Introduction

Involvement with the police can be a difficult experience for anyone, let alone a person with an autism spectrum disorder (ASD). Although there are no figures for ASD involvement in the criminal justice system (CJS) (see King and Murphy 2014), evidence suggests that individuals with ASD can and do come into contact with the legal system—both as victims (e.g., Lindblad and Lainpelto 2011; Mayes 2003), and suspects (see Woodbury-Smith and Dein 2014).

When an individual encounters the CJS, their first point of contact is typically a police officer (Bradley 2009). Procedural justice theory (Lind and Tyler 1988) provides a framework for considering how individuals with ASD perceive their encounters with the police; suggesting that perceptions of the police are not shaped by the outcome of an encounter (e.g., whether or not the perpetrator is eventually found guilty), but by whether an individual believes they were treated fairly (and with dignity and respect) throughout the process (Cascardi et al. 2000; Livingston et al. 2014; Watson and Angell 2007).

Unfortunately, there have been several widely publicised cases in which encounters with the police have been less than favourable for individuals with ASD. For example, in the case of ZH v The Commissioner of Police for the Metropolis (2012), police officers were judged to have acted ‘hastily’ and to have been ‘ill-informed’ after restraining ZH, a 16-year-old boy with severe ASD and epilepsy. A subsequent report by the Independent Commission on Mental Health and Policing (Adebowale 2013) highlighted the “lack of mental health awareness and knowledge amongst staff and officers” (p. 7) and stressed that there needs to be “sufficient information and training on communication skills for people with mental health issues or learning disabilities, including autism” (p. 45).

In England and Wales, the complex difficulties posed by ASD during public service engagement are recognised by the Autism Act 2009 (the only disability-specific legislation within the jurisdiction, H.M. Government 2009) as well as by generic equality legislation. However, the ASD-specific mandatory requirements placed on public health, social care and education services do not apply to the police, and wider CJS, equivalents. Consequently, lack of proper understanding of ASD means that those with ASD may be especially vulnerable in police and wider CJS settings. The specific issue of ASD within the CJS has been prioritised for action by the UK Government,Footnote 1 emphasising that those with ASD need support from police agencies (Cummins 2011). Support is needed in view of the considerable difficulties that may arise during arrest and detention (within general community settings as well as at formal interview). These include the sensory issues that may be faced by autistic suspects when encountering the busy, noisy police custody environment; as well as an appreciation of effective communication, to ensure that autistic witnesses and suspects do not respond to questions without understanding the implications (Autism West Midlands 2015). The Equalities Act 2010 calls for frontline police to have a proper understanding of ASD as they go about their duties. Charities in the UK emphasise that effective police responses need to be specifically tailored for ASD, and that staff should be offered training on the needs of those with ASD in all police settings (e.g., Autism West Midlands 2015; National Autistic Society 2011).

Encouragingly, psychologically-guided and evidence-based implications for practice have begun to emerge in recent years, particularly in relation to police interviewing of individuals with ASD (e.g., Maras and Bowler 2010, 2014; Mattison et al. 2015). These complement a variety of mandatory and discretionary measures to ensure that police obtain best evidence from vulnerable victims and witnesses, which includes all individuals with an ASD diagnosis (Ministry of Justice 2011).

A relatively recent development in this regard is the possibility of appointing a Registered Intermediary (RI)—a communication specialist—to facilitate the process of police officers obtaining best evidence from vulnerable victims and witnesses (via the Witness Intermediary Scheme, Home Office 1999). The intermediary role is also impartial (intermediaries are responsible directly to the court). Some police services provide “non-registered” intermediaries in some cases involving vulnerable suspects. This is separate from the mandatory provision of an Appropriate Adult (AA), whose role it is to protect the rights and welfare of a suspect throughout their time in police custody (for example, by providing emotional and practical support and facilitating communication) (Home Office 2011). The role of the intermediary extends beyond police interviews is more wide-ranging than that of the AA and also extends beyond police interviews (see Plotnikoff and Woolfson 2015); for example, it includes advising CJS professionals (e.g., legal advocates and the courts) on how best to frame questions. However, provision of these measures is reliant upon the victim, witness or suspect disclosing their diagnosis or, alternatively, the police recognising vulnerability.

