Experiences of people with intellectual disabilities encountering law enforcement officials as the suspects of crime – A narrative systematic review

It is well established internationally that there is a high prevalence of intellectual disability (ID) among people in police custody. Some people with ID may face particular challenges in negotiating the forensic formalities adopted by police at the pre-trial stage of the criminal process. These challenges need to be acknowledged and mitigated through appropriate procedural safeguards in order to, at once, preserve the fact-finding accuracy of criminal investigations and minimise the risk of securing a wrongful conviction. And yet, despite the formative role which pre-trial procedures exert over the trajectory of proceedings, little is known about the experiences of people with ID during their initial interaction with law enforcement officers. In an attempt to address this research lacuna, we reviewed six databases systematically to identify studies that explore such experiences. Seven studies with a total of 1199 participants were identified. Frequently, participants with ID describe challenges in police custody, experiencing particular difficulties in understanding and communicating information. They report a paucity of appropriate supports generally in this setting and an unmet need for the provision of procedural and emotional supports. Consistent implementation of legal safeguards is necessary, along with consistent availability of accessible practical measures to support people with ID within the criminal justice


Introduction
People with intellectual disabilities (ID) are over-represented in police custody (Murphy, 2019;Young, Goodwin, Sedgwick, & Gudjonsson, 2013) and in the prison system (Fazel, Xenitidis, & Powell, 2008;Gulati et al., 2018;Hellenbach, 2012). In the UK, the Bradley (2009) reported that the prevalence of people with ID in police custody ranged from 0.5% to 9%. This compares to a community prevalence of 2.16% of adults in the UK (MENCAP, 2020). Younger, higher functioning males with ID living in unsupported settings have particularly high rates of interaction with the legal system (Lunsky, Raina, & Jones, 2012).
While it must never be forgotten that persons with intellectual disabilities do not form a homogenous group and often exhibit important individual cognitive and behavioural characteristics (Cusack, 2017;Cusack, 2017;Edwards, 2014), it is generally acknowledged that some individuals falling within this classification encounter significant communicative, as well as cognitive, challenges in responding to allegations of criminal wrongdoing (Clare, 2003;Cusack, 2020a;Morrison, Forrester-Jones, Bradshaw, & Murphy, 2019). As a preliminary point, some people with intellectual disabilities have been found to have broad deficits in memory encoding, storage and retrieval (Kebbell, Hatton, & Johnson, 2004;Ternes & Yuille, 2008). Consequently, such witnesses have been found to face significant difficulty in providing spontaneous accounts of eyewitness events (Perlman, Ericson, Esses, & Isaacs, 1994;Tully & Cahill, 1984). Moreover, research suggests that a large proportion of these witnesses are susceptible to a range of additional debilitating psychological challenges which can significantly impair their capacity to deliver accurate accounts in forensic settings (Cusack, 2020b). Numerous studies, for example, have found that some individuals with intellectual disabilities may be more suggestible, more acquiescent, more likely to confabulate and more likely to engage in nay-saying than their counterparts within the general population (Clare & Gudjonsson, 1993;Gudjonsson & Henry, 2003;Gudjonsson & Joyce, T 2011). There is also evidence to suggest that such witnesses are more likely to obfuscate generic details about an alleged incident such as names, times and dates (Beail, 2002;Kebbell, Hatton, Johnson, & O'Kelly, 2001), that they will entertain a final option bias in response to closed-multiple choice questions (Heal & Sigelman, 1995), that their knowledge of the legal process is poor and that they struggle routinely to comprehend legal terminology (Ericson & Perlman, 2001). All of these factors conspire so that people with intellectual disabilities can feel isolated, uncertain and alone both before and after their interaction with the criminal justice system (Hyun, Hahn, & McConnell, 2014;Murphy et al., 2017).
Each of these psychological challenges experienced by some individuals with intellectual disabilities, moreover, can be significantly exacerbated by a range of wider environmental factors associated with the setting in which an individual's account of events is elicited. It is particularly apparent from the research which exists in this area, that the responses of a person with an intellectual disability will be biased by both the status of the interviewing actor and the formality of the venue in which the exchange is taking place (Cusack, 2018;Gudjohnsson, Murphy and Clare 2000;Gudjonsson & Gunn, 1982).
As such this is the constituency of the criminal justice system that is arguably most open to miscarriages of justice and violations of rights (Gulati, Cusack, Kilcommins, & Dunne, 2020). Indeed, a retrospective analysis of the US register of exonerations revealed that a quarter of exonerations arising from false confessions are from people with ID (Schatz, 2018).
Suspects with intellectual disabilities -in much the same way as victims with intellectual disabilities (Cusack, 2018;Edwards, Harold, & Kilcommins, 2012;Kilcommins & Donnelly, 2014) -stand to benefit from the adoption of a principled pre-trial forensic process. As Gudjonsson (2003: 334) points out, neither the psychological challenges experienced by some individuals with intellectual disabilities, nor any related limitations in social functioning, present evidentiary challenges that are insurmountable within the criminal process: "Persons with moderate learning disability may well be able to give reliable evidence pertaining to basic facts, even when they are generally highly suggestible and prone to confabulation". However, for best evidence to prevail, the forensic design of pre-trial proceedings is key.
In this review, we focus on the experiences of people with ID who are suspects, accused persons or subjects of interest at initial contact with Law Enforcement Officers (LEOs). To date, the literature on this topic has not been reviewed systematically. Such a review is required in order to inform training and awareness of LEOs, and to inform the development of a rights-based, forensically precise approach to policy development in this area, as envisioned in the United Nations' Convention on the Rights of People with Disabilities (2006).

