Exploration of a novel preventative policing approach in the United Kingdom to adverse childhood experiences

Adverse childhood experiences ( ACEs) have been associated with negative mental and physical health consequences. Neighbourhood police of ﬁ cers (NPOs) are thought to be well placed to identify and support children experiencing ACEs. Within this paper, we describe a qualitative exploration of an initiative deployed by a large UK police force which aimed to aid early identi ﬁ cation of young people with ACEs using police data and provide subsequent support given by NPOs. A thematic analysis was conducted using transcripts from three focus groups of NPOs, supplemented by questionnaires. Key themes identi ﬁ ed in both sets of data offer re ﬂ ections for policing and other public service provision seeking to target on the basis of ACEs, including: limitations in the usefulness of police data to identify childhood adversity; challenges using the ACEs framework as a means to target such ‘ early intervention ’ ; and debates regarding the appropriate role of the police in supporting young people in adversity. Prior to replication of this intervention elsewhere, there are several points to consider: whether police data alone are suf ﬁ cient in identifying ACEs; ensuring clarity regarding the role of the public sector agencies in such intervention; and determining an apt measure of success for the scheme.

Background R esearchers, policymakers and service providers are increasingly interested in the impact of adversity during early life experiences on future life outcomes, and the potential of 'early intervention' to prevent such outcomes (Bellis et al., 2014;Couper and Mackie, 2016;Public Health Wales, 2015;UCL Institute of Health Equity, 2015). A dominant framework for understanding this relationship is that derived from the Adverse Childhood Experiences (ACEs) study, undertaken by Felitti et al. (1998). The ACEs study originally identified seven areas relating to direct abuse or household dysfunction before the researchers added in further questions on neglect and trauma to describe ten ACEs in total (Stevens, 2012). The initial ACEs study identified that experiencing one or more of these ACEs could lead to negative long-term health and social outcomes (Felitti et al., 1998). A recent systematic review has identified that cumulative exposure to four or more of these ACEs increased the risk of developing a wide variety of negative mental and physical outcomes .
It is theorised that experiences of trauma in childhood have an impact on social, emotional or cognitive functioning, which in turn may alter health behaviours and lifestyle factors, potentially leading to an increased risk of health or social problems (Bremner, 2003;Kendall-Tackett, 2002;McEwen, 2007). As the individual and the public health burden appear vast, it is important to develop and assess methods of identification of adversity and of assuring timely support.
The contribution of law enforcement agencies to this endeavour is increasingly recognised. For example, research in the USA suggests that police officers are responsible for managing a greater proportion of reports of maltreatment (18.1%) than any other professional group, including educational (17.7%), social services (11.0%) and medical personnel (9.2%) (US Department of Health and Human Services, 2014). This is also reflected in policy discourse. In the UK, Her Majesty's Inspectorate of Constabulary (HMIC) recently stated that 'protecting children is one of the most important tasks the police undertake' (HMIC, 2015, p. 4). This is echoed in UK legislation. Under the Children Act 1989 (section 47), alongside other statutory services, the police have a responsibility to make enquiries to safeguard and secure the welfare of any child within their area who is suffering (or likely to suffer) significant harm. However, in assessing responses to child protection cases among UK police forces, the HMIC identified that, across the eight forces they investigated, 38 per cent of cases involving child protection issues were handled 'inadequately' (HMIC, 2015, p. 43). Weaknesses included: variability among forces in handling cases; concerns with officers' initial responses; ineffective detection; and poor sharing of information with other key partner agencies. The HMIC therefore suggested a need for significant improvement in how we identify and care for children at risk of abuse.
