exposure intervention, KIDNET, Ruf and colleagues [40]
found the intervention superior to a waitlist control and Catani
and colleagues [35] found it not significantly different from a
meditation–relaxation intervention. Ronan and Johnston [33]
found no significant difference between an exposure intervention
and an eclectic cognitive behavioral intervention. As
noted above, Gilboa-Schechtman and colleagues [36] found
prolonged exposure superior to time-limited psychodynamic
therapy after treatment and at 6-month follow-up, but both
interventions were efficacious and there was no significant
difference in the two interventions at 17-month follow-up.
Thus, the composite evidence failed to reveal superiority of
exposure therapy over other interventions.
Because of the controversy regarding the use of debriefing
in adults, the two studies using this intervention in children are
of interest. Thabet's team [34] found no significant difference
in outcomes for three conditions—group debriefing, teacher
psychoeducation, and no treatment; neither treatment was
determined to be efficacious. Stallard and colleagues [38]
found significant improvement in children receiving both their
individual debriefing intervention and an unstructured discussion
unrelated to the child's traumatic experience with no
significant difference in outcomes between the two conditions
and no evidence that the debriefing intervention was harmful.
Shooshtary and colleagues [49] demonstrated efficacy of a
four-session eclectic cognitive behavioral group intervention
which used debriefing as well as other techniques for
adolescents exposed to a massive earthquake in Iran.
Unfortunately, these investigators did not determine the
efficacy of any specific component of the intervention; thus,
the results do not contribute to the debate about the efficacy of
debriefing in children.
In general, studies have not sought to identify the
therapeutic component(s) responsible for improved outcomes
instead measuring outcomes associated with treatment
in general rather than any single component of the
intervention used. While not designed with that in mind, the
two studies by Layne and colleagues [28,29] have
contributed to an understanding of the importance of
dismantling interventions to identify the components
responsible for therapeutic benefit. The next generation of
studies will need more complex and sophisticated research
designs to determine the “active” ingredient in interventions.
Furthermore, studies have not addressed common factors of
interventions that may be associated with outcome. These
include, for example, the therapeutic relationship, the
expectation of therapeutic success, the process of confronting
or facing the problem, the opportunity to experience
mastery or control over the problem, and attribution of
therapeutic success or failure [50].
4.4. Statistical analysis, methodology, and power
A majority of the randomized studies conducted an intentto-treat
analysis of the collected data. This analytical
approach preserves equality between treatment groups