Monitoring Project
(see Table 2) and had contact
with CJS
Comparison with matched non-
autistic group with contact
with CJS
Type of offence: data
from n =32 with
contact with Dept for
Juvenile Justice.
Comparison group: 3
matches for each ASD
person
File review by 2 clinicians to
confirm diagnoses
Records from the Dept of
Juvenile Justice reviewed for
all participants
ASD group (n =32)/
Comparison group (n =99):
Crimes against people—38.8/
19.8 %*
Crimes against property—20.4/
28.6 %*
Drug crimes—3.9/6.7 %
Offenses against public
order—30.1/33 %
Disturbing schools—15.5/
7.1 %*
Probation violations—1.9/
7.2 %*
Status offences—5.8/7.1 %
Other—0/0.4 %
2. Hare et al.
(1999), UK
Population of three special
hospitals (high secure
psychiatric hospitals) in July
1997
AS: n =21
Autism IQ [50: n =4
Autism IQ \50: n =6
For details of full
sample see Table 1
For type of offence, data
given refers to the 31
people with AS or
autism
For some of the 31 participants
with ASD there was no index
offence, as management
difficulties in previous
placement led to transfer to
high security service
Index offence (n =31)
Asperger syndrome/autism (IQ
[50)/autism (IQ \50):
Homicide—28.56/25/16.7 %
Violence/assault—38.1/25/
16.7 %
Threatening to kill—9.52/0/
0%
Arson—19.05/25/0 %
Sexual offending—4.76/0/0 %
No index offence—0/25/
66.7 %
3. Kumagami
and Matsuura
(2009), Japan
All juvenile court cases in four
family courts in Japan
between April 2006 and
March 2007. For details of
full sample see Table 1
Type of offence: data is
from 28 people with
PDD from all four
courts
Comparison group: 289
cases no PDD, from
courts A, B, D
Semi-structured interview
School records
Court records
Property crime (stealing)—4
(14.2 %); rough crimes
(wounding)—4 (14.2 %);
robbery, attempted murder—
4 (14.2 %); sexual crime—5
(17.8 %); drug—0; arson—1
(3.5 %); status before
crime—1 (3.5 %);
trespassing—4 (14.2 %);
guns, weapons—0;
misdemeanour—0; other—5
(17.8 %)
(Non-PDD group: only
significant differences were
property crime higher in non-
PDD; sexual crimes higher in
PDD group)
J Autism Dev Disord
123
Table 3 continued
Author,
country
Study population Number of participants Methods Types of offence
(* =statistically significant
difference)
12. Mouridsen
et al. (2008),
Denmark
All adults seen as inpatients as
children with pervasive
developmental disorders
(PDD) at the Univ. Clinics of
Child Psychiatry of
Copenhagen and Aarhus
1960–1984 and had criminal
convictions as adults. See
Table 2for details of full
sample.
For type of offence, data
given is from the 29
with ASD who had
offended (out of 313)
Matched control group
of children from
general population
database, without
PDD
See Table 2Childhood autism group (1 out
of 113):
Only 1 person convicted (no
details given about type of
offence)
Atypical autism group (7 of
86)/control (37 of 252):
Violent crimes: 2.3/1.6 %;
robbery: 2.3/0.4 %;
possession of weapons: 2.3/
0.8 %; sexual offending:
0/0.8 %; arson: 2.3/0.4 %;
theft: 4.7/2.8 %; drugs: 2.3/
1.2 %; vandalism: 1.2/0.8 %;
fraud: 1.2/2.0 %; offences
against property: 2.3/1.6 %;
receiving stolen goods: 1.2/
1.2 %; driving offences: 1.2/
11.5 %*; other: 0/2.8 %
Asperger group (21 of 114)/
control (67 of 342):
Violent crimes: 1.8/2.3 %;
robbery: 1.8/0.9 %;
possession of weapons: 2.6/
0.9 %; sexual offending: 3.5/
0.9 %; arson: 4.4/0 %*; theft:
7/3.5 %; drugs: 1.8/2.3 %;
vandalism: 1.8/1.2 %; fraud:
2.6/1.5 %; offences against
property: 2.6/1.5 %;
receiving stolen goods: 0.9/
1.2 %; driving offences: 5.3/
15.5 %*; other: 3.5/5.3 %
13. Woodbury-
Smith et al.
(2006), UK
Adults with ASD living in one
Health District in England
(had to be diagnosed with
ADI-R interview and have IQ
[70). See Table 2for details
of whole sample
For types offence: 25
adults with ASD, 12
with self-reported
crime. Comparison
group: 20 volunteers
without ASD, 16 with
self-reported crime
Self-reported law breaking:
Self-reported Offending
Questionnaire
Official statistics (ASD group
only): Home Office
Offenders Index
File review
ASD group/comparison group:
Burglary =4/0 %;
robbery =0/0 %; theft:
handling stolen goods =9/
10 %; theft:
shoplifting =11/20 %; theft:
other =0/0 %; drug;
offences =11/55 %;
criminal damage =19/0 %;
violence =30/25 %
14. Woodbury-
Smith et al.
(2010), UK
ASD offenders from: medium
secure hospitals; a national
diagnostic clinic for adults
with suspected AS or HFA;
and the local health district.
