This paper explains and evaluates
the current evidence-base for risk assessment tools in the identification of
Child Sexual Exploitation (CSE) in frontline practice. Specifically, the
evidence-base for “indicators” used and of implementation of these via standardised
risk assessments commonly used in child-protection practice.
CSE is a form of sexual abuse
involving a child under 18. The abuse contains power imbalances used to coerce
and manipulate a child into sexual activity. The child in the exploitative
situation receives something in “return” and/or the perpetrator will gain a financial advantage or increased status. (Department of
Education , 2017).The
risk assessment tools central to this paper are designed to be used by
practitioners working with children for whom there are concerns regarding CSE.
The definition of
evidence-based practice relied on throughout this paper draws on Sackett et al (2000) presentation of evidence-based practice as the
integration of the best available evidence with clinical expertise and patient
values. (Sackett, et al., 2000).
A considerable focus
in this paper is placed on research evidence. Since, CSE risk assessment tools
can be said to filter “clinical expertise” and “patient values”, in that the
knowledge gained from these sources is frequently assessed against what is
“known” about CSE, collated in risk assessment tools. Whilst this hierarchal
notion of “evidence” is problematic, it is difficult to overcome. Practice requires a base to synthesise information and historically this
manifests as common “indicators” upon which information from other sources is applied.
Thus, evaluation of the evidence-base for these tools is fundamental.
The notion of evidence-based
practice being “grounded” in prior findings (Cournoyer, 2013) is presented as offering consistency
and predictability. Following this, risk indicators in CSE tools can be presented
as offering practice “grounded” in emerging patterns indicating risk of CSE,
assuming the evidence-base for these indicators is robust.
In CSE risk assessment tools there are a sizeable number of risk indicators, most
are outside the remit of this paper. Following a review of the literature, focus will be placed
on several specific indicators given the contributions their evaluation
can make to practice. The following section of the paper presents the
evidence-base for indicators and the use of standardised CSE risk assessment
tools, throughout making recommendations for future research.
Brown et al (2016) undertook
a Rapid Evidence Assessment (REA) on “indicators” used in risk assessment
tools evaluating their current
evidence-base. Noting significant variability across localities as to which
indicators and how many were needed for risk thresholds. This is problematic
since the evidence-based risk levels could
then be said to be dependent on where a
child resides rather than an evidence-based risk analysis. Furthermore, Brown concluded that of the many
indicators present only two have a robust evidence
base to present them as “indicators” of risk to CSE.
Davies et al (2000) suggest evidence-based
practice is geared towards what is
being done is “worthwhile” and carried
out in the “best possible way”. In line with practice requirements to act “in
the best interests of the child” (Children Act 1989), it is a concern if actions or
inaction lack a robust evidentiary basis. Over-reliance on unsubstantiated indicators
could result in children being “screened out” however other risk indicators
could be present but are overlooked.
then the two indicators identified as having a robust evidence-base, as per
Brown et al (2016) “Children who are
disabled” and “Children in Residential Care”.
who are disabled
Since Brown et al (2016) no further relevant studies have been identified for
this paper pertaining to CSE and disability thus, Brown can be cited as
offering an up-to-date synthesis of the current evidence-base for this
Prominent to Brown’s position, Jones et al
(2012) analysed systematic reviews and meta-analyses on the prevalence of “risk of sexual violence” in
children with disabilities. Concluding that children with disabilities were at
a higher rate than others to be victims of sexual violence. Whilst systematic
reviews are considered “gold standard” in the hierarchy of evidence the
inclusion criteria for studies in Jones et
al (2012) resulted in some studies with limited methodological rigour and
generalisability to be included (See Elbeling H et al (2002); Reiter et al
(2007). Thus, whilst systematic reviews are robust and Brown et al (2016) are justified in presenting
this as rigorous evidence for risk association this should be considered
critically for its contribution.
