1. As explained in Part A, in order to facilitate the detailed investigation and assessment of the themes for this investigation, the Inquiry selected six local authority areas as case study areas:
In addition, the Inquiry also reviewed a sample of children’s cases from each of the six areas for closer analysis.
2. A detailed thematic analysis of the evidence is set out in Parts D to L of this report. This annex sets out the Inquiry’s conclusions about the key statutory agencies in each case study area with respect to the eight themes in this investigation:
3. Durham Constabulary’s most recent problem profile for child sexual exploitation from 2019 did not provide an adequate assessment of the nature and scale of child sexual exploitation in the area or allow for analysis of the issue. Crimes “were classed as CSA rather than E” where there was no “explicit mention of grooming or exploitative techniques was mentioned”.[1] The profile also only considered data over a period of three months, which cannot be “comprehensive” as suggested.[2] As a result, as accepted by the police and local safeguarding partnership, the extent of child sexual exploitation in Durham was likely to be underestimated.[3]
4. No evidence was provided about the number of cases of child sexual exploitation by networks using the Inquiry’s definition, although such cases plainly existed.[4] Other definitions were used, but these appeared to further limit the type of group of abusers being considered by Durham Constabulary.[5]
5. Ethnicity was not recorded for suspects in 35 percent of 1,900 cases of child sexual abuse and exploitation, and for 14 percent of 1,138 victims in Durham in 2018/19.[6]
6. There were several examples of insufficient efforts to use disruption tactics to protect children. Child Abduction Warning Notices (CAWNs) were used proactively in Durham but were not visible on police systems. In 2020, Her Majesty’s Inspectorate of Constabulary and Fire & Rescue Services (HMICFRS) identified limited evidence of disruption activity at hotspot locations or in relation to known perpetrators. [7]
7. Durham Constabulary’s child sexual exploitation profile and the Council’s sexual exploitation strategy used inappropriate language suggesting that children had consented to sexual acts. There were also examples of victim-blaming language but limited evidence of it being challenged.[8]
8. Although there was a large increase in the number of recent assessments in which child sexual exploitation was identified as a factor, the out-dated checklist approach taken by Durham County Council to assessing the risk of child sexual exploitation was concerning.[9] Indicators of risk were conflated with indicators of actual harm, and the Inquiry was told that children would not be classified as high risk where there was no identified perpetrator.[10] While the Inquiry saw no evidence of this approach being put into practice, such an approach is wholly unacceptable.
9. Risk assessments relating to several of the case study children in Durham were of poor quality.[11] Only cases considered high risk were monitored via the Child Exploitation Vulnerability Tracker (CEVT), leading to a reduction in the number of cases being monitored.[12]
10. The level of intervention, protection given and the outcomes for some children were very poor. There was evidence that children were prematurely removed from the list of those discussed at multi-agency meetings when they were experiencing harm.[13]
11. There are clear areas for improvement in both the assessment forms and processes for monitoring risk in County Durham.
12. Durham’s performance in relation to missing children was mixed. Durham Constabulary’s Philomena Protocol, aimed at coordinating responses to young people in children’s homes going missing, was commended by the National Police Chiefs’ Council as an example of good practice and is being adopted by other forces. [14]However, in 2020, HMICFRS recommended that Durham Constabulary “immediately improves practice in cases of children who go missing from home”.[15]
13. Durham County Council had a lower than average percentage of children in care placed out of area. However, it was not able to provide the Inquiry with the number of sexually exploited children placed outside the area, and seven children at risk of sexual exploitation were placed in unregulated placements outside Durham in the two years to March 2019.[16]
14. A number of boys were discussed at an operational child exploitation group. No specific child sexual exploitation services were available for boys and young men, but there was timely support for one of the male case study children in line with his and his parents’ wishes.[17]
15. It is “fairly early days” in terms of the approach to children with disabilities in Durham, with its assessment form only recently including “specific recognition around any disability or additional vulnerability”.[18] The Council has also only relatively recently included disability as a specific factor on its CEVT. [19]
16. There was evidence of strong partnership working on child sexual exploitation in Durham, particularly following the formation of the police-led ERASE (Educate and Raise Awareness of Sexual Exploitation) team. While there was variability, there was also positive involvement of schools, health services and other agencies with some case study children in Durham.