ASD non-offenders from
local health services
(diagnosis of AS or HFA but
no offending)
Autistic offenders =18
men and 3 women
Autistic non-
offenders =23 men
and 3 women
Semi-structured interview
(current interests, change in
interests over time, time
spent on interest)
Health-care records from time
of index offence reviewed
(autistic offender group only)
N=21 (offender group only):
Arson =23.81 %;
deception =4.76 %;
harassment =4.76 %;
GBH =4.76 %;
ABH =4.76 %; threats to
kill =19.05 %; indecent
assault =14.29 %;
manslaughter =9.52 %;
hoax bomb
threats =4.76 %; armed
robbery =4.76 %;
assault =4.76 %; attempted
murder =4.76 %
J Autism Dev Disord
123
Only two other studies in Table 3used comparison
groups of people without ASD (studies 12 and 13),
although they did not have unbiased ASD samples. They
showed relatively few differences in types of offences
between the ASD offenders and the non-ASD offenders,
though there seemed to be a somewhat a lower rate of
driving offences (study 12) and drug offences (study 13)
among the ASD groups. Only one controlled study reported
a significantly increased rate of arson in one sub-sample
(study 12), while others reported no differences in rates of
arson compared to control groups (e.g. study 3). Enayati
et al. (2008), study 1, also reported a raised rate of arson
compared to other offences in his Asperger group, but their
rate of arson offences was no different from that in his
comparison group with no mental health diagnoses.
Other studies in Table 3neither had unbiased ASD
samples, nor did they use non-ASD control groups.
Therefore, although they show some startling figures,
such as high rates of violent conduct, threatening behav-
iour and/or arson (studies 2, 8 and 14), these effects are
almost certainly due to the source of the samples (e.g.
forensic referrals and/or hospitalised samples) rather than
true differences between people with ASD and people
without.
Co-morbid Psychiatric Diagnosis
Studies that reported on the co-morbid psychiatric diag-
noses for people with ASD who had offended (see
Table 4), all employed samples who were either already in
a mental health hospital (studies 2 and 15) or who had been
referred for forensic assessment (study 16). It was therefore
perhaps not surprising that there was a general trend of
high rates of co-morbid psychiatric diagnosis, particularly
of psychosis and personality disorder, since these were
clearly very biased samples. There were considerable dif-
ferences in the methodologies used to make the psychiatric
diagnoses, with some studies making a psychiatric diag-
nosis using a full psychiatric examination and others using
file information only.
Table 4 Psychiatric diagnosis in addition to ASD
Author,
country
Study population Number of participants Methods Psychiatric diagnosis
(PD =personality disorder)
2. Hare et al.
(1999), UK
Population of the three
English special hospitals
(high secure psychiatric
hospitals) in July 1997
For full details of the study
see Table 1. For co-
morbid psychiatric
diagnoses, only those
(n =31) with ASD
considered here
See Table 1for details of first
stages. For co-morbid
diagnoses: Psychiatric
diagnosis taken from file
information—original
diagnoses made by
psychiatric staff (pre 1995)
Autistic spectrum 9.68 %;
schizophrenia 35.48 %;
schizo-affective 3.23 %;
schizophrenia and PD
6.45 %; affective disorder
6.45 %; affective disorder
and PD 3.23 %; PD/mental
illness/learning disability
3.23 %; PD 19.35 %;
Schizoid Personality
3.23 %; organic disorder
9.68 %; other/none 9.68 %
15. La
˚ngstro
¨m
et al. (2009),
Sweden
All individuals discharged
from hospital in Sweden 1/1/
1988 to 31/12/2000, with
diagnosis of psychiatric
disorder, 15 years?in 2000.
1,421,795 files reviewed
1,089 had diagnosis of
ASC. After excluding
deceased and those
\15 years—total with
ASD 422 (317 autism;
105 AS)
Information from the
inpatient/hospital discharge
register reviewed for
diagnosis of ASC
Cross-referenced with the
National Crime Register for
people who had committed a
violent crime during
1998–2000
Current co-morbid psychiatric
diagnosis
ASD and violent crime versus
ASC no violent crime:
Schizophrenia/
psychosis =25.8/9.2 %;
depressive disorder =0/
3.8 %; substance use
disorder =16.1/0.5 %;
PD =9.7/1.5 %; other
psychiatric disorder =38.7/
12.8 %
16. Wahlund
and
Kristiansson
(2006),
Sweden
All males guilty of homicide
or manslaughter and referred
for forensic psychiatric
assessment in Sweden
1996–2001, diagnosis of
either antisocial PD or ASD
N=35–27 with Anti-
social PD and 8 with
ASC
(37 initially—2 excluded,
due to diagnostic
uncertainties)
File review
Psychiatric diagnoses made
during forensic psychiatric
assessment
Wechsler Adult Intelligence
Scale-Revised
Psychopathy Checklist-
Revised (PCL-R)
Current co-morbid psychiatric
diagnosis Autism group
(N =8):
PD: 5 (62.5 %); substance
abuse: 4 (50 %); IQ \85: 2
(25 %); IQ 85–115: 5
(62.5 %); IQ [115: 1
(12.5 %)
J Autism Dev Disord
123
Other Results
Table 5summarises results from the nine studies (seven
from UK and two from Japan) that provide data pertaining
to characteristics of offenders with ASD or risk factors in
relation to offenders with ASD. On the whole, studies have
taken rather different approaches to this issue. Allen et al.