In order to address the recommendations made by the HMIC report, several UK police forces have trialled new initiatives to improve their recording, processing and supporting of children at risk. This includes a large police force in the UK which has employed an approach adapted from the ACEs study. A predictive modelling system, known as the Tool for Intervention and Prevention Triggers (TIPT) (Hughes and Chandan, 2017), searches police 'Cumulative exposure to four or more [Adverse Childhood Experiences] increased the risk of developing a wide variety of negative mental and physical outcomes' 'A predictive modelling system [TIPT]… searches police databases to identify families and children who have multiple ACEs present' databases to identify families and children who have multiple ACEs present. The ACEs were identified solely from police databases and these included: abuse against the childphysical, psychological and sexual; neglect of the childphysical and emotional; and markers of household dysfunctionsubstance misuse, mental ill health, domestic abuse against a mother, lone parents as well as parental incarceration. Children identified as experiencing four or more ACEs are assigned to their local Neighbourhood Policing Teams (NPTs); a UK policing initiative whereby small teams of typically 10-15 police officers are assigned to a particular geographical area, with the aim of making policing visible and improving interactions between residents and police officers (Turley et al., 2012). On receipt of a 'TIPT package', Neighbourhood Police Officers (NPOs) are required to investigate the case, identify existing support, and, where appropriate, consider referral for additional support services from other local agencies. The TIPT package consists of an electronic briefing relating to information compiled by an intelligence officer once the system informs them of a child who has met the threshold of four ACEs (which can include four instances of the same ACE). The briefing consists of information relating to: demography; status (victim or perpetrator in most recent report); ACEs present and the crime reports they were derived from; vulnerabilities in the household; and sibling, parent or carer details. The briefing is at the discretion of the intelligence officer (who creates the intelligence package) and often only focuses on vulnerability in the household.
The senior police officer originally responsible for the initiative placed it in the context of a core commitment to 'early intervention', defined as an intent to 'stop problems escalating, reducing demand for police resources to manage that individual further down the line', with the overall aim of 'creating fewer victims and achieving a better quality of life for local people' (Madill, 2017). The intention is therefore that 'TIPT systemises the flagging process so [that] those in need of help are easier to pinpoint early before those adverse experiences manifest as poor outcomes later in life' (Madill, 2017). This implies a broad focus on preventing the full range of negative future outcomes, though with obvious emphasis on core police concerns of criminal offending and victimisation.
This represents an innovative approach reflected in terms of: (1) the extended role of the police to address early intervention; (2) the use of police data to inform this activity; and (3) the potential to share unique insights from this data with partner agencies. Such innovation can lead to challenges in initial implementation, which, if described, can be of great use to other police forces. The present study, therefore, provides a critical analysis of the early implementation of the scheme though qualitative research regarding the views and experiences of those NPOs who are delivering it.

Methods
The objectives of this study suited a pragmatic approach, which was the underpinning methodology adopted for the subsequent research described (Creswell, 2013).
Data collection involved two distinct approaches: focus groups and questionnaires. Three focus groups were conducted over a one-month period 'Provides a critical analysis of the early implementation of the [TIPT] scheme though qualitative research' (February 2017) in three different police districts in the region of the UK served by the police force being examined. Each focus group consisting of NPOs included purposive representation from neighbouring districts, ensuring full geographical coverage of the region. To identify an appropriate range of officers who were available for discussion, without impacting upon core staffing and service delivery, a convenience sampling method was used. In total, 35 NPOs of varying rank and length of service participated in the three focus groups. Successful recruitment of NPOs for the focus groups was supported by the neighbourhood policing senior leadership team (SLT). Verbal and email invitations were sent from the SLT to individual NPOs requesting their participation in the focus groups.
The focus groups were conducted by the second author using a semi-structured topic guide. The topic guide was developed by the principal researchers (first and second authors) based on initial discussions with senior officers, review of previous literature, and the first author's experience working in a UK police force as a volunteer Special Constable. The topic guide consisted of broad themes relating to NPOs' knowledge of ACEs and TIPT, confidence in delivering the scheme, working relationship with partner agencies in their neighbourhoods, interventions and referrals, and future outlook of the initiative. The focus groups each lasted approximately one hour and were recorded digitally.