17. There has been a significant amount of internal service audit activity on child sexual exploitation and missing children within Durham but this has not so far led to significant practice improvement by Durham Constabulary, as shown by the 2020 HMICFRS report.[20]
18. In the South Wales Police area, if the ‘exchange’ between the victim and perpetrator was not “abundantly clear from the outset”, cases were logged as child sexual abuse rather than exploitation.[21] This is not consistent with the intention of the Welsh guidance that exchange denotes the exploitation of a need.[22]
19. Neither South Wales Police nor the City and County of Swansea Council (Swansea Council) knew accurately how many children were being or were at risk of being harmed through sexual exploitation. Due to inconsistencies in reporting and use of the flag system, South Wales Police’s 2019 problem profile recognised that there may have been “significantly more occurrences that involve sexually exploiting a child than the profile reflected”.[23] While Swansea Council accepted that there was a “sharp reduction” in child sexual exploitation cases between 2018 and 2019, it was also “unable to give a clear explanation of this figure”.[24]
20. Using the Inquiry’s definition, there were examples of child sexual exploitation by networks over this period.[25] Such instances of exploitation by networks or groups should have been identified by the police and the local authority. South Wales Police incorrectly told us that there were no data to suggest sexual exploitation by organised networks, whether according to the Inquiry’s definition or at all.[26]
21. Across the South Wales Police area, including Swansea, less than half of the 56 incidents of child sexual exploitation in 2018/19 recorded the victim’s ethnicity.[27]
22. There were examples of successful disruption in Swansea. South Wales Police placed a child in police protection, disrupted a specific address on the premise of antisocial behaviour, and encouraged inspections of another location about which concerns were raised.[28] Swansea Council requested a number of CAWNs, used passport markers, engaged with the National Referral Mechanism, and communicated with housing services where there were properties of concern.[29]
23. Among the case study children, however, disruption opportunities were missed on several occasions. For example, one perpetrator was the subject of similar allegations in the past but the case was closed with no rationale given; no child sexual exploitation ‘flag’ placed on his record; and no application for a sexual risk order (SRO) was made even though “the frequency and similarities of the allegations would have made him a likely candidate” for such an order.[30]
24. Victim-blaming language and attitudes were evident in some of the Swansea case studies, despite the guidance on language provided to staff. For example, CS-A24 was described as having had “sexual partners from the age of 11”, even though anyone under the age of 13 can never legally give consent.[31]
25. From November 2019, the Welsh Government moved away from risk assessment forms to encourage a more holistic approach. Its revised guidance included prompts for physical signs, emotional signs, material signs and behavioural signs.[32] Despite this, Swansea Council continued to use the outdated Sexual Exploitation Risk Assessment Framework (SERAF), which they misinterpreted as a “national requirement”.[33] The practitioner prompts should be included in all risk assessments in Swansea as a matter of urgency.