Currently, little is known about the experiences and perceptions of the police regarding their interactions with individuals with ASD and the adaptations they may or may not make; nor do we know whether these match the views and experiences of the ASD community. The aim of this exploratory research was to survey police officers across England and Wales in order to determine: (1) Their experiences of working with victims, witnesses and suspects with ASD in their policing roles; (2) Their awareness, and use, of mandatory provisions and discretionary adjustments to support vulnerable victims, witnesses or suspects; (3) How satisfied they are in their professional dealings with the ASD community; and (4) Police training needs. A further aim was to interpret these results in the context of the views and experiences of ASD community members (adults with ASD, as well as parents), who have engaged with police.

Methods

Surveys

Three surveys (one for police officers, one for adults with ASD, and one for parents of a person with ASD) aimed to elicit both quantitative and qualitative responses regarding experiences of ASD and policing.Footnote 2 The surveys were hosted online, in order to reach a larger sample than otherwise possible. Although comparable as a whole, questions differed slightly between the surveys to suit the participant group profile (e.g., only police officers were questioned on training needs, and only ASD communityFootnote 3 members were required to confirm that a clinical diagnosis of ASD had been received).

Both surveys sought biographical information: for police officers, this included information concerning their specific role, the number of years they had worked for the police, and their geographical location; for autism community members, this included their age, gender and geographical location. Adults were also asked to specify their diagnostic label(s), their living arrangements, their education/qualifications, and their employment status. Parents were asked to provide information regarding their child’s age, diagnostic label(s), and schooling arrangements.

Next, all respondents were asked to estimate how many times they had encountered either (a) an individual with ASD in their professional role (police respondents), or (b) the police (ASD community respondents). Questions probed how frequently the individual was a victim, witness or a suspect; the type of offence(s) involved; and the age of the individual with ASD at the time of the encounter. Exploring the disclosure of an ASD diagnosis, police respondents were questioned about the point at which the diagnosis was disclosed. ASD community members were asked whether they/their dependant’s ASD diagnosis was disclosed to the police, and what influenced this decision.

Police respondents were asked to rate how easy or difficult they found various aspects of their encounters (including explaining procedures, obtaining a written statement, interviewing and providing adequate interviewing facilities), as well as how they managed different features of ASD relevant to policing (e.g., meeting communication needs, managing sensory issues and coping with unexpected changes). They were also probed for information pertaining to what adaptations, interviewing strategies and adjustments they had employed when working with individuals with ASD, and how helpful these were. ASD community members were questioned on what adjustments and accommodations were offered to them and how satisfied they were. Police officers were asked to rate their knowledge of ASD, before answering questions in relation to the training they had received, and the training they would like. Finally, all respondents rated their satisfaction with either (a) their overall professional performance in relation to ASD (for police), or (b) one specific encounter involving the police (for ASD community members). All respondents were asked to reflect on which aspects were particularly positive or challenging, and were invited to explain why.

The majority of questions included within the survey provided forced-choice options, but respondents were encouraged (throughout) to elaborate on their responses (using open-ended text boxes).

Participants

Police

Police respondents were recruited via advertisements circulated within police forces in England and Wales, the National College of Policing, social networking sites, and through snowball sampling via existing police contacts. Although 464 police respondents completed the survey, 57 were excluded as they either failed to complete questions aside from demographic information (n = 56) or did not work in England or Wales (n = 14). This resulted in a final sample of 394. The majority were at the rank of Constable (63 %, including 22 % who were Detective Constables), but all ranks up to Superintendent were represented.

Respondents had varying amounts of police service (29 % = 6–10 years, 28 % = 11–20 years, 26 % > 20 years), and declared encounters with around 20 individuals with ASD in their role (range = 0–500, mode = 10). Although most encounters involved young adults (33 % = 18–24 years; 21 % = 25–39 years) or teenagers (17 % = 12–17 years), all age groups had been encountered (2 % ≤ 4 years; 15 % = 5–11 years; 10 % = 40–64 years; 2 % ≥ 65 years).