(2008), study 8, explored the disposals used in the CJS for
people with ASD who offended and found that almost half
of the cases in their study were not addressed through the
CJS (i.e. they were diverted out of the CJS). They also
collected qualitative information from the service users
themselves (and Allen et al. is the only study to have done
this), and from staff about the predisposing and precipi-
tating factors for the offences of the people with ASD (see
Table 5). Studies 3 and 17 also examined some predis-
posing factors, adverse childhood experiences, and they
found high rates of physical abuse, neglect and adverse
experiences amongst the families of individuals with ASD
who had offended, as compared to those with ASD who
had not offended (study 17), while study 3 found the abuse
and adverse experiences rates were mostly higher in the
ASD group than in the non-ASD offenders.
Three studies (18, 19, 20), on the other hand, investi-
gated cognitive profiles, violence ratings, and theory of
mind deficits in samples of people with ASD, schizophre-
nia and personality disorder all of whom had committed
serious offences and were living in high secure care in the
UK. Murphy found that there were few differences in
cognitive test results between the convicted people with
ASD and those with personality disorder (study 18), though
people with ASD tended to have lower violence ratings
(study 18), and to score somewhat higher on one theory of
mind task than those with personality disorder, though
scoring lower on the other theory of mind task (study 19).
They also scored higher on the two theory of mind tasks
than the convicted people with schizophrenia (study 19).
Similarly, study 22 (Woodbury-Smith et al. 2005) found
very few deficits on tasks involving theory of mind, rec-
ognising emotions in others and executive functioning
between people with ASD who had offended and those
who had not.
Finally, several studies investigated the vulnerabilities of
people with ASD. Study 8 by Allen et al. (2008) was one of
the very few that asked people with ASD themselves, about
what they thought led up their offences, and about how they
felt about their arrest, the court process, prison and other
issues. Not surprisingly, the individuals (all male) reported a
range of pre-offending factors (being upset and agitated,
being impulsive, having a bad/illegal habit, family conflict,
mental health problems, work problems, bereavement) and
they often recognised that they had tried to cope in mal-
adaptive ways. While there were some positive accounts of
helpful lawyers and/or police, many of the participants found
the CJS frightening, stressful and confusing; they felt their
Asperger syndrome had often not been understood or taken
into account; and they wanted someone to explain to them
what was going on. One other study (21, by North et al. 2008)
examined other possible vulnerabilities in people with ASD
compared to those without ASD. No differences were
reported in suggestibility between the groups, but the ASD
group scored higher on anxiety and depression, paranoia, and
on fear of negative evaluation and on compliance than those
without ASD.
Discussion
Summary of Findings and Interpretation
There is some emerging research on people with ASD in
the CJS, but the poor quality of much of the research and
the variation of both methodologies and specific focus in
each study allows only tentative conclusions.
A general failing of most studies was that their samples
were small and/or likely to be biased. Sample sizes for
people with ASD who were in contact with the CJS/had
offended were less than n =40 in 17 of the 22 studies (2,
3, 5, 7, 8, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22).
Moreover, many studies were not examining unbiased
populations of people with ASD (e.g. a consecutive sample
of people diagnosed with ASD in a geographical area), nor
unbiased populations in police stations, courts or prisons
(the exceptions being studies 3, 4, 9, 10). In addition, the
finding from Allen et al. (2008) that almost half of the
people with ASD who had offended in their participant
group did not receive CJS disposals must be taken into
consideration when considering the results of studies, as it
is unlikely that results found in convicted populations, such
as in prison and forensic hospitals, are representative of all
the people with ASD who have engaged in behaviour that
could be construed as offending (Cederlund et al. 2008,
also found in their follow-up study that some of their
participants with Asperger Syndrome had had contact with
the police but it is unclear whether they had been
convicted).
The results from the seven existing studies that focused
on prevalence rates of ASD in parts of the CJS (Table 1) all
found overall rates above 1 %, at least in the more able
Asperger groups, so it seems likely that people with ASD
are somewhat over-represented in the CJS. This conclu-
sion, though, has to be tempered by the knowledge that
poor methods for diagnosing ASD were used in the studies
with unbiased samples (studies 3 and 4), while the other 5
studies almost certainly had biased samples, since they all
came from samples referred for forensic psychiatric
J Autism Dev Disord
123
Table 5 Other characteristics and risk factors in relation to offending
Author,
country
Study population Number of
participants
Methods Other information
8. Allen et al.
(2008), UK
Adults in South Wales with
Asperger syndrome who had
involvement with the criminal
justice system
16 participants for
whom staff
interviewed
6 people with AS
also interviewed
Total =16
Informant questionnaires completed
with staff
Semi-structured interview completed
with person with Asperger syndrome
Disposal:
Prison 31 %; hospital disposal 6 %;
community order 19 %; no further
action/not through CJS 44 %
Predisposing factors to offending (staff
view):
Lacked concern for outcome 94 %;
social naivety 88 %; lacked
awareness of outcome 82 %;
impulsivity 63 %; misinterp. of rules
63 %; overriding obsession 44 %
Precipitating factors for offending
(staff view):
Social rejection—69 %; bullying—
50 %; sexual rejection—50 %;
family conflict—50 %; deterioration
in mental health—31 %; change of
domicile—25 %; change in
professional support—19 %;
bereavement—13 %
17. Kawakami
et al. (2012),
Japan
ASD group: all diagnosed by child
psychiatrists using DSM-IV (age
range 6–30 years)—divided into
offenders and non-offenders.