Aim
To systematically review the published literature concerning the experiences of people with ID who are suspects, accused persons or subjects of interest at the time of their initial contact with LEOs (at prearrest, arrest, caution, initial detention, interview/charge).

Methods
Research databases MEDLINE, SCOPUS, EMBASE, CINAHL, JSTOR and PSYCINFO (inception to 1 December 2019) were searched for English-language publications using key words: "(suspect OR detainee OR prisoner) AND (intellectual disability OR mental retardation OR learning disability) AND (police OR law enforcement OR arrest OR detention OR Garda)". ('Garda' is the Irish language word for 'police').
The electronic searches were augmented by manual searching through reference lists and websites of governmental and non-governmental organisations.
One researcher (GG) screened abstracts for inclusion criteria and full texts were requested for a second stage of screening. A second researcher was consulted (BDK) where there was uncertainty regarding inclusion. PRISMA Guidelines (Moher et al., 2009) were followed in the conduct of this review.
Published studies with information about the experiences of persons with ID at their initial contact with LEOs at the pre-trial stage of criminal proceedings (pre-arrest, arrest, caution, initial detention, interview/charge) were included. Studies were included irrespective of the severity of ID. LEOs included any police officer of any rank, including trainee, whether as a responder or a custody officer. Opinion articles or reviews that did not contain primary data were excluded, as were studies that related primarily to mental illness or other mental disorders.

Results
The search strategy yielded 666 individual abstracts (Fig. 1). 604 studies were excluded at primary screening of abstracts as not relevant to the study aim. Sixty-two full-text studies were reviewed. Seven studies met inclusion criteria (n = 7).

