photo mc carthy Medium Speaker 

Jane McCarthy  

Group Medical Director, Danshell Group, UK
Visiting Senior Lecturer, King’s College London
Visiting Fellow, London South Bank University


Autism and Offending: Towards an Evidence-Based Approach

The purpose of this paper is to explore the evidence for offending behaviour in adults with autism spectrum disorder. Autism spectrum disorder (ASD) affects around 1 % of the adult population, but up to 30% of adults with intellectual disability. The prevalence is higher among men with a male to female ratio of around 3 to 1. ASD is a lifelong condition defined by DSM-5 to have two criteria of deficits in social communication and social interaction along with restricted and repetitive behavior, interests or activities.

Research into the nature and causes of ASD has grown significantly since the beginning of this century, having been first described by Leo Kanner and Hans Asperger in the 1940s. In the current decade, there has been a growing literature on people with ASD presenting to the Criminal Justice System (King & Murphy, 2014; Chaplin, McCarthy & Underwood, 2013).  This has partly been due to cases that have been prominent in the media, such as Gary McKinnon from the UK, who hacked into the U.S. government computers looking for evidence of UFOs.

The focus of research to date has been determining the prevalence of ASD among offenders within specific populations such as prisoners and how best to identify people with ASD across the Criminal Justice System (McCarthy et al, 2015). In addition, there is literature investigating types of offences committed by people with ASD and the characteristics of people with ASD who commit offences, including comorbidity. In addition, there has been some work on the vulnerabilities of people with ASD within the Criminal Justice System.

Studies on the prevalence of offenders with ASD across the wider forensic population have produced mixed results, but a recent systematic review did not conclude that people with ASD have higher rates of offending (King & Murphy, 2014).  Studies on prevalence rates of ASD in offender populations have been mainly conducted in the UK, Sweden, Japan, and more recently the USA. The U.K. studies have included the three high secure psychiatric hospitals which in 1999 found a prevalence rate of 2.4% among patients, but this compares with a study in a Scottish prison in 2012 which found less than 1 % of prisoners with ASD (Robinson et al., 2012). The study in Japan was of juveniles reporting a rate for pervasive developmental disorder to vary from 3% to 18%, with findings based on data generated in Specialist Courts. A Swedish study of young people found 27% to have a pervasive developmental disorder among those referred to a Forensic Psychiatric Service. This may indicate that the adolescence period is a high-risk time for a person with ASD to become involved with the Criminal Justice System. A recent study of a maximum security prison in the USA using the AQ-50 identified 4.4% of prisoners to have autistic traits (Fazio et al., 2012).

Literature on types of offences that people with ASD may commit has mainly relied on case studies. The limited evidence does indicate that young people with ASD are significantly more likely to commit crimes against people and significantly less likely to commit property offences. However, most studies do not use unbiased samples and thus it is not conclusive that people with ASD tend to commit specific type of offence acts, such as sexual offending or fire-setting behaviour. The risk for people with ASD to commit offences can be argued either way, as due to their reliance on and adherence to rules they may be less likely to commit criminal behavior, but due to their core difficulties with understanding social clues and impairments of empathy, this may increase their risk to commit offences against others, such as committing sexual offences or becoming aggressive. Any disruption to their routine or a sensory overload may lead a person with ASD to become very agitated, resulting in aggression towards others. Obsessional interests may lead to specific types of offending behaviour, such as fire-setting behaviour, cybercrime or obsessive harassment in the form of stalking. 

For the wider population of offenders, factors leading to offending are early social disadvantage, conduct problems such as truancy or aggression, substance misuse, and comorbid psychiatric conditions such as psychoses. People with ASD are not excluded from early social disadvantage or adversity, but there is little research on this in relation to future offending behaviour. Research based on clinic or secure hospital populations indicates that the majority of patients with ASD who have committed offences have a comorbid psychiatric disorder. However, drug and alcohol use is less common compared to patients without ASD in secure hospital populations. These are biased samples and therefore it is not surprising to see high rates of comorbid psychiatric diagnoses.