Subsequent to the focus groups, all neighbourhood teams comprising NPOs were invited to complete an online questionnaire. This questionnaire was electronic and delivered using Google Forms, a software provider accessible on all police computers. The questionnaire was sent to a designated senior NPO in each team by the overarching SLT, and either that officer or a member of their team were responsible for completion. Mirroring the key topics of the focus group, the questionnaires included a range of open-ended questions, which were intended to achieve a better understanding of the variation in practice and in context across districts and neighbourhoods, and, in doing so, address the limitations of group discussions in which the complex detail of varied practice is not readily obtained. The survey focused on several key domains which included: how they identify young people at risk currently; how they gather information on these individuals; current processes for referral and interventions; outcomes they can measure; and good practice. All neighbourhood teams were encouraged to respond. The survey was anonymous, unless the respondent chose to name their local area in their responses. A total of 20 responses was received, representing 23 per cent of neighbourhood teams. Due to the nature of the questions asked about local practice within the questionnaire, we anticipated that there was little additional value in requesting all officers within one region to fill out the form as guidelines were likely to be the same per neighbourhood team. Therefore, one response was requested per team.
Ethics Approval was received by the University of Birmingham ethical review committee (Reference number: ERN_15-1598) prior to undertaking this study. Both written and verbal informed consent were obtained from the focus group participants using consent forms and information sheets. Participants were reminded prior to the focus groups that they were under no obligation to participate in discussions. Individuals were provided with the researchers' contact details should they have wished to make any amendments 'In total, 35 NPOs of varying rank and length of service participated in the three focus groups' 'All neighbourhood teams comprising NPOs were invited to complete an online questionnaire' to their comments. No participants contacted the authors following the study. Similarly, completion of the questionnaire was voluntary, as explained to potential participants in an introductory email, and in an information sheet that acted as a cover page for the questionnaire. In both instances, all contributions were kept anonymous and confidential, with no names or areas being provided in any dissemination or discussion with senior staff. All participants were told of their right to withdraw from the study up to two weeks after completing the focus group or the questionnaire. No such requests were received. No incentives were offered to the participants of the focus groups or questionnaires.
Verbatim transcripts from the focus groups and the text from the questionnaires were analysed using a thematic approach (Pope et al., 2000). The approach consisted of the following five steps: (1) familiarisation with the raw data; (2) identifying a thematic framework; (3) indexing the transcripts according to the ideas presented; (4) charting the data with the assistance of NVivo 11.4 software; and (5) mapping and interpreting the data. Both the first and second author independently conducted initial indexing and charting of the data before a final interpretation was agreed upon. The first author led on analysis of the focus group data, and second author on the questionnaire data. Codes were then combined, and themes jointly agreed. In the following discussion, key themes are presented, with illustrative quotes where appropriate.

Results
Five main themes were identified: (1) the competing and contradictory reflections on the aims and principles of the scheme; (2) barriers to current use; (3) working with partners; (4) ongoing training and support needs; and (5) challenges in measuring success. These themes are discussed in turn.

Theme One: Reflections on Aims and Principles
Discussions in each of the focus groups reflected widespread support for the ideas inherent to the scheme, both personally and among their colleagues. Participants described their positive perceptions of the aims of the scheme in varied, but related ways.
The initiative was seen as leading to a focus on the wider impact of crime within the family, beyond the individual recorded as the victim: 'Whereas before, we might have just only looked at that one child or that one person in the family and not considered the impact on the other siblings' (FG1). Similarly, the initiative was repeatedly presented as broadening the focus of police officers to the child of an offender: 'It raises people's awareness of the child that is often ignored or overlooked and when we're going to arrest dad or we're dealing with a domestic incident, that they're affected by what's going on around them. I don't think anyone ever ignored that, but it didn't really factor into our world that much because we were dealing with the incident at hand and not, perhaps, thinking about the wider implications of it.' (FG3) This brings a step change in potential support, given that previously 'The one particular member of the family [who experienced domestic violence or 'Five main themes were identified' 'The initiative was repeatedly presented as broadening the focus of police officers to the child of an offender' abuse] has got a lot of support but actually, the other siblings… are not even thought about' (FG1). In contrast, there is now recognition that potentially 'that child is going to be the next victim here, so we need to start focusing on early intervention around that child' (FG3). As such, the initiative is recognised as 'a future investment' (FG3) for the force.
