Introduction

The management of offending type behaviour in persons with an intellectual disability poses a major challenge to service providers. This challenge is further compounded by the fact that many persons presenting with such behaviour are non-adjudicated by the criminal justice system.

There is currently no Forensic Intellectual Disability Service in Ireland (Irish College of Psychiatrists, 2007). However, the first post in Ireland for a Consultant Forensic Psychiatrist with a special interest in intellectual disability has been sanctioned by the health Service Executive and this is a welcome development and the first step in the development of a national service.

There is also a significant lack of acute inpatient mental health services (approved centres) for people with intellectual disability and severe mental health problems (Leonard et al. Reference Leonard, Hillery and Staines2007). These factors and the continuing absence of appropriate enacted capacity legislation in Ireland (Leonard & McLaughlin, Reference Leonard and McLaughlin2009) further complicate the clinical management of offending behaviour in people with an intellectual disability as this practice largely occurs outside of a legal framework.

The association between ‘criminal offending’ and intellectual disability has for many years been the subject of scrutiny and some speculation. The received wisdom of the early and mid 19th century was based on opinion and prevailing social mores rather than methodologically sound research. The following quotation from Lindsay (Lindsay, Reference Lindsay2002) is attributed to L.M. Terman in 1916 and illustrates the thinking of the time:

A number of factors have complicated attempts to investigate and establish the precise relationship between offending and intellectual disability. These mainly comprise inconsistencies in the definition and measurement of both ‘intellectual disability’ and ‘offending’.

In particular, the inadequacy of brief IQ measures has previously been reported (Thompson et al. Reference Winter, Holland and Collins1979). Thompson et al. found that the Slosson Intelligence Test and the Wechsler Adult Intelligence Scale (WAIS) over-estimated the prevalence of intellectual disability in a prison population when compared with the WAIS full scale IQ.

It is also a challenge to distinguish a criminal offence from ‘challenging behaviour’ (Simpson & Hogg, Reference Simpson and Hogg2001a ). Challenging behaviour has been defined by Emerson as: ‘culturally abnormal behaviour of such an intensity, frequency or duration that the physical safety of the person or others is likely to be placed in serious jeopardy, or behaviour which is likely to seriously limit use of, or result in the person being denied access to, ordinary community facilities’ (Emerson, Reference Emerson1995).

For a crime to be committed criminal intent known in legal terminology as mens rea is required in addition to a specific criminal behaviour known as actus rea (Lindsay, Reference Lindsay2002). Those with an intellectual disability may be considered by carers and close contacts as incapable of criminal intent due to lack of decision-making capacity and their behaviour is then viewed within the ‘challenging behaviour’ paradigm and consequently never reaches the attention of the criminal justice system (Simpson & Hogg, Reference Simpson and Hogg2001a ).

However, the issue of decision-making capacity is not black and white. The person’s ability to make a decision depends on the match between the complexity of the decision and the cognitive abilities of the person in question at the material time. Other important factors are the context of the decision and the demand characteristics or level of stress or duress the person experiences at the time of making the decision.

Added to this, the concept of varying degrees of impairment of decision-making capacity has now been enshrined in the Irish Assisted Decision Making (Capacity) Bill 2013 (Irish Statute Book, 2013). This Bill envisages varying levels of impaired decision-making capacity being matched with corresponding decision making supports rather than the current binary situation where a person is either deemed to be legally capable of decision making or lacking decision-making capacity. The Irish statute which deals with ‘fitness to be tried’ is the Criminal Law (Insanity) Act of 2006 (Irish Statute Book, 2006) and this Act still applies binary a approach to ‘fitness’; however, this too may change with future revision of the law.

The extent of the reluctance to report behaviours as criminal acts amongst those working with the intellectually disabled has been highlighted by Lyall et al. (Reference Lyall, Holland and Collins1995) who reported that ‘tolerance levels towards offending behaviour was extremely high’ in the group homes and day centres that he surveyed. This reluctance was so intense that ‘staff at one residential establishment said they would hesitate to report rape’.

Consequently, because this behaviour is often viewed within a non-legalistic paradigm, a wealth data is available with respect to rates of ‘challenging behaviour’ in this population rather than specific criminal offences.

