the assessment instruments used and the examination of their
psychometrics if applied to the study described; the
methodological rigor of the studies including the use of
blinded assessment and random assignment as well as the
type of control; and the use of assessor training. The quality
of the description of the intervention, including the use of
manuals; the use of methods to assess treatment adherence;
and the adequacy of the power to detect small, medium, and
large treatment effects were also examined. Finally, the
efficacy of the interventions with respect to PTSD and/or
posttraumatic stress symptoms or reactions, was determined.
2.2. Target symptoms
In operationalizing “clearly defined target symptoms,”
Foa and Meadows [13] require the study of “significant”
trauma-related psychopathology such as PTSD or a related
condition to assess improvement associated with the
intervention (p. 453). They describe the importance of
using diagnostic status and a threshold of symptom severity
as well as inclusion criteria for beginning treatment and
exclusion criteria to exclude participants with, for example,
comorbid conditions. Because disaster interventions are
commonly used in nonclinical populations of children in the
general community population who suffer a range of
exposures to a disaster, the use of diagnosis as an enrollment
criterion or an outcome measure is not always appropriate.
Thus, in this analysis, which focused on posttraumatic stress
outcomes, posttraumatic stress symptoms or reactions were
identified as the primary outcome if the study did not address
the diagnosis of PTSD. The use of inclusion and/or exclusion
criteria was determined as yes or no for each.
2.3. Reliable and valid measures
The instruments used in the studies were documented as
was psychometric information if it was assessed for the
sample studied.
2.4. Blinded assessment
Use of blinded assessment was recorded as yes or no.
2.5. Assessor training
Assessor training was evaluated to determine if assessors
were trained (coded as yes or no), to indicate if the assessor
was or was not a mental health professional (clinician,
psychologist, psychiatrist, school counselor, graduate psychology
student), and to ascertain if inter-rater reliability was
calculated and if procedures to prevent evaluator drift were
used during the study (each coded as yes or no).
2.6. Manualized treatment and intervention description
Studies were coded as having used a manual, welldescribed
treatment, or neither (yes, well described, and no).