Facilitating the effective and early identification of children at risk for developing serious behavioral and emotional problems

In at least 30% of preschool and kindergarten children in the general
population, behavioral and emotional problems are reported to some degree at
any time. However, only a minority of 5-10% of all children will develop high
and persistent levels of behavioral /emotional problems. This minority must be
identified as early as possible because these children are at highest risk of
developing serious negative outcomes already during elementary school, which,
once fully developed, are likely to persist well into adolescence and young
adulthood. In fact, the majority of the societal costs caused by
behavioral/emotional problems are accounted for by these children.
Well-validated instruments to assess behavioral and emotional problems in young
children are available. However, the available instruments are very imprecise
in their identification of young children at risk for developing serious
problems. Meta-analyses showed that the majority of the children with high
levels of behavioral/emotional problems at a certain point in time will not
develop serious or chronic problems, and referral for mental health services is
not justified based on elevated problem scores alone. In addition, about 10% of
the children identified as not needing help, in fact do.
The purpose of the proposed study is to facilitate the effective identification
of children at true risk. To this end, 1,000 children will be followed over the
period from kindergarten to grade 1 of elementary school (age 4 to 7 years).
Behavioral as well as emotional problems will be assessed by multiple raters
(parent-, teacher-, peer reports). These will be complemented by an assessment
of multiple risk and protective factors from the child (e.g., cognitive
abilities), parent (e.g., parent-child relationship, parenting stress), teacher
(e.g., teacher-child relationship), peer (e.g., peer rejection, peer
victimization) and school (e.g., academic achievement) domain. An unbiased
assessment of clinical risk status at outcome age (age 7) will be based on (1)
scores in the clinical range of CBCL and TRF, (2) a judgment by an expert panel
(i.e., child psychiatrist, clinical child psychologist, youth care worker).
In addition to the early identification of children at risk, the project
focuses on the development pathway to psychopathology. Such a possible pathway
may be the influence of social experiences in the school context on stress
regulation.

In Phase 1 of the project, a systematic approach will be used to identify which
factors add to the prediction of children*s clinical risk status. First, we
will determine how well multiple assessments of behavioral problems
(opposition, aggression, ADHD symptoms) indicate the development of behavioral
problems into the clinical risk status at outcome age. In the second step, we
will determine which risk factors assessed form age 4 to 7 are associated with
clinical risk status at each age (4, 5, 6, 7 years). We will then test to what
extent adding information on each of these risk factors to the repeated
behavioral/emotional problems assessment improves the identification of
children at true risk at age 7. In the third step, we will test how risk
factors from the third step are linked to each other at the different ages and
across ages, possibly setting off interlocked chains of risk influences. We
anticipate that the addition of risk variables that (a) have an additive effect
on behavior problem development and/or outcome prediction, and (b) are part of,
or set off a chain of risk influences, will improve the predictive accuracy at
each any of the assessed ages. The same procedure will be followed for
emotional problems (anxiety, depressive symptoms). In the final step we will
test the predictive validity of both sets of behavioral/emotional problem
curves and risk variables sets regarding age 7 clinical risk status.
In Phase 2 of the project we will construct an efficient screening tool to
identify children at risk in everyday practice across ages 4 to 7 years. Using
a Markov chain model, the factors identified in Phase 1 will be used to come to
a decision tree, in which each step specifies what factors should be assessed
to come to an increasing precision in predicting children*s clinical risk
status.
In addition to the early identification of children at risk, the project
focuses on the development pathway to psychopathology. Such a possible pathway
may be the influence of social experiences in the school context on stress
regulation. This path can pass through the influence of social aversive
experience in DNA methylation of promoter regions of genes that are responsible
for regulating stress, which leads to changes in the reaction of the HPA-axis
to stress.

Longitudinal observational study

No risk associated with participation. Burden of participants is low.