ASD Community

ASD community members who had encountered the CJS were recruited through advertisements placed on ASD websites (including the websites of charities, e.g., the UK’s National Autistic Society), via social media (Twitter, Facebook) and by snowball sampling within the ASD community. Of the 31 adults with ASD (16 males, 15 females) who completed the survey, 10 % were aged 18–24, 55 % = 25–39 and 35.5 % = 40–64. The majority (74 %) described their diagnosis as Asperger syndrome, 16 % as autistic disorder and 10 % as ASD. Co-occurring diagnoses were reported for 23 adults in the sample (74 %). Of these 23 adults, the most common co-occurring conditions were: mood disorder (60 %), anxiety disorder (43 %), and other developmental disorders (26 %). Most of the adult group were educated in mainstream schools (87 %), and held at least GCSE level (age 14–16) qualifications (86 %). Half the ASD participants (50 %) were in paid/voluntary employment, and 29 % were currently in education. Based on their ability to complete an online survey independently, it can be assumed that this sample was relatively high functioning (verbally and intellectually).

To capture the perspectives of those who support individuals with ASD, as well as to sample the experiences of people with ASD who are not able to respond to online questions themselves (e.g., due to their young age or severity of their autism), we also invited parents to share their experiences of autism and the CJS. Of the 49 parents who completed the survey, four were male and 43 were female (two respondents did not provide information on gender), and the majority (83 %) were aged 40–64 years (one respondent did not provide their age). Their dependents tended to be male (92 %) and were most commonly diagnosed with Asperger syndrome (66 %) or autistic disorder (28 %). Less frequently, the children were diagnosed with atypical autism (2 %), childhood autism/childhood disintegrative disorder (2 %) or Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS) (2 %) (two parents did not provide information on diagnostic labels). Co-occurring diagnoses were reported for 33 children (70 %) [Two parents did not provide information on additional diagnoses]. Of these 33 children, the most frequent co-occurring conditions were: developmental disorder (45 %), anxiety (33 %), or mood (30 %) disorders. Providing information on education (n = 48), parents reported that 60 % of the children were educated in mainstream schools; the remainder were educated in specialist schools/units (35 %) or at home (4 %).

Ethical approval for this study was granted by Royal Holloway, University of London and the University of Bath research ethics committees. All participants provided their informed consent before commencing the survey. In addition, respondents from the ASD community were required to confirm that they were no longer experiencing distress from the event.

Results

Data Analysis Plan

Not all participants answered every question, so responses do not always tally to the total number of participants. Missing data were not reconstructed. Responses were largely analysed using descriptive statistics; presenting the percentage of responses per category for each question. Most questions were answered on a 5-point scale (e.g., ‘very difficult’; ‘quite difficult’; ‘neutral’; ‘quite easy’; ‘very easy’); however, for ease of interpretation, we report findings condensed into 3-points (e.g., ‘easy’; ‘neutral’; ‘difficult’) as the overall patterns remained the same. Respondents were given opportunities to elaborate on their responses to closed questions and to provide additional comments throughout. Responses to these open questions were analysed collectively using a thematic analysis approach (Braun and Clarke 2006). This involved identifying and understanding each respondent’s perspective on their experience, in order to identify overarching themes. These were assimilated and accommodated as they emerged. Analyses were independently conducted by one of the authors (TH), with LC and KM each also coding the data independently. Findings were reviewed and discrepancies resolved before nine themes and subthemes were identified. These are presented in Table 1.

Table 1 Themes identified from open questions in online survey
Full size table

Experiences and Views of Police Officers

Nature of Reported ASD Involvement with PoliceFootnote 4

Reflecting on their professional encounters with individuals with ASD, police respondents (n = 259) estimated that 36 % involved the person as a victim, 43 % as a suspect, 10 % as a witness, and 13 % in a different capacity (only two officers provided detail of what this referred to: one reported that a police colleague was diagnosed with ASD; the other reported that the individuals encountered were involved in antisocial behaviour or domestic incidents). Respondents were asked to provide an indication of the kinds of crimes with which they had been involved. Officers (n = 263) reported that their most frequent encounters with individuals with ASD were in relation to crimes of violence: suspects (17 %), victims (21 %), witnesses (21 %). The next most frequent encounters were: for suspects, criminal damage (16 %) and sex offences (12 %)/domestic violence (12 %); for victims, sex offences (16 %) and harassment (13 %); for witnesses, domestic violence (14 %) and criminal damage (11 %).