Findings
Seven studies that describe the views and/or experiences of people with ID on contact with LEOs were identified, relating to a total of 1199 participants from English, Australian and Canadian samples. The seven studies are summarised in Table 1. Cockram (2000) sought to study the experiences of 843 people with ID through the Australian justice system and to compare these with 2442 other offenders from the general population. According to this study, people with ID were no more likely to be arrested and charged with a criminal offence than others in the general population. Upon release, however, people with ID were subsequently re-arrested at nearly double the population rate. Moreover, at sentencing, people with ID were treated differently with respect to the types of penalties imposed; i.e. different penalties imposed for similar offences. Accused people with ID were more likely to have charges dismissed. However, individuals with ID were more likely to receive custodial sentences from lower courts in respect of offences such as sexual offences and drink driving when compared to other offenders. Leggett, Godman, and Dinani (2007) interviewed 15 people with ID about their experiences of police custody in the UK. Eleven of the 15 participants reported that they had an appropriate adult (AA) present during their interview. More than a quarter were not afforded the safeguard of an AA. There was a variable level of understanding of the role of the AA; two people did not know what an AA was while two others knew that the AA sat in on the interview but could say nothing about their purpose. A number of participants expressed concerns about the manner of the police officers during the interview. These concerns included intimidating tactics such as shouting. There was also concern about being treated unfairly by the police.

Narrative review
Many participants in this study focused on issues concerning the environment and refreshments offered to them while in custody. In a number of cases, these practical concerns took precedence over factors such as whether or not they had an AA present. A variety of emotional states were described by participants; most expressed negative feelings (n = 11). While some felt angry and frustrated at some aspect of their treatment, some were able to identify positive aspects of the situation, such as feeling that they were being listened to and understood. Talbot (2008) interviewed 154 people identified by prison staff as having learning disabilities or learning difficulties in a UK setting. At the time of arrest, people with ID were almost twice as likely as the comparison group to be unemployed. Over half had attended a so-called "special" school and they were three times as likely to have been excluded from school compared to the comparison group. At the police station, fewer than a third received support from an AA during police interview and half said they didn't know what would happen once they had been charged. Some said they had been beaten or handled roughly by the police and felt manipulated into agreeing to a police interview without support. In court, over a fifth did not understand what was going on; some didn't even know why they were in court or what they had done wrong. A majority reported that the use of simpler language in court would have helped. Raina, Arenovich, Jones, and Lunsky (2013) examined the outcome of police responses to 138 people with ID in crisis in a Canadian setting. Following police intervention, 15 were arrested, 76 were taken to the emergency department (ED) and 47 experienced on-scene resolution. Outcome was significantly related to history of forensic involvement, residence at time of crisis and crisis involving physical aggression. Those with a forensic history were more likely to be arrested and less likely to be taken to the ED. Arrest was also more likely for those who lived in less supported settings and for those whose crisis involved physical aggression. All those presenting with suicidal behaviour were sent to the ED. While this study did not include qualitative data directly from people with ID, it is included in this review because it provides indirect evidence of people with ID in crisis and coming into contact with law enforcement. As demonstrated in this study, this situation can culminate in arrest and, thereby, possibly result in experiences analogous to those identified in the other papers described here.
Jessiman and Cameron, 2017 interviewed eight people with ID as part of a cohort of 13 people with mental illness or ID, looking at their experience of AAs in police custody in the UK. Although all 13 participants had experience of being interviewed in custody at least once, only six had had an AA present. Those who had been arrested more than once reported that they had been offered an AA on some occasions but not others. They reported that this depended on whether the custody sergeant had a sympathetic attitude towards their situation.
Most participants reported that that their experience of custody was one of confusion and incomprehension, which was attributed to both their own mental state and the lack of explanation by the police. Their primary need was for support in understanding various aspects of their situation, including why they were in custody, how long they would be there, the questions that were being asked of them, and what their rights were. Most respondents were of the view that once they understood this, they would be better able to manage the situation for themselves. Some, however, felt that they would still need support to communicate effectively with the police and, in particular, to prevent misinterpretation of their verbal responses and non-verbal behaviour.
The second most common need in this study (Jessiman and Cameron, 2017) was for emotional support. Respondents' experience of custody was overwhelmingly negative as they recalled feeling intimidated, frightened, dehumanised, bullied and isolated. They wanted someone on their side and to protect against humiliation. A few respondents wanted help managing their physical well-being, including access to regular rest breaks and medication. Many complained about the effect that the environment had on them, including noise, smells and poor food. Attributes of an ideal AA in this situation were described as being caring, confidential, knowledgeable, understanding, trustworthy and a good listener and communicator.
The Scottish Equality and Human Rights Commission (2017) interviewed 38 people about their experiences of the criminal justice system, including people with ID. This study found that criminal justice agencies have no consistent way to identify people with ID. Some people with ID volunteered information about their disability during their first contact with police but others assumed that agencies were already aware. Some feared they would be treated differently because of their ID, including the range of sentencing options that could be used.
Some participants in this study reported a lack of suitable legal representation. Speaking with a solicitor over the phone in advance of a police interview was seen as inappropriate for people with ID. Such a phone call provides a solicitor with little or no time to assess the intellectual competence of the person or to determine whether they understand and are capable of following their advice. In most circumstances, the person will be advised to answer "no comment" to all police questions. However, people with ID highlighted that they found this advice difficult or impossible to adhere to. Participants also noted inconsistent use of the AA and that the court environment could be challenging; they often did not understand what was being said or what was happening around them. The solicitor was expected to provide all the required support in court. Participants in this study also expressed concerns over the perceived inconsistency in how cases in the system were diverted or dropped. Ellem and Richards (2018) interviewed three young people with cognitive disabilities (n = 3) about their interaction with the police in Australia as part of a wider study that included gathering views from service providers. The researchers found that it was challenging to recruit participants for this study. The themes elicited included a need for the young people to participate as citizens, be treated with dignity and be treated with fairness. A female participant reported that she valued having her voice heard and that she perceived the police as punitive in their approach when they fail to listen to her. Another young man found it hard to understand what was being said. The participants suggested that respect involved making allowances and giving second chances. Some spoke of positive experiences they had with the police such as being given only a warning when travelling on a train without a ticket. Some participants reported feeling unfairly targeted by the police.