The important additional contribution of the police in ensuring support for young people in adversity was recognised. For example, one questionnaire respondent highlighted that, without such an initiative: 'Schools are missing out on a lot of information that the police have access tothis information could be used to help support the child. Whether its drugs, disorder, DV [domestic violence] or any other criminality at an addressit all impacts on their school work and life in general. ' (Q12) The particular role of the police, and more specifically of NPOs, in enabling support to young people experiencing ACEs was therefore broadly accepted. There was less agreement, however, regarding the precise role that the police should be playing in the support of these young people. In each of the three focus groups, concerns were raised by several NPOs as to whether the role that they were being asked to play was appropriate to the remit of a police officer: 'This is social work and I'm surprised that it's the Police that's taking it up' (FG2). Concerns were also raised about the barriers NPOs faced using the tool.

Theme Two: Barriers to the Use of TIPT
The positive support for the aims of the initiative to assist young people experiencing ACEs was frequently countered by concerns regarding current practice in relation to the effective identification of young people undergoing ACEs. The perceived barriers to the effective use of the TIPT approach include concerns with: the quantity of packages [number of referrals from the TIPT system] received being unreflective of known levels of need; frequent targeting of young people already known to local services; and the inherent limitations in only using police data.
Some NPOs worked in very deprived areas, and therefore expected large numbers of packages; however, this was not always realised, and officers were unclear as to why. While it is understood that a prospective system of identification would not immediately identify a large number of young people, the lack of cases to date appears to be leading to a lack of confidence in the efficacy of the initiative.
For other respondents, the concern was with the relevance of the young people identified through TIPT. Several NPOs report that the majority of packages relate to young people who are already well known to local services, including many already in the child protection system: 'The majority of cases are either: late teens; got a social care intervention, got health problems and they've got all sorts of things' (FG2). The police data are therefore seen as identifying children far later than other professionals who have existing relationships with the child or family.
Questionnaire respondents highlighted a range of other ways in which they felt that the targeting of young people based purely on police data was flawed. One NPO suggested that the young people identified to date are generally 'Positive support… was frequently countered by concerns regarding current practice in relation to the effective identification of young people undergoing ACEs' 'persistent missing persons' (Q10), which might therefore serve as a more effective indicator of adversity. Another respondent listed known ACEs that were not readily being highlighted, despite connections to criminal behaviour, including 'children whose parents are drug abusers' (Q3) and 'children whose parents have been in prison' (Q3). Concerns were also raised about the lack of cultural sensitivity of the TIPT approach. In particular, concern was raised that young people experiencing ACEs within Muslim communities may not be so readily identified by current approaches. This was due to a belief that these communities 'tend to under report' (Q7) certain forms of crimes, including domestic violence, and are less affected by more visible ACEs related to drug and alcohol misuse.
Concerns with the limitations of the current TIPT approach were amplified by recognition that efforts in implementing the initiative were at the expense of alternative methods of identification of children and young people in need of support. In some cases, this was presented as countering established good practice. Such comments were particularly raised by NPOs who discussed previous modes of working that they felt had been more effective than the current approach. Several NPOs suggested that schools, youth clubs and children's homes can often work well in partnership supporting the project as they can provide valuable insights into the lives of young people. They often identify difficulties before they result in police activity, including a broader range of ACEs than those reflected in police data. In some areas, informal partnerships have emerged; for example, in one neighbourhood, officers had asked each school 'for three names' (FG2) to target for support.
NPOs also highlighted their own tacit knowledge about families in their neighbourhoods, including through home visits that illuminate a range of experiences of adversity that are not currently captured by the TIPT approach: 'When you walk into a house and you look round and think, "That nappy hasn't been changed…" [and] mum and dad, I happen to know, are both Class A users' (FG1). However, officers also recognised that this knowledge was not currently captured in police data in a way that could inform a robust targeting system: 'A Neighbourhood Officer would have all the information in their head. It never gets put into a system, so then when somebody else comes to look and says, "What do you know about this family?" the systems won't tell you. ' (FG3) In parallel, NPOs felt that there were competing priorities being requested by their supervising officers, directing attention away from this activity. This was seen to be highly detrimental to this type of work due to the rapport required with partners, and therefore the importance of the continued involvement of specific personnel. This context of reduced resources and additional service pressures led to questions as to whether this scheme represented 'core business' (FG2) for the police which requires focus on key priorities (determined by the public and locally elected officials). In particular, it was recognised that, if there was a particular rise in crime in a local area, the preventative activity 'goes off your radar and it's just disappeared' (FG3) potentially resulting in a loss of support to a child or family. Unless support for young people in adversity is established as core business for the police force, the concern is that it will be side-lined and eventually discontinued. Hence, in an environment where the 'Efforts in implementing the initiative were at the expense of alternative methods of identification of children and young people in need of support' question of the role of the police is questioned, it is important that their working relationship with partners is explored.