Recognition and Disclosure of ASD

Where a disclosure was made, most police respondents (90 % of n = 260) reported that a person’s ASD diagnosis was ‘always’ or ‘sometimes’ made known to them prior to, or during, first contact. The most frequently reported source of the disclosure was the individual with ASD (always = 11 %; sometimes = 79 %) or the parent (always = 15 %; sometimes = 47 %).

Use of Appropriate Adults (AA)

An AA is a person appointed by the police to act as an independent safeguard for a vulnerable suspect in police custody. Almost all of the 192 police who responded to this question reported that an AA had always (67 %) or sometimes (31 %) been involved. Views on the utility of the AA involvement (n = 189) were mixed: 51 % of respondents found them helpful whereas 30 % reported that they were unhelpful (20 % = neutral). Responses to the open-ended questions (n = 40) indicated that negative views related to the AA not knowing the person (and therefore lacking understanding of their personal condition) or not fully understanding their roles. As can be seen in Table 1 (Section 3.6), most police respondents felt that family members and key workers, who knew the individual well, were useful AAs because they could use their knowledge of the person to provide individualised support.

Adjustments and Provisions for ASD

Police responded to questions about measures and adjustments used when interviewing someone with ASD (n = 199). As illustrated in Fig. 1, the most frequent adjustments that police made were avoiding long-winded or multiple part questions (92 %), allowing extra time to process questions (91 %) and being mindful of the vocabulary used (89 %). Responding to the question regarding how easy or difficult it was to make these adjustments (n = 175), 49 % reported it was easy and 19 % said they found it difficult (32 % = neutral). As can be seen in Table 1 (Sections 1 and 8), themes that emerged from police respondents’ open-ended responses (n = 47) indicated that factors such as time constraints and a lack of training were viewed as major barriers in enabling them to make appropriate adaptations and adjustments to support individuals with ASD in their policing roles.

Fig. 1
figure 1

Frequency of measures and adjustments reported by police respondents. A = Avoid long winded, multiple part, negative or tag questions (n = 194). B = Allow extra time to process questions/adapt the pace of the interview (n = 191). C = Use or avoid particular vocabulary (e.g., literality) in questioning (n = 194). D = Provide extra breaks or timeout in interviews (n = 186). E = Change the interview time (n = 188). F = Change the interview location (n = 187). G = Change the chronology of topics asked about in questioning (n = 182). H = Change the interview room layout (n = 184). I = Break down questioning into multiple interviews (n = 184). J = Use props in questioning (n = 189)

Full size image

Use of Intermediaries

To reiterate, the role of an intermediary is to facilitate communication during police interview, as well as throughout the criminal justice process (Ministry of Justice, 2012). Of 130 police respondents, 12 % reported that an intermediary was always involved in cases concerning individuals with ASD; 55 % stated ‘sometimes’, and 33 % stated ‘never’. The perceived helpfulness of the intermediary was mixed: 42 % (of n = 113) found them helpful and 23 % said they were unhelpful (35 % = neutral). Facilitating communication was cited as an important reason for the utility of intermediaries, but many police respondents commented that the delay in obtaining them was unhelpful (Table 1, Section 3.6).

When asked to rate how well they knew/understood the Witness Intermediary Scheme (Youth Justice and Criminal Evidence Act; Home Office 1999) and how comfortable they felt working with an intermediary, most police respondents (61 %, of n = 240) felt they had little or no knowledge of the scheme (30 % reported having some or excellent knowledge (10 % = neutral) but 38 % reported feeling comfortable working with an intermediary (45 % = neutral; 17 % = uncomfortable).