Discussion
This review sought to systematically study the published literature on the experiences of people with ID who are suspects, accused persons or subjects of interest at the pre-trial stage of the criminal process.
The most striking finding from our systematic review is the paucity of published studies that include a qualitative description of the experiences of people with ID when they interface with LEOs at the initial stages of the criminal justice system. The few studies that were found had diverse objectives and were largely from the UK, save for one Canadian and one Australian sample. This raises the possibility of publication bias and English-language "Tower of Babel" bias. In essence, we know very little about the experiences of people with ID during their interface with law enforcement, especially in developing countries and, indeed, in developed, non-common law jurisdictions. Some of the included studies were conducted prior to the adoption of the UNCRPD and findings should therefore be interpreted in light of this.
While this systematic review focused on people with ID who are accused persons or suspects of crime, the themes elicited in this research are also of direct relevance to the experiences of people with ID as victims of crime (Cusack, 2018;Edwards et al., 2012;Spaan & Kaal, 2019). For victims, no less than for those accused of crime, the forensic approach adopted by the police in the early stages of a criminal investigation can have a uniquely formative impact on the trajectory of an allegation of criminal wrongdoing (Burton, Evans and Sanders, 2007). The interview process, for instance, will yield a preliminary narrative of events, will inform the police's identification of corroborating witnesses, will facilitate early determinations around witness competence and credibility and will provide valuable insights for the public prosecutor in terms of the testimonial supports which a witness might be likely to require in court (Cusack, 2018;McLeod, Philpin, Sweeting, Joyce, & Evans, 2010). These pre-trial police interactions therefore occupy a crucial role in ensuring that the legitimate needs, and human rights of persons with ID -both as victims and as suspectsare met throughout the criminal process (Keilty & Connelly, 2001). They also emphasise the importance of having specialist interviewers who can better undertake interviews with suspects with intellectual disabilities (Inclusion Ireland, 2017).
We identify three key challenges arising from the literature we have reviewed: identification of people with ID (Scottish Equality and Human Rights Commission, 2017), communication (Ellem & Richards, 2018;Jessiman et al., 2017;Talbot, 2008) and support (; Leggett et al., 2007;Scottish Equality and Human Rights Commission, 2017;Talbot, 2008). Addressing these three challenges would, not only promise to meaningfully enhance the pre-trial experience of suspects with ID and better protect their rights, but also substantially improve the forensic accuracy of modern investigative processes.