ASD: 175 (147
men, 28 women)
36 with criminal
history and
matched group of
139 without
Logged: Childhood adversity: parent
mental illness, substance abuse or
criminality, family violence, physical
or sexual abuse, neglect. Plus:
parental death, divorce, or loss, life
threatening childhood illness,
extreme economic adversity
Types of criminal behaviour included
theft (55 %), sexual misconduct
(25 %), violence (25 %), running
away (19 %), arson (11 %),
blackmail (6 %), other (spoof emails)
(3 %). ASD criminal group
significantly higher than ASD non-
criminal group on age at diagnosis
and on childhood adversity in: family
violence, physical and sexual abuse
and neglect, parental death, divorce
and other parental loss. Regression
showed age of diagnosis, physical
abuse and neglect most important
3. Kumagami
and Matsuura
(2009), Japan
All juvenile court cases in family
courts in Japan between April
2006 and March 2007
See Table 1for
details
28 with PDD
Semi-structured interview (including
the Adverse Childhood Experiences
questionnaire)
School records
Court records
Adverse childhood experiences
PDD group (n =28)/non-PDD
(n =289):
Recurrent physical abuse—21.4/
11.8 %; recurrent emotional abuse—
25/10.7 %; sexual abuse—0/0.3 %;
alcohol/drug user in house—10.7/
5.5 %; mother treated violently—
10.7/9.7 %; mental illness in the
home—10.7/10.4 %; one or no
biological parents—50/42.7 %;
incarcerated household member—
7.1/7.6 %; neglect by parents—17.8/
8.6 %
18. Murphy
(2003), UK
3 groups from high secure hospital,
all with history of serious
offences: AS group and random
selection of those with
schizophrenia or PD
N=13 males in
each group: AS;
schizophrenia; PD
All 20–40 years old
WAIS-R; NART-R
Weschler Memory Scale; adult memory
and information processing battery;
classical Weigl; Violence Rating
Scale
AS group and PD group younger on
admission than schizophrenia group.
AS group less likely used alcohol
than the other groups; had lower
violence ratings than the other
groups; had higher WAIS than
schizophrenia group and higher
reading scores than both other groups
AS group and PD group: both higher
scores on some subtests of WAIS c.f.
schizophrenia group. No differences
on test scores on Weigl. Very few
differences in scores between AS
group and PD group
J Autism Dev Disord
123
assessment or samples resident in forensic psychiatric
facilities.
Equally, those six studies that focused on the prevalence
of offending in people with ASD (Table 2) are challenging
to interpret because so few are without major methodo-
logical faults. However, the 4 studies that had non-ASD
control groups all reported that people with ASD com-
mitted the same number of offences or fewer offences than
those without ASD, suggesting that people with ASD are
less likely to offend than other people of the same age and
gender (or that, if they show offending type behaviour, they
are dealt with outside the CJS). Of course, most people
with ASD prefer to operate in the social world by strict
rules, and it may be that this reduces their likelihood of
offending, leading to the findings of these well-controlled
studies.
Considering the types of offences committed by people
with ASD (seven studies, Table 3), it is again difficult to
draw firm conclusions based on the research to date, since
over half of the studies either had no controls (three
Table 5 continued
Author,
country
Study population Number of
participants
Methods Other information
19. Murphy
(2006), UK
Male patients detained under the
Mental Health Act (1983) in high
secure psychiatric care in England
As above. Total—
39: AS 13;
schizophrenia 13;
PD 13
Diagnoses made by clinicians using
ICD-10 criteria
WAIS-R; theory of mind tasks: The
revised eyes task and modified
advanced theory of mind test
Theory of mind task 1:
% of answers correct: Asperger 94.4 %;
schizophrenia 79.5 %; PD 89.7 %
Theory of mind task 2:
% of answers correct: Asperger 52.7 %;
schizophrenia 33.3 %; PD 74.3 %
20. Murphy
(2007), UK
A group of male patients with AS
detained in high security
psychiatric care in England.
13 Diagnosis of Asperger made by
experienced clinicians using the
Gillberg and Gillberg criteria. Theory
of mind tasks: the revised eyes task
and modified advanced theory of
mind test
Results =% of answers correct on
tasks
Theory of mind task 1—76.9 %
Theory of mind task 2—30.8 %
21. North et al.
(2008), UK
People with ASD from diagnostic
clinic and existing ASD database
Non-ASD group from a control data
base
26 people with ASD
(21 men, 5
women): 8 had
history of arrest
27 matched people
without ASD (21
men, 6 women): 4
had history of
arrest
Gudjonsson Suggestibility Scale;
Gudjonsscon Compliance Scale;
Hospital Anxiety and Depression
Scale; Brief Fear of Negative
Evaluation Scale; Paranoia Scale
ASD versus control group:
No significant differences on any GSS
scores (recall, confabulations, yield,
shift)
ASD group significantly higher than
controls on compliance; on anxiety
and depression; on fear of negative
evaluation and on paranoia
22. Woodbury-
Smith et al.
(2005), UK
3 groups
(a) ASD offenders from services
(b) ASD non-offenders: from local
health district
(c) General population: volunteers
ASD offenders: 18
men ?3 women
ASD non-offenders:
20 men and 3
women
General population:
17 men ?6
women
Theory of mind: Eyes Test Revised
Executive functioning: behavioural
assessment of dysexecutive
syndrome; facial expressions of
emotion stimuli and tests (FEEST)
IQ: WASI
ASD offenders versus ASD non-
offenders:
No significant differences on total
scores of any test
ASD offenders versus general
population:
No significant differences on total
scores of any test; significantly worse
on fear only in FEEST
ASD non-offenders versus general
population:
Non-offenders significantly worse on
dysexecutive syndrome tests and
Eyes test; also worse on some
emotions in FEEST (though not on
total FEEST scores)
14. Woodbury-
Smith et al.