Challenges and Aspects of Policing that Work Well

Police respondents rated (on a five-point scale) how easy or difficult they found various aspects of policing in relation to people with ASD (n = 239). As seen in Fig. 2, police reported that they found obtaining a written statement and conducting interviews with someone with ASD to be the most difficult. Understanding ASD and conducting identification parades with individuals with ASD were reported to be the easiest.

Fig. 2
figure 2

Police ratings of ease/difficulty for their experiences with people with ASD in different aspects of policing. A = Obtaining a written statement (n = 128). B = Interviewing (n = 186). C = Managing distress (n = 209). D = Managing unexpected changes to procedures (n = 185). E = Explaining procedures (n = 228). F = Managing sensory issues (n = 161). G = Managing memory issues (n = 184). H = Helping to focus, engage and/or cooperate (n = 233). I = Meeting communication needs (n = 236). J = Building and maintaining rapport (n = 230). K = Providing adequate interviewing facilities and environment (n = 189). L = Finding appropriate support (n = 214). M = Carrying out identification parades (n = 28). N = Understanding ASD (n = 233)

Full size image

Police Knowledge and Training

Rating their overall knowledge of the disorder, almost half (48 %) of the 238 officers who answered this question felt they were knowledgeable (20 % = little or knowledge; and 32 % = neutral). Around half of respondents (48 %, of 237) also indicated they felt well equipped to work with individuals with ASD, but 29 % reported feeling poorly equipped (24 % = neutral).

A total of 242 police officers responded to question on police training: 37 % of officers reported that they had received training on ASD; and, of these, 13 % had received further refresher training. This covered a wide range of topics, but largely involved improving general knowledge of ASD (86 %), as well as techniques to enhance communication (66 %) or minimise distress (64 %). Training tended to be provided by police services (70 %), or by charitable or ASD-specific organisations (16 %). Almost half (49 %) of police officers who had received ASD training were satisfied, but 28 % reported that they were unsatisfied’ (23 % = neutral).

Police (n = 82) indicated which aspects of training they felt were positive. After general knowledge of ASD (79 % of respondents), the top three were: minimising distress of an individual with ASD (51 %); practical application and usefulness for their police role (44 %); and modifying interviewing techniques (35 %). Aspects of training that police (n = 58) deemed not satisfactory (top three) were: lacked sufficient focus on ASD in the CJS context (47 %); overly simplistic (45 %); and lacked practical application and relevance for their policing role (34 %).

Of 151 officers who had not received ASD training, 92 % reported that it would be useful (particularly general training on ASD = 86 %). The top three training needs identified by police (n = 138) were: enhancing communication (82 %); minimising distress (78 %); and training for working with someone with ASD that was specific to their policing role (71 %).

Qualitative analyses of the responses to open-ended questions (see Table 1) reiterated these findings, with police officers reporting a lack of existing training (Section 1.1 of Table 1) and demonstrating a strong desire for high-quality training on the topic of ASD, especially training that is role-specific (Section 1.2 of Table 1). Nevertheless, a minority of officers reported having received excellent training on ASD (Section 1.3 of Table 1). This highlights that pockets of good practice do exist but need to be more widespread.

Overall Experiences

Questioning police (n = 230) on how satisfied they felt with the way they had worked with individuals with ASD in their role, 42 % felt satisfied, and 21 % were not satisfied (37 % = neutral). A variety of factors were cited in relation to this satisfaction/dissatisfaction, as summarised in Table 1.

Experiences and Views of the ASD Community

Overall Satisfaction

Asking adults (n = 26) to describe one particular experience of the CJS (including their encounters with police) and rate how satisfactory they found it, responses were largely negative: 69 % = unsatisfactory; 11.5 % = neutral; 15 % = satisfactory; 4 % = not sure. Parent (n = 38) views were equally negative: 74 % = unsatisfactory; 13 % = neutral; 13 % = satisfactory.

Focusing on policing specifically, adult and parent respondents were asked how satisfied they were with: (1) the treatment they received when they first encountered the police; (2) the way they were treated at the initial stages of the investigation; (3) the explanation of what would happen during the police investigation; and (4) whether they felt they were treated appropriately by police interviewers. Data are presented in Table 2.