United Nations convention on the rights of persons with disabilities
The UN Convention on the Rights of Persons with Disabilities was adopted in 2006. It applies established human rights principles from the UN Declaration on Human Rights to the situation of people with disabilities. States which ratify the UN Convention commit themselves to delivering civil and political rights to people with disabilities. Addressing each of the challenges identified by this review would be consistent with spirit of the Convention. Ensuring adequate support and facilitating communication is key to ensuring that people with intellectual disabilities can exercise their capacity as enshrined in Article 12(3) of the Convention. The provision of information in an accessible format to people with a disability is required by Article 9 of the Convention. Procedural safeguards such as the provision of Appropriate Adults in legal proceedings have an important role in ensuring access to justice for people with disabilities, as envisaged in Section 13(1) of the Convention. Training programmes play an important role in combatting attitudinal barriers among criminal justice professionals. The findings of this study may usefully inform the development of such programmes, as envisaged by Article 13(2) of the convention.

Appropriate adults: support and communication
In the wake of public unrest at the treatment of two child suspects who confessed under duress to involvement in the death of Maxwell Confait in 1972 (Price & Caplan, 1977) and, against the backdrop of the publication of the Royal Commission on Criminal Procedure report in 1981, the Police and Criminal Evidence Act 1984 (together with a series of complementary non-binding Codes of Practice) was enacted in England and Wales with a view to statutorily regulating police powers and safeguarding the rights of criminal suspects (Dehaghani, 2016). One of the principle safeguards envisaged by this legislative framework was the introduction of an "appropriate adult" safeguard; for which provision was made in a series of accompanying Codes of Practice (Dehaghani & Bath, 2019).
According to paragraph 1.15 of Code C: Code of Practice for the Detention, Treatment and Questioning of Persons by Police Officers (2018), a person of any age who is suspected by an officer to either be "vulnerable", or a juvenile (i.e. under 18 years of age), must not -save in limited circumstances -"be interviewed regarding their involvement or suspected involvement in a criminal offence or offences, or asked to provide or sign a written statement under caution or record of interview, in the absence of the appropriate adult" (Code C, para.1.15). According to Dehaghani: "The AA performs a number of different, but overlapping and complementary, functions. He or she must support, advise and assist the suspect; ensure that the police act properly and fairly, informing a senior officer if not; assist with communication whilst respecting the right to silence; and ensure rights are protected, respected and understood by the suspect" (Dehaghani & Bath, 2019). The AA has a role not only in interviews but whenever the suspect is given or asked to provide information or participate in any procedure: "This includes when warnings in relation to adverse inferences are given, when rights and entitlements are explained, when samples-such as fingerprints, photographs and DNA-are to be taken, when strip or intimate searches are to be conducted, and during charge, bail and police cautions" (Dehaghani & Bath, 2019).
AAs may benefit from training to help them better understand their role and communicate this (Jessiman and Cameron, 2017) to a suspect with an intellectual disability. Indeed, this was a point explicitly noted by an expert committee established in December 2014 specifically to "examine current AA arrangements for vulnerable adults, identify shortcomings in provision, and develop recommendations for ensuring provision for all who need it" (National Appropriate Adult Network, 2015). Following a review of existing literature, data from police forces, liaison and diversion services, AA services and custody officers and interviews and consultation involving senior stakeholders and individuals with direct experience of the criminal justice system, the expert group concluded: "Findings of the research conducted for this project, and existing G. Gulati, et al. International Journal of Law and Psychiatry 71 (2020) 101609 research evidence, strongly suggest that there are two elements to the under-identification of need for AAs. One element is that the police frequently lack the expertise and training to recognise when a suspect may have a vulnerability. For example, many forms of vulnerability are complex and sometimes hidden and custody officers operate under significant time pressures. Moreover, they do not have access to appropriate screening tools to assist this process. The second element is that officers opt not to request AAs for a substantial proportion of those whom they do identify as vulnerable or who identify themselves as such. This appears to reflect officers' concerns about lack of availability of AAs and the likelihood of delay associated with securing them; a lack of clarity over the threshold for requiring an AA; and/or officers' scepticism about the need for AAs in some cases (National Appropriate Adult Network, 2015)." Other jurisdictions might do well to consider the AA model, once it is implemented correctly and consistently, underpinned by appropriate training and awareness. O'Mahony (2010) cites the usefulness of registered intermediaries in facilitating communication between defendants who need additional support and law enforcement. Such communication must include the provision of information in an accessible form. It is not known, for example, what proportion of police departments internationally provide information on rights in an accessible format, such as easy-read leaflets developed for people with ID.
The studies included in this review highlight a need for practical and emotional support for people with ID in police custody. In the UK, AAs address many aspects of this need, at least in theory, but there is a lack of consistent application of this safeguard as well as confusion among people with ID about the precise roles of the AA.