(2010), UK
ASD offenders and ASD non-
offenders—see Table 3for details
Autistic
offenders =18
men and 3 women
Autistic non-
offenders =23
men and 3 women
Semi-structured interview on interests
Health-care records from time of index
offence reviewed (autistic offender
group only)
Special circumscribed interests
investigated in relation to crimes (for
n=21 offender group only)
More of ASD offender group had
special interests involving violence
(19 %), than ASD non-offender
group (0 %). Only one had a linked
offence though
For n =2 others: clear links between
special interest and offence
J Autism Dev Disord
123
studies) or biased (forensic) samples (three studies) or both
(three studies). Moreover, although some of the controlled
studies appeared to report higher levels of violent crime in
the ASD group (e.g. study 10), others did not (e.g. study
12). Similarly, reports of high rates of arson in the ASD
group in uncontrolled studies (e.g. study 2, study 14), were
sometimes supported (study 12) but sometimes not sup-
ported by the controlled studies (study 3). Given the liking
of people with ASD for rules, one interesting finding from
a well-controlled study (study 10) was that people with
ASD were less likely to commit probation violations than
those without ASD. However they seemed more likely to
commit crimes involving school disturbances (study 12),
perhaps reflecting the difficulties people with ASD have in
coping with the school environment.
It is important to consider the role of co-morbid psychiatric
diagnoses in people with autism spectrum conditions who
offend, especially as some researchers have proposed that
offending in this group is best seen as a function of their co-
morbid diagnoses, rather than their ASD (Newman and Gha-
ziuddin 2008; Woodbury-Smith et al. 2005). There did appear
to be a trend of higher rates of psychosis and personality dis-
order diagnoses, rather than other mental health diagnoses.
Nevertheless, the fact that these studies were all conducted in
mental health settings may simply mean that such settings are
very likely to include people with dual diagnosis.
Very few studies that met the inclusion criteria for the
current review carried out any research into the experiences
of people with ASD in the CJS, and the one study that did
(Allen et al. 2008, study 8) interviewed only 6 people.
Nevertheless, the data were illuminating since the individ-
uals provided some support for the ideas that social func-
tioning deficits, life events, mood disturbances and poor
emotional coping skills contributed to their offending. The
results need to be interpreted with caution, however, due to
the small numbers involved. Interestingly though, Allen et al.
(2008) also gathered information on staff opinions of the
precipitating and predisposing factors for offending behav-
iour and these supported most of the suggestions put forward
by Howlin (2004) as to the factors that might predispose
people with ASD to offend, i.e. that social naivete
´, misun-
derstanding of social situations, lack of understanding of the
rules, and obsessional interests might help explain why some
people with ASD offend. Other studies suggested psycho-
social adversity was also an important factor.
It might be expected that, given their communication
deficits and social functioning difficulties, people with
ASD might struggle to cope in police interviews and in
court, like people with ID do (Murphy and Mason 2014).
Evidence from Allen et al’s (2008) study suggested that
this was indeed the case, and yet North et al. (2008)
reported no differences in suggestibility between those
with ASD and those without, though people with ASD
were more compliant. Interestingly, this lack of difference
in suggestibility has been confirmed by Maras and Bowler
(2012) (a study not included here because, although the
participants had ASD, they were not involved in the CJS).
Surprisingly few deficits in theory of mind were also found
in a number of studies (Murphy 2006,2007; Woodbury-
Smith et al. 2005).
Limitations
In terms of limitations to the current review it should be
considered whether the search terms used were able to
capture all available research in this area. The search terms
were very broad, covering a wide range of terms used for
ASD and for involvement in the CJS. Only one additional
study was found by hand-searching the reference lists of
the other papers selected after the use of the search terms,
suggesting that the original search was reliably targeting
the relevant papers. Only one more study not detected by
any of the search strategies was later found to report some
data on offending type behaviour: Cederlund et al. (2008)
commented on the numbers of his participants who had
been in contact with the CJS as part of their follow-up of
people with Asperger Syndrome and autism.
In terms of limitations of the research included in the
review, individually the studies were of reasonable quality
and met the inclusion criteria, but the large variety in
populations used, from which to draw participants, and the
diverse methodologies made direct comparison difficult.
Some studies used ‘offender only’ populations, whilst
some used populations within a geographical area, which
will clearly impact on how and where the results can be
applied. Within those studies that drew their participants
from ‘offender only’ populations there was a likely vari-
ability in the reliability of the ASD diagnosis and there
were a considerable variety of types of settings from which
participants were drawn: court, prison or forensic hospital,
which would affect the implications of the findings. Not all
studies used comparison or control groups and none of the
studies attempted to match groups for factors known to
affect criminal offending such as social deprivation,
physical abuse and neglect. It was also noticeable that there
was a distinct lack of research concerning women with
ASD in the CJS. Furthermore, the studies reviewed origi-
nated in different geographical areas, and the country in
which the research was carried out may have an important
influence on the results, as jurisdictions differ in the pos-
sibilities for diversion from the CJS, for those seen as
vulnerable. Additionally, some countries, such as Sweden,
Japan and the USA, used information from national health
registers or court registers to include much larger popula-
tions of people with ASD than studies carried out in the
UK. Even using national registers, however, may not
J Autism Dev Disord
123
capture an entire population of people with ASD as there
are likely to be many people who have undiagnosed ASD.