Table 2 ASD community satisfaction ratings with various aspects of policing
Full size table

Elaborating on these ratings, qualitative analyses of open-ended responses provided by ASD community members (adults n = 31; parents n = 49) highlighted: police officers lacked awareness and knowledge of ASD; the needs of individuals with ASD (physically, emotionally, cognitively and emotionally) were not being met; there was a lack of information, and explanation, given by police officers; delays at various stages were unacceptable; and some individuals felt victimised or discriminated against by police officers (see Table 1 for details).

Disclosure of Diagnosis

Questioning adults with ASD (n = 28) on whether they disclosed their ASD diagnosis revealed mixed results: 39 % = always; 36 % = never; 25 % = on some occasions, but not others. Often, this decision was linked to a fear of discrimination or victimisation by police officers (see Table 1, Section 5). In contrast, the majority of parents (n = 41) reported that their dependant’s diagnosis was disclosed either ‘every time’ or ‘sometimes’ (76 %). Only 10 % of parents deliberately chose not to disclose their dependant’s diagnosis, with 12 % reporting that the investigation led to their dependant’s diagnosis. The issue of disclosure was also identified in the qualitative analysis as an over-arching theme throughout survey responses (see Table 1, Section 4).

Adjustments and Provisions

Upon disclosing a diagnosis of ASD, individuals are entitled to receive mandatory support provisions and reasonable adjustments. They are, additionally, eligible for a range of discretionary provisions (throughout engagement with police and the broader CJS) to ensure their personal support needs are met. Adults and parents were questioned on whether they received such adjustments or provisions at police interview and, if so, how helpful these were. As illustrated in Table 3, adults and parents were largely dissatisfied by interviews being held at the police station and few were satisfied with the number of breaks provided during interviews. In contrast, they tended to be fairly satisfied with the support provided at interview in the form of AAs and Interview Supporters.Footnote 5 There were also many mixed responses (satisfied with some aspects of the adjustments and adaptations, but dissatisfied with other aspects).

Table 3 ASD community satisfaction with adjustments/adaptations provided
Full size table

Discussion

Little is known about how well equipped police officers consider themselves to be when dealing with individuals with ASD. Exploring the perspectives of almost 400 police officers, and 80 members of the ASD community, this study revealed discrepancies between their opinions and experiences: our police sample was fairly satisfied with how they worked with people with ASD while the ASD community sample was generally dissatisfied.

More specifically, police respondents were largely satisfied (42 %) with their dealings with the ASD community; with only 21 % reporting dissatisfaction (37 % gave a neutral response). Further, 47 % of police reported that they were ‘fairly knowledgeable’ about ASD. However, the possibility of a discrepancy between police self-reported competence and professional conduct in practice cannot be discounted. For instance, Dando et al. (2008) found that officers frequently reported using communication-enhancing techniques at interview (such as establishing rapport, providing full explanations of procedures and allowing an uninterrupted account), yet a national evaluation of investigative interview training found the overall standard of interviews to be poor (Clarke and Milne 2001).

The majority of parents (74 %) and autistic adults (69 %) were dissatisfied with their experiences, reporting discrimination (e.g., Table 1, Section 5), a lack of clarity and explanation (e.g., Table 1, Section 6), and feeling that their needs were not met (e.g., Table 1, Section 3). In particular, many ASD community respondents felt that an inappropriate physical environment (e.g., interview rooms, custody suites) coupled with a lack of appropriate support and explanation led to emotional stress, along with breakdowns in communication (Table 1, Sections 3.1–3.4). Moreover, both police respondents and autism community members expressed frustration over delays in the process, for example in waiting times to obtain an intermediary (Table 1, Section 7).