Identification of people needing support
Although lauded by Medford et al. (2003: 262) for providing "an additional safeguard for vulnerable suspects"; the studies included in this particular review suggest that the application of the "Appropriate Adult" safeguard is notably inconsistent (Jessiman et al., 2017;Leggett et al., 2007;Talbot, 2008), possibly due to a failure to identify ID among people brought into custody. In this respect, the current review echoes concerns raised by a report published by the Prison Reform Trust in the United Kingdom in 2007, where individuals were learning disabilities who participated in police interviews recounted that, "police officers did not routinely ask if they had learning difficulties and did not always believe them when they said they had" (Prison Reform Trust, 2007;Inclusion Ireland, 2017). A separate, contemporaneous study by Talbot (2008) arrived at a similar finding: "there is no routine or systematic procedure for identifying prisoners with learning difficulties or learning disabilities". Addressing this issue will present its own challenges and efforts to identify people with ID in acute law enforcement encounters or custody suites is often fraught with difficulty (Hoyano, 2015;Jacobson, 2008;Palmer & Hart, 1996). Screening tools, for instance, often lack large-scale validation or may be too time-consuming or complex to administer in busy police settings (Ali & Galloway, 2016). Police custody screening tools often focus only on acute mental illness rather than ID (Noga, Walsh, Shaw, & Senior, 2015). Close and Walker (2010) propose a "checklist of indicators" that might alert criminal justice professionals to potential ID and they recommend the involvement of a "forensic special educator" to assist police in these cases. Examples of the proposed indicators include when a suspect changes version of events in response to questions from police more than once, when a suspect agrees without objection to waiver of Miranda rights with arresting police officers, when they cannot explain Miranda rights to police officers, when a suspect agrees to allow police officers to search or confiscate personal property without a search warrant, agrees to meet and be interviewed by police officers without an attorney present, when a suspect's version(s) of events does not match information obtained by police, when a suspect does not understand the questions being asked by police or when a suspect responds to investigator queries with irrelevant and/or highly improbable versions of events (Close & Walker, 2010). The use of any screening protocol or checklist will have important considerations around training of LEO's and the privacy of the individual.

Conclusions
Identifying people who need additional supports, facilitating communication and providing practical and emotional support would go a long way towards ensuring that the rights of people with ID are better respected during their contacts with LEOs. Such parity of treatment and equity of justice is a necessary undertaking for countries that have ratified the United Nations' Convention on the Rights of People with Disabilities (2006). It is also, more broadly, an exigency of both due process as well as the maxim of natural or constitutional justice. It is imperative that research in the field of intellectual disabilities incorporates qualitative or "lived" experiences of people with ID (McNamara, 2020). The findings of this review and further research in this area could usefully inform training of LEOs and policy development in this area.

Declaration of Competing Interest
GG is Chair of the Faculty of Forensic Psychiatry at the College of Psychiatrists of Ireland. The views expressed are his own. BDK, AC, SK and CPD have no conflicts of interest to declare.