26. Swansea Council also “identified a theme about premature removal”, with two case study children (CS-A24 and CS-A25) removed from its child sexual exploitation protocol while still at significant risk.[34]
27. There were specialist services available for sexually exploited young people in Swansea, through Barnardo’s and the South Wales National Society for the Prevention of Cruelty to Children (NSPCC) Protect and Respect project.[35]
28. During an almost three-year period, 10 children went missing in excess of 50 times each; one went missing 115 times.[36] The local safeguarding children board identified that missing children and young people were only offered return home interviews (RHIs) if they were “perceived as medium to high risk” and that information from RHIs was not routinely shared.[37]
29. It was not possible to determine the numbers of sexually exploited children from Swansea placed outside their home area from the data provided.[38] When a child is placed out of area, there should be a cross-border meeting between police forces (and sometimes a social worker), but such meetings do not always take place.[39]
30. Positive practice in the Swansea area included the fact that suitable accommodation and specialist care was given to two of the case study children beyond their 18th birthday, when they left care.[40]
31. The often hidden nature of child sexual exploitation against boys and young men is recognised in Wales.[41] Swansea Council anticipated that its contextual safeguarding pilot with a wider brief on criminal exploitation would enable it to identify more males at risk of sexual exploitation. A January 2020 systems review found that Swansea Council was “identifying extra familial harm within current practice better than the other sites that they looked at across the UK, and responding to this in a coordinated manner”.[42]
32. While Swansea Council’s risk assessment identified whether a child at risk of sexual exploitation has a disability, it did not prompt the assessor to consider the impact of disability on the child and any vulnerability to sexual abuse.[43]
33. At the time of our hearings, there was no multi-agency child sexual exploitation team or meeting to monitor threats to children in Swansea. Instead there were stated to be close informal working relationships between the key professionals.[44]
34. Systems for audit and improvement were not generally well used in Swansea.[45]
35. The 2019 Warwickshire problem profile looked at perpetrators and victims but did not record any groups or networks involved in or perpetrating child sexual exploitation in the county.[46] Organised criminal gangs exploiting young people were said to be “relatively rare”, although it was recognised that exploitation was more often by ‘disorganised’ groups with loose associations.[47]
36. It appears that Warwickshire Police conducted little, if any, analysis of the extent to which child sexual exploitation was being committed by groups or networks.[48] As the extent of perpetrator networks and the threat of child sexual exploitation was not properly understood, exploitation – for example that reported by CS-A2 in April 2017 – could not be properly addressed.[49]
37. In the Warwickshire 2019 profile, the ethnicity of over 40 percent of 137 perpetrators and over 50 percent of 162 victims of child sexual exploitation was unknown.[50]
38. Disruption tactics were not all formally recorded, as a result of which Warwickshire Police could not assess their effectiveness and revise them as required. It issued a number of CAWNs in 2018/19 but accepted that it had been too slow to implement other approved disruption tactics against perpetrators in some cases.[51]
39. Training aimed at eradicating victim-blaming language and attitudes had been given to frontline teams in Warwickshire County Council and Warwickshire Police. Victim-blaming language was seen in respect of two of the case study children but generally, where victim-blaming language was used, it was challenged and managed. [52]
40. There was a more progressive approach to risk assessment in Warwickshire, which uses chronologies in all assessments and has adopted the National Working Group (NWG) Child Sexual Exploitation Assessment Tool. The use of this assessment was shown to have improved professional judgement, and promoted greater involvement of children and parents/carers in assessment and planning.[53]
41. There were concerns about the operation of the risk assessment process in Warwickshire in the case of CS-A1. Despite earlier concerns from her family, CS-A1 was only later identified as at high risk of sexual exploitation. Warwickshire County Council accepted the delay and stressed subsequent changes in the risk assessment process.[54]
42. There was evidence of difficulties for children in accessing Child and Adolescent Mental Health Services (CAMHS) in Warwickshire.[55] However, there were examples of positive support, including in prevention with a Barnardo’s CSE Training and Outreach project worker, targeted support from youth workers for children deemed to be at low risk of sexual exploitation, and specialist provision by Barnardo’s for sexually exploited young people.[56]
43. There were poor outcomes for CS-A1, who was cared for in a succession of unsuitable placements in Warwickshire and the risks of placing her in proximity to offenders were ignored.[57] This highlighted major and relatively recent failings in the systems of both Warwickshire County Council and Warwickshire Police for identifying sexually exploited children in care and taking effective action to protect them.