Theme Three: Working with Partners
Reaffirming policing priorities and 'core business' (FG2) also led NPOs to question whether they were typically best placed to engage with children and families given the possible negative public perception towards the police. As one focus group participant explained: 'Although we've seen the changing face in policing, they haven't. As far as a lot of these families are concerned, we're there to lock them up' (FG1). While NPOs are tasked with developing strong relationships in their local areas, it was recognised that members of the public do not necessarily distinguish the local officer from colleagues in other parts of the Force: This can lead to blurring of the roles of various teams within the police, and confusion for families that requires careful explanation. Similarly, it is recognised that the young person may be resistant to engaging with an officer due to fear of peer perceptions: 'When you turn up to one of these vulnerable people in uniform, their guard is up straightaway, especially if their friends are outside. You're just down as a grass then, regardless of whether the police are there to help you…' (FG1) Reflecting this concern, there was a strong and repeated narrative that police involvement should be necessarily, and appropriately, restricted to identifying need, with interventions widely seen as best delivered elsewhere; this is discussed further in the following theme.
It was evident that, at present, there is no clear and consistent approach in place across the Force regarding how to act upon a TIPT package referred to a neighbourhood team. This was recognised as indicative of local variation in partnership arrangements, and, in particular, the availability of agencies and schemes to which to refer young people. Whilst some areas appear well served by referral processes, elsewhere officers suggested that there are no clear routes, meaning referrals were being made for each case in a bespoke manner, with officers necessarily developing relationships as needed.
In areas lacking established and robust means of referral, some NPOs were frank in their assessments of the efficacy of their decision-making deeming it often as inadequate. For example, one questionnaire respondent presented his approach as 'shocking' (Q2): 'I just decide. I haven't been trained. There is no external scrutiny of my decision. What if I make the wrong decision? We are talking about the daily life of a child… I am not a social worker. Police deal with crime and ASB [antisocial behaviour]. The specific interventions required to support the daily life of a child who is vulnerable for any reason should be determined by properly qualified professionals. We would not ask a social worker to police a demonstration, but police can act as semi-social workers.' (Q2) 'NPOs… question[ed] whether they were typically best placed to engage with children and families given the possible negative public perception towards the police' 'There was a strong and repeated narrative that police involvement should be necessarily, and appropriately, restricted to identifying need'

Theme Four: Ongoing Training Needs
This concern with a lack of professional expertise and experience is echoed elsewhere: 'I am not trained or expert in dealing with ACEbut me and my team are expected to make meaningful interventions into the lives of broken children and save them from themselves sometimes.' (Q2) NPOs underwent a two-day training course in relation to the role, with sessions regarding ACEs and prevention of harm. NPOs' views on the delivery of training were mixed, with concerns frequently voiced that the training was too theoretical or abstract and therefore not appropriate to their practical needs. NPOs were concerned that the training did not clearly promote 'a consistent approach' (FG2) as to how the TIPT initiative should be delivered, which in turn could lead to different practices across the force. This was echoed by colleagues in other focus groups who were clear about the limitations in their skill set, relative to other professionals: 'I went through an 18-week training course, when I joined this job… With these referrals and the social services side of the job, I've had no training. ' (FG1) Indeed, in some instances concerns of negative impact were voiced: 'An adverse effect occurs when we send people into situations who aren't qualified, and I include myself when I'm saying that. Part of the art of safeguarding is knowing when you don't know…' (FG3)

Theme Five: Measuring Success
