The PICAS-study: Validation of the Sexual Knowledge Picture Instrument (SKPI); about the verbal- and the nonverbal ways sexually abused children reveal sexual knowledge.

Sexual abuse in children is a major problem worldwide[4]. From all victims of
sexual abuse, a quarter to a third are under 7 years of age[5]. By potentially
affecting physical, mental and sexual health of the child, it can be a burden
for many years into adulthood[2]. In a 2010 Dutch study, up to 9% of surveyed
high school students reported ever being sexually abused[1]. A meta-analysis
combining prevalence figures worldwide, shows a CSA prevalence of 12.7% in
self-report studies and 0.4% in informant studies[4].
The differences in these prevalence numbers can be explained by study method,
self-report vs. collecting data amongst professionals or organizations and
varying definitions of CSA[6].

Despite the growing attention from pediatricians and other health care
professionals, it is difficult to either reject or confirm a suspicion of CSA.
There are almost never specific signs or *red flags* in the examination of
children suspected to be sexually abused, and the genital examination is often
non-contributing[7].
Unfortunately, most children are unable or reluctant to talk about the abuse.
Spontaneous disclosures are rare for multiple reasons, such as a conflict of
loyalty, shame, guilt and limited verbal capacities of the child[8]. To assist
in confirming the diagnosis or to make a risk assessment on CSA, a few
questionnaires are available internationally, but none of these tools are
validated in the Netherlands[6]. Also, these questionnaires mostly focus on
child abuse in general, and not on CSA specifically[6].
Another problem is that young children are unable to fill in these
questionnaires themselves. Therefore, at this point, a clinician*s confirmation
of suspicions of CSA in young children is a highly subjective matter and mainly
based on expert opinion.

Sexual Knowledge Picture Instrument
The Sexual Knowledge Picture Instrument (SKPI) was originally developed as a
tool to reveal sexual knowledge in young children[3]. It is a child-friendly
picture atlas, in which in addition to a number of illustrations about everyday
routines, the following topics are shown: Physical differences between boys and
girls, gender identity, genitals and their functions, reproduction, sexual
behavior of adults and boundaries between physical intimacy and sexual acts[3].
The use of the SKPI with semi-structured questions from a developed guideline
makes it possible for a trained clinician to conduct a conversation with the
child about these matters, and to observe the child*s behavior at the same
time.
In 2005, a preliminary SKPI study was performed by Brilleslijper-Kater[3].
Hypothesis was that sexually abused children experienced inadequate sexual
behavior and therefore would show more deviant sexual knowledge. Pre-school
children normally appeared to have very little knowledge of sexuality. They
only exhibited basic knowledge of genital differences, gender identity, sexual
body parts and (non-sexual) functions of the genitals. Non-abused children*s
reactions were relaxed, open-minded and unprejudiced taking their own points of
view and experiences as points of reference. This in contrast to abused
children. At first glance they appeared to disclose (even) less information
than non-abused children. Further analyses showed that abused children were
much more reserved in revealing their knowledge. The most remarkable finding,
however, was their nonverbal behavior. Examples of this, such as fidgeting,
ducking under the table, or theatrical behavior seemed to distinguish the
abused group from the non-abused group. To further investigate this, a second
study, specifically aimed at the differences in nonverbal behavior between
abused and non-abused children was conducted. Results showed that abused
children exhibited significantly more nonverbal behavior compared to non-abused
children[3].

Because of the non-verbal character of the SKPI we hypothesize that it is
suitable for all children with a developmental age of 3 to 8 years old,
regardless of their ethnic background. The SKPI is currently being used on a
daily basis at the outpatient clinic of the Social Pediatrics department in the
Emma Children's Hospital. In our clinical practice the use of the SKPI makes an
important contribution to the diagnosis of sexual abuse in young children.
To date, despite its frequent use, the SKPI has only been explored in a limited
extent in the above-described preliminary study. Therefore it is the aim of our
study to determine the accuracy and reliability of the SKPI, showing its
potential as a valuable addition to the clinicians examination in young
children suspected of CSA.
Currently a guideline on diagnostics in suspected CSA is developed by a
multidisciplinary team commissioned by the Nederlandse Vereniging voor
Kindergeneeskunde (NVK). This guideline will be published in 2016, under the
leadership of A.H. Teeuw, pediatrician in the AMC and president of
TASK-Amsterdam. While developing the guideline, professionals of child
protection services, vice squad, forensic medicine and other health care
professionals stated a need for accurate diagnostic instruments to investigate
suspicions of CSA in children.

1) The validation of the SKPI used in children with a developmental age of 3 to
8 years old.

Hypothesis:
- By using the SKPI the examiner will be able to distinguish children who are
victims of CSA from children who are not.

2) The demonstration of the reliability of the SKPI by determining intra- and
inter-observer reliability.

Hypothesis:
- The SKPI is a reliable instrument with high intra- and inter-observer
reliability.

3) The validation of the SKPI in children with a non-Dutch ethnicity.

Hypothesis:
- By using the SKPI the examiner will be able to identify victims of CSA
despite their ethnic background.

First, a systematic literature review of the *state of the art* on the
following topics will be performed:
- The signs in young children during the investigation of CSA, and possible
ethnic differences.
- The available tools contributing in detecting CSA.
- The use of interview techniques in young children.

Subsequently, our main study can be divided into three projects:

1) A (validation)study, establishing the diagnostic accuracy by calculating the
predictive value, sensitivity and specificity of the SKPI in three different
groups of children:

I. Case group (n=50): Known sexually abused children from 3-8 years, enrolled
in cooperation with the Dutch vice squad.
II. Control group (n=100): Children who are not suspected to be sexually
abused, enrolled in Dutch primary schools and health care centers (matched with
cases on age, gender and ethnic background).

After determining the diagnostic accuracy of the SKPI in the complete sample, a
subgroup analysis will be performed to compare the diagnostic accuracy of the
SKPI in children with a Dutch ethnicity to children with a non-Dutch ethnicity.

III. Mixed CSA and non-CSA group (n=100): Children who come into care because
of a (either low or high) suspicion of sexual abuse, recruited at our own
outpatient clinic.


We consider it most relevant for the SKPI to be able to differentiate in this
*real* daily practice population, and distinguish sexually abused children from
the children who were not sexually abused, but may have experienced other
(different) types of abuse. For these children, initially, it is not known if
children are sexually abused. To determine abuse status, we will follow the
children up and assess the reports of the child protection services and the
police to determine if CSA has been confirmed or ruled out.

In all three groups, the researcher using the SKPI will be blind to the abuse
status or any other diagnostic information.

2) A second (validation) study of the reliability of the SKPI is carried out,
to determine the inter-rater and intra-rater reliability by scoring video
recordings of interviews with the SKPI in our own outpatient clinic.

3) After finishing the validation study, a training program with certification
will be developed. This is the first step in the implementation phase of the
SKPI in the Netherlands. By completing this training, the pediatrician or other
professional is able to use the SKPI along with its semi-structured interview
technique. A pilot trial of this training in 20 pediatricians will be performed
in our own clinic.

The researcher will be instructed by the manual, and tries to comfort the child
as far as possible. The sexually abused children , however, may be fearful or
otherwise affected during the investigation. If the child exhibits any signs of
this, immediate psychiatric help will be consulted.