Following this Roberts et al (2015) conducted a case-control
study on exposure to childhood sexual abuse in women with autism. Finding that
rates of sexual abuse correlated with increased severity of autism. Critically,
again, however, all the participants were mothers from a predominately white
ethnic background, which causes issues for generalisability. Nevertheless, the
study was longitudinal which has marked benefits for understanding CSE risk given
the fluctuating nature of risk and disability additionally the sample size was promising
The nature of CSE is distinctive
from other forms of child abuse. Whilst the “best evidence” available will
undoubtedly result in drawing on research from other disciplines and on
research on others forms of child abuse, it is essential to recognise the
distinct differences between types of abuse, and distinct indicators.
After Children (LAC) and Children in Residential Care
Children in residential care have been
held to be particularly vulnerable to CSE. Becketts (2011) research analysed
historic risk assessments that found concerns
relating to CSE. Disproportionately, around three-quarters of these assessments
concerned children who were “looked after” at the point of assessment. Equating
to almost 70 percent of the 12-17-year-old LAC in the study location. Given the
disproportionate representation of LAC children in assessment where CSE
concerns were indicated, it is viable
that such a connection be drawn.
Again, it is important to be mindful
of the co-existing factors that may place a child at risk, aside from their LAC
status. Drawing on theories such as attachment (Bowlby; Ainsworth) or identity
formation (for example Erikson) whilst LAC status may be associated with CSE,
it is not the cause. Contact with services forms part of a child’s LAC status
practitioners are arguably more likely to assess and identify CSE in LAC
children. Although Serious Case Reviews (SCR’s) such as Rotherham and
Oxfordshire have highlighted that LAC status may not associate with successful identification
Nevertheless, statistics from
practice indicate that children who are in residential care are
disproportionately represented in cases involving CSE. Pearce (2014) sent questionnaires
to 144 Local Safeguarding Children Boards (LSCB) and found that 684 of the 1065
CSE cases involved children with LAC status of which 54 percent were in
Furthermore, Coy (2009) used
narrative interviews to explore women’s perceptions of risk to CSE and
residential care. The women reported an association
between feelings of destabilisation from multiple placement moves that impacted
on their capacity to form trusting relationships and CSE. This study can be
used to draw on service user voices to understand and build an evidence base
for “residential care” as a risk indicator and aids in the explanation as to why this may be the case.
Both disability and residential care
were presented by Brown et al (2016)
as evidence-based indicators. The inclusion of these in risk indicators allows
for evidence-based decision making. However, as this paper has shown an
exploration of the original sources relied upon and wider research demonstrates
gaps remain. As outlined the evidence base for CSE risk indicators mirrors the
complexity seen in CSE, the vast number of indicators can be viewed a positive
reflection of this. However, the evidence relied upon to present these as
indicators does not sufficiently acknowledge this complexity. Illustratively,
running away as an indicator, although widely referenced in the assessment
tools has a contradictory evidence base.
longitudinal study analysed data from two groups of adolescents. The groups
were established based on whether they had “sold” sex. This links with CSE
given the age the women became involved in sex work. Those who reported
involvement in sex work had significantly higher rates of running away in
childhood than their counter group. However, Klatt (2012) conducted an in-house
audit of data collected from voluntary organisations that work with children at
risk of CSE. Whilst they observed many variables associated with CSE the study
concluded that running away was not found to significantly increase the risk to CSE. These examples are not exhaustive of
the vast research base on “running away” and CSE the purpose is to illustrate
the nature of the confliction that exists within this base.
issue being that practice is presented as rooted in an evidentiary basis, but holds
a contradictory one in the research. Despite this these indicators are
referenced online (see for example Safe & Sound); in policy (Department of Education , 2017) and in practice.
Further research should explore what evidence bases these sources are using and
how they are being evaluated and monitored as little explanation is offered.
profile media cases have led to stereotyped CSE; a heterosexual female victim
(white, from a disadvantaged background), with the perpetrators being male of
“Asian” decent (Fox, 2016).