44. Warwickshire County Council appeared to have good procedures in place for communicating with relevant agencies before placing a looked after child out of the area. Children from other areas placed in Warwickshire were also monitored. [58]
45. A multi-agency campaign in Warwickshire – ‘Something’s Not Right’ – aimed to increase awareness of the sexual exploitation of boys and young men and encourage them to speak out about their experiences and seek support. Dedicated resources have been put in place for male victims, with a Barnardo’s team targeting boys and young men with services to prevent and disrupt sexual exploitation.[59]
46. The processes used in Warwickshire form a foundation for effective identification of disabilities but the low rate of identification and recording of disabilities in sexually exploited children is a concern. The disability of one of the case study children (CS-A300) was not properly recognised or recorded by the Council.[60]
47. A multi-agency, co-located child exploitation and missing team has brought together the statutory agencies, Barnardo’s and missing children practitioners in Warwickshire. Warwickshire County Council has engaged the Tackling Child Exploitation Support Programme to help improve its service. [61]
48. Warwickshire had a well-established audit and review process.[62]
49. Data collection about child sexual exploitation was problematic in St Helens. The child sexual exploitation problem profile had not been updated and appeared to be based on out-of-date analysis.[63] All of this inhibited the ability of Merseyside Police to identify perpetrators and protect victims.
50. St Helens’ 2018 problem profile considered organised criminality but did not identify any organised crime group (OCG) engaged in child sexual exploitation.[64] However, as at September 2019, Merseyside Police had identified 11 OCGs for the Merseyside region with child sexual exploitation as a primary or secondary crime type.[65] There were other instances of child sexual exploitation which involved a “group/gang”.[66] As there was only “a limited amount of resource” to identify children who are vulnerable to sexual exploitation by a group not formally classed as an OCG, networks in St Helens not formally classed as OCGs did not appear to be a focus of the policing response to child sexual exploitation.[67]
51. Between April 2017 and March 2019, in St Helens, Merseyside Police noted that 41 percent of 435 victims and 28 percent of 217 alleged perpetrators of child sexual abuse and exploitation did not provide their ethnicity.[68]
52. Merseyside Police accepted that the information on the extent to which it had used each of the disruption tactics from the toolkit “was not easily available”.[69] In 2018, HMICFRS identified “limited use” of CAWNs as a tactic for disruption and prevention. Improvement has been slow; Merseyside Police accepted that there were “a very limited number of CAWNs issued” and that there had been “missed opportunities”, as highlighted by the experiences of several of the case study children.[70] It also appeared to have made relatively little use of civil orders.[71]
53. Although training and guidance around victim-blaming had been given to frontline children’s social care and police staff in St Helens, victim-blaming attitudes continued to be evident for some time. By the time of an audit in September 2020, however, only one example of inappropriate language was found.[72]
54. Risk assessment failed to protect several of the case study children who had been sexually exploited in St Helens. Children in need of support to prevent sexual exploitation were not always classified as sufficiently high risk to merit intervention from Catch22 (which works alongside young people, their families and the police in St Helens to find out what has caused them to run away, and prevent them from running away again in the future).[73]
55. There were examples of placements and assessments of case study children in care which failed to keep them safe. In 2018 and 2019, HMICFRS found that police work around missing children in St Helens was too-often desk-based.[74]
56. St Helens Council has greatly reduced the number of children placed at a distance of more than 30 miles from St Helens.[75]
57. Boys and young men referred because of sexual exploitation concerns and discussed at multi-agency meetings in St Helens accounted for 14 percent in 2017/18 and 11 percent the following year of the total number of children considered.[76] St Helens Borough Council made no specific child sexual exploitation services available for boys and young men.[77]
58. St Helens Council accepted that there was limited scope to identify learning disabilities or needs, or for practitioners to record the effect of disabilities on the child, and no guidance for staff on how to assess and record disability. The disabilities of two children in the St Helens case study group (CS-A27 and CS-A71) were not sufficiently identified and taken into account.[78]
59. The quality of the partnership working in respect of the case study children in St Helens was mixed.[79]