The other methodological difference that made comparison
across studies difficult was the diversity of ways in which
‘offending’ data were gathered, with some studies using
file information, some using self-report measures, and
others using national statistics on convictions. It must be
concluded that the examination of the relationship between
ASD and offending is in its infancy. In many ways the
studies found mirror the kinds of studies on ID and
offending some years ago. This review will hopefully go
some way to indicating where improvements can be made
to the methodology of studies in the future.
References
References preceded by * are those studies included in the systematic
review tables.
*Allen, D., Evans, C., Hider, A., Hawkins, S., Peckett, H., & Morgan,
H. (2008). Offending behaviour in adults with Asperger
syndrome. Journal of Autism and Developmental Disorders,
38, 748–758.
American Psychiatric Association (APA). (2013). Diagnostic and
statistical manual of mental disorders (5th ed.). Washington,
DC: APA.
Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T.,
Meldrum, D., et al. (2006). Prevalence of disorder of the autism
spectrum in a population cohort of children in South Thames:
The special needs and autism project (SNAP). Lancet, 368,
210–215.
Baron-Cohen, S. (1988). An assessment of violence in a young man
with Asperger’s syndrome. Journal of Child Psychiatry and
Psychology, 29, 351–360.
Barry-Walsh, J. B., & Mullen, P. E. (2004). Forensic aspects of
Asperger’s syndrome. Journal of Forensic Psychiatry & Psy-
chology, 15, 96–107.
Bean, P., & Nemitz, T. (1994). Out of depth and out of sight. London:
Mencap.
*Brookman-Frazee, L., Baker-Ericzen, M., Stahmer, A., Mandell, D.,
Haine, R. A., & Hough, R. L. (2009). Involvement of youths
with autism spectrum disorders or intellectual disabilities in
multiple public service systems. Journal of Mental Health
Research in Intellectual Disabilities, 2, 201–219.
Brugha, T., McManus, S., Meltzer, H., Smith, J., Scott, F. J., Purdon,
S., et al. (2007). Autism spectrum disorders in adults living in
households throughout England report from the adult psychiat-
ric morbidity survey 2007. England: The NHS Information
Centre for Health and Social Care.
Cashin, A., & Newman, C. (2009). Autism in the criminal justice
detention system: A review of the literature. Journal of Forensic
Nursing,5(2), 70–75.
Cederlund, M., Hagberg, B., Billstedt, E., Gillberg, I. C., & Gillberg,
C. (2008). Asperger syndrome and autism: A comparative
longitudinal follow-up study more than 5 years after original
diagnosis. Journal of Autism and Developmental Disorders, 38,
72–85.
*Cheely, C. A., Carpenter, L. A., Letourneau, E. J., Nicholas, J. S.,
Charles, J., & King, L. B. (2012). The prevalence of youth with
autism spectrum disorders in the criminal justice system. Journal
of Autism and Developmental Disorders, 42, 1856–1862.
Chen, P. S., Chen, S. J., Yang, Y. K., Yeh, T. L., Chen, C. C., & Lo,
H. Y. (2003). Asperger’s disorder: A case report of repeated
stealing and the collecting behaviours of an adolescent patient.
Acta Psychiatrica Scandinavica, 107, 73–76.
Clare, I. C. H., & Gudjonsson, G. H. (1993). Interrogative suggest-
ibility, confabulation, and acquiescence in people with mild
learning disabilities (mental handicap): Implications for reliabil-
ity during police interview. British Journal of Clinical Psychol-
ogy, 32, 295–301.
Clare, I. C. H., & Gudjonsson, G. H. (1995). The vulnerability of
suspects with intellectual disabilities during police interviews: A
review and experimental study of decision-making. Mental
Handicap Research, 8, 110–128.
Cooper, S.-A., Mohamed, W. N., & Collacott, R. A. (1993). Possible
Asperger’s syndrome in a mentally handicapped transves-
tite offender. Journal of Intellectual Disability Research, 37,
189–194.
Department of Health. (2010). Fulfilling and rewarding lives, the
strategy for adults with autism in England. London: Department
of Health.
Dickson, K., Emerson, E., & Hatton, C. (2005). Self-reported anti-
social behaviour: Prevalence and risk factors amongst adoles-
cents with and without intellectual disability. Journal of
Intellectual Disability Research, 49, 820–826.
*Enayati, J., Grann, M., Lubbe, S., & Fazel, S. (2008). Psychiatric
morbidity in arsonists referred for forensic psychiatric assess-
ment in Sweden. Journal of Forensic Psychiatry and Psychol-
ogy, 19, 139–147.
Everall, I. P., & Lecouteur, A. (1990). Firesetting in an adolescent boy
with Asperger’s syndrome. British Journal of Psychiatry, 157,
284–287.
Fazel, S., Xenitidis, K., & Powell, J. (2008). The prevalence of
intellectual disabilities among 12,000 prisoners: A systematic
review. International Journal of Law and Psychiatry, 31,
369–373.
Freckelton, S. C. (2012). Expert evidence by mental health profes-
sionals: The communication challenge posed y evidence about
autism spectrum disorder, brain injuries and Huntingdon’s
disease. International Journal of Law and Psychiatry, 35,
372–379.
Freckelton, S. C. (2013). Autism spectrum disorder: Forensic issues
and challenges for mental health professionals and courts.
Journal of Applied Research in Intellectual Disabilities,26,
420–434.