NPOs repeatedly highlighted the lack of data being collected by which to measure the success of their interventions, despite institutional recognition that this was important. It was widely recognised that it is very hard to measure success in a preventative intervention such as this. This was well articulated by one focus group participant: 'So you're working with little Johnny, who's eight or ten years old, and you work with him for two or three years and you're doing really well with him; you're helping him and he's doing this and he's growing and then… little Johnny starts getting involved in something and goes down the wrong path. Does that count your four years as a failure? Ultimately, is that the way we're going to look at the job because you've put your time and money into little Johnny? In four years down the line, he's going to then become arrested for something and get locked up. Is that how you judge it? Is that a failure then? ' (FG2) This was of particular concern because of the range of complex contextual factors that might limit the long-term effectiveness of such activity but are beyond the influence of the police, including pressures that appear in adolescence, such as peer pressure, gang involvement, educational disengagement and unemployment. Even where some measure of success might be apparent, NPOs also recognised the challenge of attribution: 'If we say, in 10 years' time, that 99 out of 100 haven't fallen in 'Training was too theoretical or abstract and therefore not appropriate to [NPOs'] practical needs' 'NPOs repeatedly highlighted the lack of data being collected by which to measure the success of their interventions' the CJ [Criminal Justice] system, can we prove it's been because of this?' (FG1).
While recognising the challenges in determining appropriate outcome measures, a lack of such indicators was argued as likely to put the initiative at risk of being discontinued, with senior managers asking: 'How am I justified giving you more resources for your department when I'm judged on these performance figures [i.e. key performance indicators for the Force]?' (FG2). These concerns imply a need for short-and medium-term indicators of realistic progress that might serve to illustrate impact in a relevant timescale, and in a robust and measurable manner.

Discussion
The TIPT initiative represents an innovative policing practice based upon the application of the increasingly prominent ACEs framework to inform targeting early interventions at young people in adversity. An exploration of its implementation offers an opportunity to reflect on a number of key themes in current policing and broader public service provision, including: limitations in the usefulness of police data to identify childhood adversity; challenges in the use of the ACEs framework as a means to target such 'early intervention'; and debates regarding the appropriate role of the police in supporting young people in adversity.
In targeting on the basis of administrative police data, a narrow criminal justice framework is applied to understand young people's experiences of adversity. Police data offers insight into a limited range of ACEs, including domestic violence, abuse or neglect, or where a parent has been incarcerated. Furthermore, experiences are only identified when a crime is reported, or there is a request for a police response, suggesting partial understanding of such experiences, affected by a multitude of social and cultural factors that can influence whether crime is reported (Black, 1976), including local concerns about relationships with minority communities. In most cases, then, a young person is being identified for support only when their family has come into formal contact with the police. The identification of adversity is dependent on the criminalisation of a member of the familytypically the father. The barriers this may pose to engagement with support provided by the police were noted by several research participants. The conflation of adversity with criminalisation may mean that the reason for engagement is not understood by the family or young person. Thus, there is a need to understand the perspectives of young people engaged by the initiative in this respect.
The implementation of this initiative also offers useful insight into the operationalisation of the ACEs framework in targeting early intervention. This police force employs an additive model of ACEs: where four or more experiences of ACEs are apparent, intervention is considered. This is informed by an interpretation of the epidemiological evidence that suggests the relative risk of negative outcomes to be particularly heightened when four or more ACEs are observed .
Concerns regarding the application of population-level epidemiological research findings to decisions regarding service provision to individuals are well-rehearsed. These include the need for caution in drawing inferences about 'In targeting on the basis of administrative police data, a narrow criminal justice framework is applied to understand young people's experiences of adversity' individuals from characteristics of social groups to which individuals belong. In this context, a simple counting of ACEs is seen to offer little insight into the qualitative experience of trauma and adversity for the young person, including the severity or longevity of exposurefor example, being unable to distinguish between 'one-off' events and prolonged experiencesor the individual and family factors that might offer protection from or resilience to these experiences (Bellis et al., 2018;Bridgewater, 2018;Edwards et al., 2018;Kinner and Borschmann, 2017). In addition, the areas where there is limited information in police systems, such as lone parents or socio-economic status, their effects may be demonstrative of a multiplicative effect on negative outcomes rather than just purely additive.