The prevalence of ‘challenging behaviour’ in persons with an intellectual disability has been estimated between 5.7% and 14% but again is subject to variations in case definition and sample populations (Xenitidis et al. Reference Xenitidis, Russell and Murphy2000). Rates of aggressive and self-injurious behaviour have also been estimated to be higher in those with more severe intellectual disability.

It is worth noting that criminal offending in the general population is similarly underreported. This was recognised over half a century ago by Grunhut who referred to the ‘dark figure (of unreported crime)’ (Holland et al. Reference Holland, Clare and Mukhopadhyay2002).

Over the last four decades research evidence has emerged exploring an association between intellectual disability and offending. Hodgins studied a historical cohort of >15 000 Swedish-born children (Hodgins, Reference Hodgins1992). Those with an ID were defined as those who were ‘placed in special classes for intellectually deficient children in high school and were never admitted to a psychiatric ward’. Those men who fell into this category were reported to be three times more likely to have a conviction by the age of 30 and women were nearly four times more likely. These conclusions are weakened by the fact that intellectual disability was not defined in terms of generally accepted criteria, that is, impairment of overall intelligence and impairment in adaptive behaviour, which were manifest in the developmental period.

Levels of Intellectual Disablilty may be further catagorised in terms of standardised IQ score: Mild=50–69, Moderate=35–49, Severe=20–34, Profound=<20 (World Health Organisation, 1992). The further category of ‘ borderline’ intellectual disability is not contained in ICD-10 but is widely used and is denoted by an IQ score of between 70 and 80 (American Psychiatric Association, 1994).

West & Farrington (Reference West and Farrington1973) studied two groups with IQ over 110 and IQ <90. Although the latter group does not precisely map onto an Intellectually disabled group the results were interesting in that the less able group were twice as likely to have a police record and 10 times more likely to re-offend. However, it has been reported that some of the association between lower IQ and delinquency may be accounted for by social deprivation that links both variables (Simons, Reference Simpson and Hogg1978).

However, Moffit et al. (Reference Moffit, Gabrielli, Mednick and Schulsinger1991) used data from a Danish birth cohort and found a small but significant correlation between IQ and delinquency, which was independent of social deprivation.

It should be noted that the association between IQ and offending is not linear as it is likely that below a certain level of ability the risk of offending drops off significantly. An early study illustrates this point. In the late 1950s a group of American investigators studied under-privileged boys in Massachusetts (McCord & McCord, Reference McCord and McCord1959). The group was split into a treatment and control group and into bands according to IQ levels. The investigators found that those with an IQ >110 had a 26% conviction rate while those with an IQ of between 80 and 90 had a 44% conviction rate and those with an IQ of <80 had a conviction rate of 35%, which is intermediate with respect to the other groups.

There have been varying estimates of the proportion of people held on police custody who have an intellectual disability. These range from 1% (Winter et al. Reference Xenitidis, Russell and Murphy1997) to 8% (Gudjonnson et al. Reference Gudjonnson, Clare, Rutter and Pearse1993).

The proportion of those with an intellectual disability presenting to court has also been studied by Susan Hayes (Reference Hayes1993, Reference Hayes1996) in Australia and has been estimated as high as 24% for urban and rural Australian Magistrates’ Courts. However, in rural areas the aboriginal group was overrepresented within the population having an IQ <70 and also constituted a population for whom the test battery had not been validated.

The estimates of the prevalence of intellectual disability in the prison population have varied greatly both within and between jurisdictions. The conclusions of these studies are often undermined by the methodology being limited to a non-standardised clinical assessment of intellectual disability or estimates of IQ, which are of variable validity.

An English-based study (Gunn et al. Reference Gunn, Maiden and Swintion1990) found that <1% of those in the prison population had an intellectual disability but those studied were deemed to have an intellectual disability based on ‘clinical impression’.

In another study (Coid, Reference Coid1988) it was reported that 5.1% of those requiring psychiatric reports while on remand were known to Intellectual Disability Services.

A later study (Murphy et al. Reference Murphy, Harrold and Carey1995) studied 157 men on remand and did not find any to have an IQ <70.

In contrast, Brown & Courtless (Reference Brown and Courtless1968) studied the US prison population and found that 10% may have had an intellectual disability. In a later publication the same investigators estimated that 1.6% of inmates within a US prison study sample had an IQ of <50 (Brown & Courtless, Reference Brown and Courtless1971).