Fujikawa, Y., Umeshita, S., & Mutura, H. (2002). Sexual crimes
committed by adolescents with Asperger’s disorder: Problems of
management by the viewpoint of probation officers at a family
court. Japanese Journal of Child and Adolescent Psychiatry, 43,
280–289.
Gomez de la Cuesta, G. (2010). A selective review of offending
behaviour in individuals with autism spectrum disorder. Journal
of Learning Disabilities and Offending Behaviour, 1(2), 47–58.
Gudjonsson, G. H. (1991). The ‘notice to detained persons’, PACE
codes and reading ease. Applied Cognitive Psychology, 5, 89–95.
Gudjonsson, G. H. (1992). The psychology of interrogations, confes-
sions and testimony. Chichester: Wiley.
Gudjonsson, G., Clare, I. C. H., Rutter, S., & Pearse, J. (1993).
Persons at risk during interviews in police custody: The
identification of vulnerabilities. The Royal Commission of
Criminal Justice, Research Study no. 12. London: HMSO.
*Hare, D. J., Gould, J., Mills, R., & Wing, L. (1999). A preliminary
study of individuals with autistic spectrum disorders in three
special hospitals in England. London: National Autistic Society/
Dept of Health.
Hart-Kerkhoffs, L. A., Jansen, L. M., Doreleijers, T. A., Vermeiren,
R., Minderaa, R. B., & Hartman, C. A. (2009). Autism spectrum
J Autism Dev Disord
123
disorder symptoms in juvenile suspects of sex offences. Journal
of Clinical Psychiatry, 70(2), 266–272.
Haskins, B. G., & Silva, J. A. (2006). Asperger’s disorder and
criminal behavior: Forensic–psychiatric considerations. Journal
of the American Academy of Psychiatry and the Law, 34,
374–384.
*Hippler, K., Viding, E., Klicpera, C., & Happe
´, F. (2010). Brief
report: No increase in criminal convictions in Hans Asperger’s
original cohort. Journal of Autism and Developmental Disorders,
40, 774–780.
Hodgins, S. (1992). Mental disorder, intellectual deficiency and
crime: Evidence from a birth cohort. Archives of General
Psychiatry, 49, 476–483.
Hodgins, S., Mednick, S. A., Brennan, P. A., Scgulsinger, F., &
Endberg, M. (1996). Mental disorder and crime. Archives of
General Psychiatry, 53, 489–496.
Howlin, P. (2004). Autism: Preparing for adulthood (2nd ed.).
London: Routledge.
*Kawakami, C., Ohnishi, M., Sugiyama, T., Somekl, F., Nakamura,
K., & Tsujii, M. (2012). The risk factors for criminal behaviour
in high-functioning autism spectrum disorders (HFASDs): A
comparison of childhood adversities between individuals with
HFASDs who exhibit criminal behaviour and those with HFASD
and no criminal histories. Research in Autism Spectrum Disor-
ders, 6, 949–957.
*Kumagami, T., & Matsuura, N. (2009). Prevalence of pervasive
developmental disorder in juvenile court cases in Japan. Journal
of Forensic Psychiatry and Psychology, 20, 974–987.
*La
˚ngstro
¨m, N., Grann, M., Ruchkin, V., Sjo
¨stedt, G., & Fazel, S.
(2009). Risk factors for violent offending in autism spectrum
disorder: A national study of hospitalised individuals. Journal of
Interpersonal Violence, 24, 1358–1370.
Lerner, M. D., Haque, O. S., Northrup, E. C., Lawer, L., & Bursztajn,
H. J. (2012). Emerging perspectives on adolescents and young
adults with high-functioning autism spectrum disorders, violence
ad criminal law. Journal of the American Academy of Psychiatry
and Law, 40, 177–190.
Lyall, I., Holland, A. J., Collins, S., & Styles, P. (1995). Incidence of
persons with a learning disability detained in police custody: A
needs assessment for service development. Medicine, Science
and the Law, 35, 61–71.
MacEachron, A. E. (1979). Mentally retarded offenders: Prevalence
and characteristics. American Journal of Mental Deficiency, 84,
165–176.
Maras, K. L., & Bowler, D. M. (2012). Brief report: Suggestibility,
compliance and psychological traits in high-functioning adults
with autism spectrum disorder. Research in Autism Spectrum
Disorders, 6, 1168–1175.
Mawson, D. C., Grounds, A., & Tantam, D. (1985). Violence and
Asperger syndrome: A case study. British Journal of Psychiatry,
147, 566–569.
Mayes, T. A. (2003). Persons with autism and criminal justice: Core
concepts and leading cases. Journal of Positive Behavior
Intervention, 5, 92–100.
McBrien, J., Hodgetts, A., & Gregory, J. (2003). Offending and risky
behaviour in community services for people with intellectual
disabilities in one local authority. Journal of Forensic Psychi-
atry, 14, 280–297.
Medford, S., Gudjonsson, G., & Pearse, J. (2000). The identification
of persons at risk in police custody: The use of appropriate
adults by the metropolitan police. London: Institute of Psychi-
atry and Metropolitan Police.
*Mouridsen, S. E., Rich, B., Isager, T., & Nedergaard, N. J. (2008).
Pervasive developmental disorders and criminal behaviour: A
case control study. International Journal of Offender Therapy
and Comparative Criminology, 52, 196–205.
Mourisden, S. E. (2012). Current status of research on autism
spectrum disorders and offending. Research in Autism Spectrum
Disorders, 6, 79–86.