In contrast to the underpinning research (Felitti et al., 1998), this police force has chosen to allow the same specific ACE to be counted multiple times, rather than requiring four distinct experiences. This goes some way to address concerns regarding the limitations of the ACEs framework in understanding the duration of a particular experience, though this decouples the initiative from its purported evidence base. Furthermore, it still serves to simplify the experience into a dichotomous variable, with no automatic consideration to the qualitative complexity of experience, including the resilience and resources within a family to respond to such adversity.
In recognition of the limitations in the reliance on police data and the ACEs framework, NPOs were encouraged to investigate a family identified by the TIPT process, including by engaging other statutory services who could provide a more holistic understanding of the needs of the young person, potentially including the identification of other ACEs. This process was not formalised or prescribed. The lack of a framework for assessing a young person beyond the identification of ACEs places significant emphasis on the ability of NPOs to understand, interpret and respond to complex situations of adversity within often very challenging family contexts. It also clearly results in inconsistent levels of engagement with and participation of the young person and their family in considering the appropriate response and support that is needed.
Despite a general agreement and shared understanding among those implementing the scheme of the overall purpose in offering early support to young people in adversity, there was a lack of clarity as to its specific aims, particularly with regard to the scheme's definition of success. Most commonly, the scheme is presented (implicitly or explicitly) as intended to prevent future criminality among young people who have been identified as at higher risk due to adversity in childhood. This reflects the primary focus and professional discourse of serving police officers regarding crime reduction but represents a narrow view of the evidence base regarding long-term outcomes following early adversity. These aims do not encapsulate the opportunities the scheme provides in terms of addressing the parallel risk of victimisation to crime and the broader public services agenda highlighted by addressing ACEs to improve educational disengagement, and poor physical or mental health.
The lack of clarity or shared understanding regarding the aims of the scheme calls into question the appropriateness of the subsequent intervention that is therefore put into place. Put simply, an intervention intended to prevent criminality might look very different to one intended to prevent victimisation. However, the rationale for specific approaches to intervention were typically poorly described and determined by the individual's or team's thinking and 'NPOs were encouraged to investigate a family identified by the TIPT process, including by engaging other statutory services' 'An intervention intended to prevent criminality might look very different to one intended to prevent victimisation' professional approach, rather than any known evidence base. Whilst initial identification was determined through a consistent approach across the police force area (even if open to dispute), the subsequent determination of intervention appeared to vary greatly. The scheme is predicated on a generalised model of elevated risk of negative outcomes, rather than an understanding of the specific mechanisms by which specific experiences of adversity might influence future behaviour, including expressions of criminality, or indeed other negative outcomes to be prevented (Corcoran and McNulty, 2018;Kinner and Borschmann, 2017;Taylor-Robinson et al., 2018).
The use of the ACEs framework, and the evidence base that underpins it, is therefore of limited value to NPOs in determining intervention. Indeed, as has been highlighted elsewhere (Early Intervention Foundation 2016), the evidence base for such early intervention remains under-developed. In this context, there could be legitimate concerns about the negative impact of formal criminal justice-related responses due to the labelling of young people who have experienced ACEs as at risk of criminality. In particular, given long-established understandings of the potential for the self-fulfilling prophecy of perceived labels of risk regarding criminality (Becker, 1963), recognition is needed as to the negative potential of well-meaning early intervention, particularly from within a criminal justice paradigm.