Given the impact the media has on perceptions in society, future research
should evaluate “commonly” held views on indicators. Equally, practitioners must be proactive and reflective in their
knowledge base, considering the indicators critically. Yet, organisational
systems gear practitioners towards the use
of such tools thus the evidence base for them is critical.
additional issue with the evidence-based indicators arises when working with contextually
marginalised groups, here the decline in a focused evidence-base alters the “effectiveness”
of established risk indicators.
majority of reported CSE involves females, thus the sizable proportion of the evidence-base for practice focuses on male-to-female
abuse. As research on other forms of child abuse has shown there are clear differences
in victimisation between sexes (see for example Asscher et al (2015)). Currently,
practice tools, are disproportionately informed by research on females but
applied to both sexes, thus are likely to
be less effective in their applicability
to males. As Fox (2016) presents, practitioner’s abilities to effectively work
with such cases is inconsistent, this then creates the impetus for filling these gaps through practice research.
Coy (2017) explains the evidence-base for CSE regarding Lesbian, Gay, Bisexual
and Transgender (LGBT) communities is also limited and highlights practitioners
lack of knowledge when working with victims who identify as LGBT. Observing disputes
between practitioners on the “appropriateness” of asking a child about their
sexuality and its implications for practice.
is envisioned then, that future research should explore the effectiveness of asking
such questions in risk assessments to all children. However, ensuring service
users who identify as LGBT form a central part of
this exploration. Effective risk assessment tools should be inclusive, yet
future research should be targeted and focused towards remedying these identified
gaps in knowledge and skills.
the evidence-base for indicators of risk must be established, effective practice also requires reliable and
equally evidence-based methods to implement this knowledge. The next section of
this paper will explore the evidence-based for implementation of indicators
through standardised risk assessment tools.
Currently, practice tool score risk by assessing the number
of indicators against risk thresholds. Some argue scored tools encourage
consistency in decision-making.
For instance, Andretta et al (2016) study in the US evaluated screening tools used in “Commercial
Sexual Exploitation of Children” (CSEC) identification. Arguing that identified
risk indicators could be used more effectively via a quantitative
decision-making system, the Sex-Trafficking Assessment Review (STAR). STAR is a
quantitative tool used to determine risk, based on responses and traits
associated with risk of CSEC (identified through a prior research).
Questions yield a categorical response, for
instance, regarding where the child had resided the previous night. A response
of “at home”; resulted in a risk score of 0 whilst a response of at “a
boy/girlfriend’s house” resulted in a score of 2. The overall score produces an
associated risk level. STAR’s effectiveness centres on it providing large-scale “effective” screening.
In the UK, the socio-economic context of
practice has resulted in the drive for efficient tools. However, such efficiency
must both empowerment and safeguard
children. CSE is a complex multi-faceted issue and those who are at risk have
multi-layered characteristics, many of which interact. Scoring in this way has
the potential to imply merit to experiences of children as “deserving” and
“undeserving” of support. Pertinently, there is a lack of robust evidence base
for CSE risk indicators thus a quantitative tool that assigns rigidity to
decision making based on a flawed evidence-based,
would fail to meet its aims.
in Andretta et al’s study no analysis
was made of outcomes; whether risk thresholds assigned correlated with
identified CSEC. Therefore, risk assessment in CSE
must move beyond tick box practice to effectively utilise and contextualise the
current evidence-base on CSE. With the move
towards a structured qualitative approach.
et al (2017) outline different
approaches to risk assessment (See Table in Appendix 1). The current CSE risk
assessment tools follow an actuarial risk process; the table shows comparisons
between this approach and the structured professional judgement (SPJ). Actuarial
approaches have been viewed as limiting professional discretion, which can lead
to undesirable outcomes a structured approach to the incorporation of
“professional expertise” is important in working with CSE. Given the complexity
of CSE discretion and dissent will form an important part of evidence-based
risk assessments however this requires structure as Reisel (2016) shows
professional discretion in CSE can lead to bias and ill-informed decision
making. Thus, SPJ can bridge the gap between absolute professional judgement
and no professional discretion.
has been effectively used in other areas of safeguarding children such as
neglect (Bortoli, et al., 2017) and is more in line with
the definition of EBP including professional expertise and service users
feedback. Positively, an SPJ tool
developed for CSE risk indicators, like
the current tools aims to contain “aides-memoir” (Bortoli, et al., 2017, p. 663) a list of
evidence-based indicators grounded in several sources. However, unlike the
current system, these would not be scored1.