*Murphy, D. (2003). Admission and cognitive details of male patients
diagnosed with Asperger’s syndrome detained in a special
hospital: Comparison with a schizophrenia and personality
disorder sample. Journal of Forensic Psychiatry and Psychol-
ogy, 14(3), 506–524.
*Murphy, D. (2006). Theory of mind in Asperger’s syndrome,
schizophrenia and personality disordered forensic patients.
Cognitive Neuropsychiatry, 11, 99–111.
*Murphy, D. (2007). Hare psychopathy checklist revised profiles of
male patients with Asperger’s syndrome detained in high
security psychiatric care. Journal of Forensic Psychiatry and
Psychology, 18, 120–126.
Murphy, G., Beadle-Brown, J., Wing, L., Gould, J., Shah, A., &
Homes, N. (2005). Chronicity of challenging behaviours in
people with severe intellectual disabilities and/or autism: A total
population sample. Journal of Autism and Developmental
Disorders, 35, 405–418.
Murphy, G., & Mason, J. (2014). People with intellectual disabilities
and offending behaviour. In E. Tsakanikos & J. McCarthy
(Eds.), Handbook of psychopathology in intellectual disability.
New York: Springer.
Murrie, D. C., Warren, J. I., Kristiansson, M. D., & Park, E. (2002).
Asperger’s syndrome in forensic settings. International Journal
of Forensic Mental Health, 1, 59–70.
National Autistic Society. (2011). Autism, a guide for criminal justice
professionals. London: National Autistic Society.
Newman, S. S., & Ghaziuddin, M. (2008). Violent crime in Asperger
syndrome: The role of psychiatric co-morbidity. Journal of
Autism and Developmental Disorders, 38, 1848–1852.
*North, A. S., Russell, A. J., & Gudjonsson, G. H. (2008). High
functioning autism spectrum disorders: An investigation of
psychological vulnerabilities during interrogative interview.
Journal of Forensic Psychiatry and Psychology, 19, 323–334.
Perske, R. (2011). Perske’s list: False confessions from 75 persons
with intellectual disability. Intellectual and Developmental
Disabilities, 49, 365–373.
Radley, J., & Shaherbano, Z. (2011). Asperger syndrome and arson: A
case study. Advances in Mental Health and Intellectual
Disabilities, 5, 32–36.
*Robinson, L., Spencer, M. D., Thomson, L. D. G., Stanfield, A. C.,
Owens, D. G. C., Hall, J., et al. (2012). Evaluation of a screening
instrument for autism spectrum disorders in prisoners. PLoS
One, 7, 1–8.
Schwartz-Watts, D. M. (2005). Asperger’s disorder and murder.
Journal of the American Academy of Psychiatry and the Law, 33,
390–393.
*Scragg, P., & Shah, A. (1994). Prevalence of Asperger’s syndrome
in a secure hospital. British Journal of Psychiatry, 165,
679–982.
*Siponmaa, L., Kristiansson, M., Jonson, C., Nyden, A., & Gillberg,
C. (2001). Juvenile and young adult mentally disordered
offenders: The role of child neuropsychiatric disorders. Journal
of American Academy of Psychiatry and the Law, 29, 420–426.
Smith, S. A. (1993). Confusing the terms ‘‘guilty’’ and ‘‘not guilty’’:
Implications for alleged offenders with mental retardation.
Psychological Reports, 73, 675–678.
Soderstrom, H. (2005). Clinical neuropsychiatric symptoms in
perpetrators of severe crimes against persons. Nordic Journal
of Psychiatry, 59, 246–252.
*Soderstrom, H., Sjodin, A. K., Carlstedt, A., & Forman, A. (2004).
Adult psychopathic personality with childhood-onset hyperac-
tivity and conduct disorder: A central problem constellation in
forensic psychiatry. Psychiatry Research, 121, 271–280.
J Autism Dev Disord
123
Talbot, J. (2008). Prisoners’ voices: Experiences of the criminal
justice system by prisoners with learning disabilities and
difficulties. London: Prison Reform Trust.
Toichi, M. (2002). Sexual crimes by adolescents with Asperger’s
disorder: Relation to cognitive task findings. Japanese Journal of
Child and Adolescent Psychiatry, 43, 290–300.
*Wahlund, K., & Kristiansson, M. (2006). Offender characteristics in
lethal violence with special reference antisocial and autistic
personality traits. Journal of Interpersonal Violence, 21(8),
1081–1091.
Wing, L. (1981). Asperger’s syndrome: A clinical account. Psycho-
logical Medicine, 11, 115–129.
Wing, L. (1996). The autistic spectrum. London: Robinson.
*Woodbury-Smith, M. R., Clare, I. C. H., Holland, A. J., & Kearns,
A. (2006). High functioning autistic spectrum disorder, offend-
ing and other law-breaking: Findings from a community sample.
Journal of Forensic Psychiatry and Psychology, 17, 108–120.
*Woodbury-Smith, M. R., Clare, I. C. H., Holland, A. J., Kearns, A.,
Staufenberg, E., & Watson, P. (2005). A case–control study of
offenders with high functioning autistic spectrum disorders.
Journal of Forensic Psychiatry and Psychology, 16, 747–763.
*Woodbury-Smith, M. R., Clare, I. C. H., Holland, A. J., Watson, P.
C., Bambrick, M., Kearns, A., et al. (2010). Circumscribed
interests and ‘offenders’ with autism spectrum disorders: A
case–control study. Journal of Forensic Psychiatry and Psy-
chology, 21, 366–377.
J Autism Dev Disord
123