Indeed, there remains significant debate among NPOs as to the appropriate role of the police in support of young people with ACEs. This includes a commonly expressed view that it is infeasible for sufficient training to be provided for police officers to offer appropriate support, given the complexity and extent of training of social workers and allied professionals, and that the role of the police should only ever be restricted to identification and referral to other agencies or to existing multi-agency partnerships or forums for coordinating 'early help'. This argument included recognition that police data may offer additional insight to that of other services, for example, in identifying young people experiencing domestic violence. However, there was also recognition that young people identified by TIPT thus far are already well known to local services, with police data, therefore, providing little additional value to date, and the existence of four ACEs in such data ill-serving as an appropriate indicator for 'early intervention'. There is a clear need to align the aims of the multi-agency working approach across the whole region to prevent duplication and to also address many of the barriers mentioned. There are several suitable recommendations that can be made as a result of the exploration of this intervention: 1 The benefits of data sharing -It is clear from this exercise that police data have inherent limitations in identifying markers of early adversity. However, if a data sharing approach was used and ACEs were introduced into the algorithm from other public sector data sources, it is likely that at-risk children would be identified at a much earlier stage to allow for early intervention. Although a public sector data sharing approach is an important step, this study also highlights the need for a wider collaborative effort amongst the community to address ACEs. The role of the general public within this should not be underestimated. Efforts taken to improve the knowledge and understanding of ACEs amongst the general public and public sector services will not only trigger opportunities for intervention through earlier identification of negative exposure and consequences, but also be key to improving data recording and longitudinal follow up.
'The use of the ACEs framework… is therefore of limited value to NPOs in determining intervention' 2 The best person for the role -Once the at-risk children are identified, it is possible that NPOs may not be the most appropriate individuals to conduct the follow-up visits. Instead, it may be better placed that the role of the NPO is to share their rich local knowledge and provide a supporting role to another public sector service which may be better placed to inform or deliver preventative advice to the child or family. Alternatively, there may be a benefit in refocussing the aims of the TIPT programme for NPOs. Rather than the emphasis on NPOs being in the role to identify ACEs, perhaps just a greater understanding of ACEs amongst NPOs as a result of the TIPT tool may provide useful insight into the reasons for inappropriate or anti-social behaviour demonstrated by adults. This may trigger consideration by NPOs into alternative positive police action aside from the criminal justice system. 3 Improved communication -This study also highlighted the duplication of work across public sector services. This will likely only reduce if there is clear communication between the services to ensure that a plan is in place for each family, with the roles of each multi-agency organisation clearly defined.
This type of approach to studying this intervention had inherent strengths and limitations. One possible strength was that the primary author works in the police force in a voluntary capacity and, as a result of being part of the social group being studied, he was able to share many of the experiences, assumptions and beliefs of the group to improve the analytical understanding. However, there is a chance that this may have imported an interviewer bias on his analysis, therefore there was an independent analysis conducted by the second author who agreed on the final framework. In addition, another strength was that participants chose to take part from a variety of stations across the force area which added to the strength of the study as NPOs could give views from working in different socio-economic settings. The authors believe that, although further interviews could have been conducted, theoretical saturation was likely reached, as with the sample focus groups, as no new ideas had emerged. Although a limitation of the study is that, as a small qualitative piece of work, it is only able to provide an exploration of the views surrounding the topic rather than generalisable results, it provides insight into whether an approach like this could be adapted for implementation elsewhere and the barriers to be considered when doing so.

Conclusion
Our qualitative exploration of an implemented local policing initiative to support young people in adversity highlights a range of issues that warrant further reflection to improve future delivery of such a service. These were concerns as to whether police data alone are sufficient for identifying adversity, and if the use of four experiences of ACEs is an appropriate framework for identifying cases.
The limitations of the ACEs framework also provoke concerns as to whether it offers the most appropriate method of identifying children at risk, or whether alternative sources of information should be routinely drawn upon, including NPOs' tacit knowledge of their community and the families within it, or other proxy indicators of adversity readily available to police but outside of the ACEs framework, such as reports of a child running away from home.
More fundamentally though, the views of officers implementing the scheme provoke concern regarding intervention. The ACEs framework is useful in 'Once the at-risk children are identified, it is possible that NPOs may not be the most appropriate individuals to conduct the followup visits' sensitising police officers to the potential of childhood adversity in shaping current and future experiences of criminality and victimisation, and an increased awareness of adversity among NPOs is to be commended. However, adversity should not be seen as deterministic of future criminality or victimisation, or as a reason to draw young people into a criminal justice system or paradigm. Careful attention is needed as to how to determine whether a young person would benefit from support, and what sort of support is appropriate. The well-intentioned public service intent to support young people in adversity must therefore be advanced with caution.