60. In 2019, Ofsted found that children’s social care in St Helens was “inadequate” overall (and had declined in quality since 2014).[80] St Helens Council told us that it had started to develop audit and quality assurance systems for its work with respect to child sexual exploitation but these were at an early stage of development.[81]
61. The Metropolitan Police Service’s pan-London profile in 2019, which included Tower Hamlets, identified gaps in intelligence in several areas, including due to inconsistent flagging.[82] 31 percent of reports in Tower Hamlets did not specify the type of child sexual exploitation suffered by a victim.[83] There was also little information about the level of risk to a child (which was missing in 39 percent of crime reports considered) and the vulnerabilities of the child.[84]
62. The Metropolitan Police Service told the Inquiry that there were no cases or issues with child sexual exploitation by networks in Tower Hamlets, using the Inquiry’s definition.[85] Given the breadth of the Inquiry’s definition of a network, this cannot be right. The 2019 problem profile identified that 9 percent of child sexual exploitation reports for Tower Hamlets in 2018/19 showed links to gangs or organised groups; it also accepted that “young people affected by, or associated with, gangs are at high risk of sexual exploitation”.[86]
63. The London Borough of Tower Hamlets also indicated that it had not identified any cases of sexual exploitation by networks applying the definition of an OCG, although four operations within the Borough would fall under the Inquiry’s definition.[87]
64. Ethnicity was not recorded in the 2019 profile for 86 percent of offenders involved in 147 reports of child sexual exploitation and 14 percent of 166 victims of child sexual exploitation in the Central East Area BCU.[88]
65. The Children’s Society (which provides a commissioned service to support missing and looked after children and children with disabilities in Tower Hamlets) was critical of the London Borough of Tower Hamlets’ practice with respect to disruption.[89] Although disruption tactics should be aimed at perpetrators and locations where exploitation is occurring, the Metropolitan Police Service used CAWNs in Tower Hamlets to disrupt groups of children who were victims of child sexual exploitation.
66. There was some evidence of victim-blaming in relation to CS-A22 and CS-A77.[90]
67. There were failings by the London Borough of Tower Hamlets in the assessment of the risk of child sexual exploitation prior to 2019, such as in its assessment of the risk of CS-A81.[91] More recently it has developed a rounded assessment form that moved away from a tick box approach.[92] However, concerns remain as to how long risk assessments take to be completed and reviewed.[93]
68. In 2016, external inspection found that the Metropolitan Police Service response to children who go missing was poor. By 2018, there were some signs of improvement. Among the case study children, CS-A77 and CS-A90 frequently went missing from home or care without any effective responses.[94]
69. The London Borough of Tower Hamlets accommodated 80 percent of its children in care within 20 miles of the town hall. By October 2020, there were no Tower Hamlets children in unregulated placements.[95]
70. The London Borough of Tower Hamlets accepted that the number of males referred to its child sexual exploitation team was low, and stated that it had put in place several improvements for the identification of male survivors.[96]
71. In common with other areas, improvement is required to systems in Tower Hamlets to increase the identification of sexual exploitation risk for children with disabilities.[97]
72. The London Borough of Tower Hamlets considered that there are effective structures in place, although in the past there had been some disjointed working, in part as a result of changes to partnership meeting structures and a high turnover of staff.[98]
73. The Borough has undertaken substantial work to address Ofsted’s 2017 finding that its children’s social care department was “inadequate”.[99] Similarly, the 2019 HMICFRS post-inspection review of the Metropolitan Police Service area found that there had been structural change and improvements in the approach to child protection.[100]
74. In 2018, HMICFRS identified concerns about the extent to which children reporting sexual exploitation in Tower Hamlets were listened to by Metropolitan Police Service officers and staff.[101]
75. In Bristol, cases flagged as sexual exploitation could have included criminal exploitation cases, until the system was changed so that cases could be flagged for both criminal and sexual exploitation.[102]
76. Despite a significant number of adults who posed a risk to children, there was less information about offender networks.[103] Avon and Somerset Police’s 2019 problem profile noted that:
The second and third groups would be regarded as an organised network according to the Inquiry’s definition.[106] Avon and Somerset Police explained that networks using the Inquiry’s definition were difficult to “segment”.[107]
77. In Bristol, there were a series of issues leading to inaccuracy in the flagging of criminal cases for child sexual exploitation.[108]