Removal of this scoring could result in a more qualitative assessment. An issue with the current risk assessment tools
evidence-based approach is the failure to see the risk to this child, irrespective of their risk in relation to “others”. Evidence-based practice implemented via SPJ
risk assessment allows for evidence to
inform and guide decisions that are grounded in research whilst considering
the child in their environment using professional judgement and including their
however, in an era of austerity and
budget cuts, there is merit for the use
of “rankings” of risk; in settings where time, practitioners and resources are
stretched. However, resource-led practice whilst a reality is not a goal, and
assessment should reflect this. Drawing on Holland et al (2000), a critique of the risk assessment tools and their
goal of “efficiency” can be found in the concept of technical rationality. The
position that “social issues” such as CSE can be dealt with through rational
technicality where efficiency and resource-led agendas disproportionately
become the criteria for “resolving” problems.
it has been argued that SPJ risks assessments render tools ineffective and
causes the conclusions reached to lack empirical support, (Guyton & Jackson, 2007) which appear contra
to the roots of evidence-based practice. Whilst SPJ allows for case-specific
factors, that are not included in the list of evidence-based indicators, to be assessed
this is structured. Helibrun et al (2010)
explain when such factors are included
the decision must then be explained by the practitioner. This allows for
professional judgement, autonomy and “local knowledge” (which have been
highlighted in SCR’s as important in CSE cases) to form the evidence base for decision making in a
structured manner. Additionally, this may remedy a critique of evidence-based
practice “de-professionalisation” of social workers. This paper presents SPJ as
an alternative method for risk assessments in CSE from an evidentiary and value-based perspective.
Finally, evidence-based practice
goes beyond having a knowledge-base for indicators and an effective assessment
tool to include in practice skills. Survivors of CSE have highlighted the
significance of the practitioner relationship and of skills such as
non-judgemental communication (Anonymous Girl A, 2013). Evidence-based
practices “effectiveness” will be heavily determined
by the relationship between the practitioner; the child and those who possess
the information that forms our evidence-base.
future research will explore all intersects
of society. Enabling personalised assessments to support the identification of CSE across each intersection
of identity, correspondingly, however, somewhat
unobtainable given the lack homogeneity within these groups. Professional
practice despite a rigorous evidence-base will be uncertain given the
unpredictable and complex nature of society. However, this does not negate the
need for a robust evidence-base for working with risk in CSE. There is a clear
need for a robust and representative evidence base to ensure “risk assessment”
tools if used, are informed and up-to-date.
tools discussed in this paper form components of the “evidence-base”, upon which knowledge from other sources is assessed.
What is “known” about the child is then applied
to the risk assessment indicators. Thus, often the tools act as a “paper
gatekeeper” to the recognition of risk,
this is questionable given their lack of an established evidentiary basis. Whilst
it should be acknowledged that practitioners are not robots, the tools provide a
key function in the assessment process. Given the complex evidence-base, further research should explore
how confident practitioners are in dissenting from conclusions drawn by these
stated CSE risk assessment should be based on “the best available evidence” and
be conducted in the “best interests of the child” CA 1989. This paper has outlined gaps in the current
evidence base which impedes this. Throughout the paper, research recommendations have been presented positioned
towards better understanding and assessment of CSE risk. SPJ assessments have
been presented as a holistic and effective method of implementing evidence-based
risk indicators into practice. Ultimately evidence-based
practice can support practitioners to engage in effective and “grounded”
practice in a complex and difficult area of work, however, this has yet to be successfully realised.
The current model includes a ranking/scoring system of indicators which once
calculated lead to distinct levels of risk to be attributed depending on the
number of “present” indicators.