In addition, studies have investigated the vulnerabilities of people with ASD. Many people with ASD are frightened and confused by their experiences at the police station, court or prison. People with ASD are vulnerable to bullying and less able to cope in prison. For many, their contact with the Criminal Justice System may be the first time the diagnosis of ASD is made. Evidence to date indicates that people with ASD are not more suggestible to leading questions but may be more compliant, which may be problem when being interviewed at a police station.

It is a national policy in England to divert vulnerable offenders, such as those with mental illness, intellectual disability and autism spectrum disorder, away from the Criminal Justice System to hospital care, if indicated, or to rehabilitate them in the community rather than in  custodial options as prisons. However, in order to achieve this, we need to identify individuals with ASD early within the Criminal Justice System, such as police stations or courts, while being able to use a validated suitable screening tool in these settings will undoubtedly be a challenge.

The final section of this paper describes a study undertaken in a prison in South London to identify the extent of autistic traits among male prisoners and, secondly, the association between autistic traits and common mental health problems. The full details of the methodology are described in the paper by McCarthy et al., 2015. In total, 240 male prisoners were screened with an age range of 20 to 72 years with a mean age of 34 years. 45% of the participating prisoners were from a black minority ethnic background. We used the 20-item Autism Spectrum Quotient (AQ-20) to measure the autistic traits. Thirty-nine of the prisoners had significant autistic traits as defined by a score of greater than 10 on the AQ-20. Follow-up diagnostic assessment was carried out using the Autism Diagnostic Observation Schedule (ADOS) and, where possible, the Autism Diagnostic Interview (ADI-R). Those with autistic traits had higher rates of depression of 30% compared to 6% for prisoners with no neurodevelopmental disorder (p<0.001), and current anxiety disorders in 27% of prisoners with autistic traits compared to 11% of prisoners without a neurodevelopment disorder. Prisoners with autistic traits also had higher rates of thoughts about self-harm and self-harm behavior than prisoners without neurodevelopmental disorders. There are a numerous limitations of this study thus lowering the generalisability, but the strength was the use of standardized assessment tools in face-to-face interviews.

The evidence to date indicates that people with ASD exist in the Criminal Justice System and that they are vulnerable in terms of their health and well-being. Therefore, we need to improve our recognition of people with ASD early in the Criminal Justice System, not just through screening but the training of police officers, court staff and prison staff to recognise and support people with ASD. This does require a policy commitment for health services to work more closely with the Criminal Justice System to divert people with ASD away from prison settings. Services need to support interventions in the community that reduce the risk for reoffending in the future through the development of social skills and better mental health care for adults with ASD. The evidence base to date is still limited, partly due to biased samples and the way we undertake comparisons with offenders who do not have ASD. Future evidence must not only focus on adults in populations at risk, such as in hospitals or prisons, but should also focus on how to improve life chances and well-being for young offenders with ASD. In addition, more evidence is needed on how to best support people with ASD with their vulnerabilities at the police station and within the court system.


Fazio R. L., Pietz C. A. & Denney R. L. (2012). An estimate of the prevalence of autism-spectrum disorders in an incarcerated population. Open Access Journal of Forensic Psychology, 4, 69-80.

Robinson L., Spencer M. D., Lindsay D. G., Stanfield A. C., Owens D. G. C., Hall J. & Johnstone E. C. (2012). Evaluation of a screening instrument for Autism Spectrum Disorders in prisoners. PLoS ONE, 7, e36078.

Chaplin, E., McCarthy, J & Underwood, L (2013) Autism Spectrum Conditions and offending: An introduction to the special edition, Journal of Intellectual Disabilities and Offending Behaviour, 4(1/2), 5-8. DOI 10.1108/JIDOB-05-2013-0012

King C. & Murphy G. H. (2014). A systematic review of people with Autism Spectrum Disorder and the Criminal Justice System. Journal of Autism and Developmental Disorders, 44, 271-273.

McCarthy J., Chaplin E., Underwood L., Forrester A., Hayward H., Sabet J., Young S., Asherson P., Mills R. & Murphy D. (2015). Screening and diagnostic assessment of neurodevelopmental disorders in a male prison. Journal of Intellectual Disabilities and Offending Behaviour, 6(2), 102-111.