Fazel conducted a systematic review of the prevalence of intellectual disability amongst 12 000 prisoners (Fazel et al. Reference Fazel, Xenitidis and Powell2008). Although no summary estimate of prevalence was calculated due to study heterogeneity, the authors found that typically 0.5–1.5% of the prison populations studied internationally were diagnosed with an intellectual disability (range 0–2.8% across studies).

There is extremely limited evidence in relation to the prevalence of intellectual disability within the Irish prison population. Murphy et al. (Reference Myers2000) reported from a sample of 10% of the Irish prison population that as many as 28.8% of Irish prisoners may have a Learning Disability as measured using the Kaufmann Brief Intelligence test, wide range achievement test and the vocabulary subtest of the Weschler Adult Intelligence Scale–Revised. On the face of it this appears to be a very high figure and the limitations of brief tests have already been noted above.

If this figure is accurate it may reflect historical service provision issues, in that those with a mild intellectual disability have not traditionally been catered for within Intellectual Disability Services in Ireland and so are more likely to have instances of unacceptable behaviour viewed within a criminal offending paradigm.

Simpson & Hogg conclude their Systematic Review of the evidence regarding offending amongst people with ID with the following comments: ‘there is no convincing evidence that the prevalence of offending among people with ID is higher than for the wider population … offending among those with IQ less than 50 is rare’ (Reference Simpson and Hogg2001a ).

People with an intellectual disability have historically been associated with particular offences, most notably sex offences and arson. With this in mind it may be more useful to ask the question: ‘is there a link between levels of Intellectual Disability and specific offence types?’.

Lindsay (Reference Lindsay2002) reviewed the literature on sex offenders with Intellectual Disabilities and concluded that there is ‘no clear evidence for the over or under representation of people with developmental disabilities amongst sex offenders’. Although Simpson & Hogg concluded that ‘there is some evidence to suggest that the relative prevalence of sexual offending, criminal damage and burglary, although not theft, are higher among people with an IQ in the “borderline” range than the general population’ (Reference Simpson and Hogg2001a ).

The characteristics of offenders with learning disability have been the subject of much interest. Holland et al Reference Holland, Clare and Mukhopadhyay. (Reference Holland, Clare and Mukhopadhyay2002) have outlined the characteristics that offenders with an intellectual disability display and these are for the most part in common with their counterparts in the general population. They are overwhelmingly young, male, and have histories of psychosocial disadvantage, offending in other family members, self-reported behavioural problems dating back to childhood and high rates of un-employment.

Simpson & Hogg commented that the evidence linking psychiatric history in those with ID and offending behaviour is ‘highly variable and no conclusions are suggested’ (Reference Simpson and Hogg2001b ).

In his review of ‘dual diagnosis’ in offenders with ID O’Brien described high rates of mental illness but the evidence is limited to case reports and clinical samples that are prone to selection bias (O’Brien, Reference O’Brien2002).

O’Brien also discussed the link between Autism spectrum disorders and offending. He emphasised that this evidence is limited by its poor quality and is based on case reports including Asperger’s original observations. Putative mechanisms for offending behaviour in those with autism were also described by O’Brien in terms of the core clinical features of autism.

It has not been established that predisposing factors are the same for all offences. A case in point is sex offending. Lindsay (Reference Lindsay2002) makes particular reference to the propensity of abusers to have suffered abuse themselves. Despite this it should be noted that the evidence for victim–offender cycles in sexual offending in the general population is inconclusive and such a link can unwittingly lead to the stigmatisation of those who have been abused as potential offenders. Lindsay also describes an inclusive list of group characteristics that again reflect social deprivation, family dysfunction, poor social adjustment and poor understanding of sexuality.

To date no previous study has been untaken to estimate the level of offending behaviour amongst service users of Irish mental health services who have an intellectual disability.

The Forensic Intellectual Disability Working Group of the Irish College of Psychiatrists was established in order to establish the level of need for a Forensic Intellectual Disability Service in Ireland and develop a College position paper. Basic data from this national survey was combined with focus group data and feedback from plenary conference proceedings to inform the conclusions and recommendations set out in the position paper People with a learning disability who offend: forgiven but forgotten? (Irish College of Psychiatrists, 2007).