78. The agencies in Bristol used algorithmic mechanisms to help with assessing risk and problem profiling. Avon and Somerset Police’s Topaz Risk Assessment Profile (TRAP) system identifies suspects based on a score indicative of their risk of carrying out child sexual exploitation offences, taking into account various factors.[109] Officers then consider whether to flag an individual based on that score and other intelligence.[110] From TRAP, a weekly ‘offender’ list is circulated between multi-agency partners, which includes the reason why each subject has been identified.[111] Bristol City Council uses its Think Family Database to create a profile of children at risk of sexual exploitation, bringing together 35 different datasets about children and known risk factors (such as episodes of going missing).[112] The ‘Insight’ team, funded jointly with the police, use that data to create a predictive risk model to analyse the extent to which children are at risk of sexual exploitation.[113] It produces a weekly, automated list of children at heightened risk of sexual exploitation.[114]
79. The use of predictive analytics to identify children at risk in Bristol and elsewhere has been subject to press comment and academic review.[115] Concerns have been raised about the quality of the data inputted into the model, the risk of reinforcing the errors and biases of those making the original records, and the potential to focus on factors linked to socio-economic and racial discrimination.[116] Bristol City Council’s statistical analysis of the model in September 2020 found it to have ‘Very Strong’ precision and ‘Very Strong’ recall. However, other research was more cautious about predictive models in other areas, finding that four of every five children at risk were missed (false negatives) and, of the children the models identified as being at risk, they were wrong six out of ten times (false positives). The improved collection and use of data is critical to the response to child sexual exploitation but it is important that agencies do not over-rely upon it. On its own, predictive analytics produces too many misleading assessments; it may be a helpful supplement but should not be used as a principal tool.[117]
80. The ethnicity of 28 percent of 137 suspects and 19 percent of 474 children at risk of child sexual exploitation in Bristol was unknown or unrecorded.[118] Internal research by Avon and Somerset Police in 2020 highlighted that some of its systems did not make it easy to record ethnicity; it was sometimes not clear who is responsible for recording ethnicity; there was some lack of knowledge about how to record ethnicity and why it is important; and some officers were not confident in asking members of the public to define their ethnicity.[119]
81. There were regular Operation Topaz meetings where the threats posed by perpetrators and risks to victims were shared. Operation Topaz had been involved in intelligence-led disruption activity against perpetrators and evidence-led prosecutions, in partnership with the Crown Prosecution Service. This had included the use of civil and criminal orders, arrests, and the investigation of associated offences of drugs and theft. Operation Topaz had successfully focussed partnership efforts on tracking and apprehending offenders.[120]
82. Some examples of victim-blaming language were identified in the Bristol evidence.[121]
83. Over the past three years, there has been a steady decline in the number of assessments in which child sexual exploitation was identified as a factor in Bristol. It is unlikely that this reflects a real decline in exploitation. It is more likely that this is related to issues with data capture.[122]
84. The Barnardo’s Against Sexual Exploitation (BASE) project has developed specific ways to support child victims. Examples included sexual health clinics for vulnerable children offered from BASE premises several times a week and a dedicated CAMHS nurse in the BASE project to provide quick and flexible mental health services.[123]
85. Completion of RHIs in Bristol was at a very low level in April 2018, when only 29 percent of children who went missing accepted an RHI. By June 2020, as a result of improvement action by the Council, 92 percent of 114 eligible children had been offered an RHI and 52 percent of them had accepted an RHI.[124]
86. One solution to the under-representation of male victims being adopted in Bristol has been for Barnardo’s BASE (a specialist service that supports young adults aged 18 to 25 who are at risk of being sexually exploited or where there are known concerns around sexual exploitation) to lower the threshold for entry into sexual exploitation support services when boys and young men are referred.[125]
87. In 2019, Bristol City Council proactively audited 12 cases relating to children with a disability who were considered to be at risk of sexual exploitation. This work identified strengths and tangible areas for improvement, which have been implemented.[126]
88. There was evidence of good partnership working in Bristol.[127]
89. Bristol City Council and Avon and Somerset Police have undertaken extensive internal auditing of their practice in relation to child sexual exploitation, including working with the NWG.[128]