Despite this, it is important to stress that police officers did not consider themselves naive to the difficulties faced by individuals with ASD within police settings. As noted in Table 1, for example, many police officers acknowledged the importance of making adjustments to work fairly with individuals with ASD. Many also recognised the need for very careful planning and preparation (Table 1, Section 2) to be able to cater for the diverse abilities and disabilities associated with ASD; but this was often hampered by legislative requirements (e.g., time constraints, Table 1, Section 8) and the police environment (e.g., a lack of suitable facilities, see Table 1 Section 3.2). Police officers also reported frustration with a lack of training, information and organisational support (Table 1, Sections 1 and 8), thereby limiting their capability to respond effectively to individuals with ASD. Yet police officers generally felt they were doing the best they could in the face of these constraints.

The majority of police respondents reported using AAs and intermediaries, but the perceived helpfulness of these was mixed. Further, 61 % of police officers reported having little or no knowledge of the Witness Intermediary Scheme (Home Office 1999). It is worth noting, however, that, typically, only specialist police interviewers have access to Intermediaries, warranting some caution with these findings (as the majority of police respondents were frontline uniformed officers, with no responsibility for specialist interviewing). This suggests that some police respondents may have been interpreting the term ‘intermediary’ more broadly, as a third party who acts as a mediator between people. Indeed, confusion regarding terminology was an issue in this survey (see footnote 5 for further discussion).

It is also worth noting that, although there seems to be a general consensus amongst specialist interviewers that intermediaries are useful (notwithstanding delays in obtaining them), there is no published research on the impact of the presence of an intermediary (or indeed AAs or interview supporters) in ASD interviews. There is concern amongst police, however, over whether AAs always fully understand their role and whether they might, in some instances, be detrimental to an interview (e.g., by intervening inappropriately or even offering the suspect an alibi during interview) (Cummins 2011; Medford et al. 2003). Nevertheless, the mere presence of an AA has been shown to have some beneficial effects for vulnerable adult suspects, such as decreasing interrogative pressure in interviews, and encouraging the presence and better ASD informed involvement of legal representation (Medford et al. 2003). Moreover, many within the ASD community in the present study felt that the presence of a supportive third party was, or would have been, beneficial.

Encouragingly, police respondents reported implementing a wide range of adjustments (e.g., avoiding particular question types and allowing extra time for interviewees to process questions during interview) and many appeared astute to the challenges facing people with ASD (see Table 1). For example, many officers observed that the “tell me everything” ‘gold standard’ interview mnemonic for victims and witnesses (Milne and Bull 2002) is unhelpful and that more concise parameters need to be set. The location of the interview itself was also reported to be an area of concern for both police and ASD community respondents, and this was often linked to the sensory issues associated with ASD (American Psychiatric Association 2013). Interviewing the person outside the police station (e.g., in a Vulnerable Person Interviewing Suite, or in their own home) was reported to be a useful solution, although this was not always an option (e.g., for those who are suspected of committing crimes that necessitate a custodial environment; due to legal representatives’ reluctance to attend; or due to the possibility that a suspect with an ASD will not realise the seriousness of the interview). However, officers recognised the diversity across the Autism Spectrum and the importance of tailoring interactions to the individual, as far as possible (see Table 1, Sections 3 and 9).

It is also important to stress that police responsibility for adapting procedures and protocols on the basis of vulnerability arises either following disclosure of an individual’s diagnosis of ASD, or on suspicion that an individual may be vulnerable. Our survey highlighted that many members of the ASD community (adults more often than parents) chose not to disclose an ASD diagnosis to police.Footnote 6 This is a factor largely outside the police officer’s control, but appears to depend on how the individual perceives the police officer’s knowledge and understanding of ASD (see Table 1, Section 4). Many adult and parent respondents indicated that, in some cases, they chose to disclose the ASD diagnosis so that their vulnerabilities would be better understood and to ensure that their behaviours were not construed negatively (see also Brosnan and Mills 2015). Conversely, others chose not to disclose their diagnosis for fear of being negatively stereotyped. This fear would not be unfounded considering the widespread sensationalism and misconceptions around ASD (Huws and Jones 2011), and this is an important issue to address in police training.

Disclosure of ASD diagnosis is also important in cases of victimisation that are directly related to a person’s vulnerability due to their ASD, as reflected in the comments of our parent and adult respondents (see Table 1, Section 4). A recent report suggests that victimisation in individuals with ASD and intellectual disability is not always reported to the police (Beadle-Brown et al. 2014). For victimised individuals to report alleged crimes against them, it is imperative that their experiences with police are positive. Failure to disclose their diagnosis may hamper investigations of crime where an incident relates specifically to the person’s ASD (i.e., disability hate crime or targeting of their related vulnerability). Moreover, non-disclosure may result in victims and suspects not receiving the support to which they are entitled (e.g., an AA or intermediary). Our finding that many people with ASD are unwilling to identify themselves to police has implications for possible police reliance on ‘Autism Alert Cards’, which are offered by many ASD charities (e.g., the UK’s National Autistic Society) to individuals with ASD (to carry and use when engaged with the police and other CJS professionals).

Notwithstanding disclosure, inadequate police understanding of ASD is also likely to result in inadequate support. Despite these concerns, nearly all of the police respondents in our sample had some ASD awareness, and many believed themselves to be capable of making appropriate adjustments for individuals with ASD. There were also examples of positive outcomes from police, adults and parents (see Table 1, Section 3). Clearly, police training is crucial for ensuring that individuals with ASD are treated fairly (Adebowale 2013; Bradley 2009; Department of Health 2010, 2014; Equality Act 2010), as identified by both police and the ASD community respondents in this study (see Table 1, Section 1). A lack of understanding of ASD may lead to intentions and behaviours being misinterpreted. In this survey, only 37 % of officers had received police training on ASD, and over a quarter expressed dissatisfaction with this training (often in relation to its brevity and lack of relevance; or a lack of engagement and flexibility when delivered online). Police felt that several key barriers to effective management of ASD (such as failing to manage distress, sensory issues and communication needs) could be directly attributed to a lack of appropriate training. Our findings corroborate the Independent Commission on Mental Health and Policing (Adebowale 2013) report, which recognises a lack of ASD awareness amongst police and the need for sufficient information and training. These results also support the UK Government’s aim for the ASD community to be consulted when developing training programmes, particularly from those who have encountered the police and the wider CJS (Department of Health 2010, 2014). Indeed, there has been growing recognition and awareness of the importance of actively including such input as an integral aspect of ASD research and service development more generally (e.g. Milton 2014; Pellicano et al. 2014).

Overall, this study is a foundational step, intended to promote further investigation into the implications of ASD for policing; however, it is not without many of the limitations commonly associated with initial exploration. In order to reach as wide a sample as possible, participants were recruited online. This had three main drawbacks. First, participants were self-selecting; therefore it is likely that the sample may have included a disproportionate number of respondents from the ASD community who were unhappy with their experiences, as well as police respondents who already had some knowledge of, and possibly also personal interest in, ASD. This latter point may also play a part in the apparent discrepancy between the overall satisfaction ratings of the police and ASD community samples. Moreover, given potential discrepancies between police self-reported behaviours and actual practice (e.g., Clarke and Milne 2001; Dando et al. 2008) it would be useful to carry out direct observations of how police procedures (e.g., interviews) are conducted with individuals with ASD. Second, it was not possible to verify the diagnostic status of the ASD sample (although participants were asked to confirm that they had received a formal diagnosis). A third issue that is worth noting is our adult sample necessarily comprised verbally and intellectually able individuals, not representative of all ages and abilities of the wider autism population. Future research developments, with carefully selected cross-sectional samples and alternative methodologies (aside from online surveys) will be important. Yet it is important to stress the exploratory nature and uniqueness of the present survey.

In sum, there is an urgent need for the development of national evidence-based guidelines and training on ASD for police services and, indeed, the wider CJS. An evidence base is only still emerging, and there is a need for further research that can be used to inform the development of police strategies for effective engagement with the ASD community (and, more specifically, police protocols tailored to ASD). The present findings indicate effective and attainable support and adjustments during interview as a research priority. In addition, we echo previous calls for widespread police training on ASD (Adebowale 2013; Department of Health 2010, 2014). This too should be evidence-based, informed by those personally affected by ASD, and tailored to specific policing roles, to ensure its suitability for improving job performance across the wide range of police settings.