Objective: Increasingly, children are being diagnosed with psychiatric disorders, including preschool-age children. These diagnoses in young children raise questions pertaining to 1) how diagnostic algorithms for individual disorders should be modified for young age groups, 2) how psychopathology is best detected at an early stage, and 3) how to make use of multiple informants. The authors examined these issues in a prospective longitudinal assessment of preschool- and elementary school-age children who were exposed to a traumatic event.
Method: Participants were 114 children (age range: 2-10 years) who had experienced a motor vehicle accident. Parents and older children (age range: 7-10 years) completed structured interviews 2-4 weeks (initial assessment) and 6 months (6-month follow-up) after the traumatic event. A recently proposed alternative symptom algorithm for diagnosing posttraumatic stress disorder (PTSD) was utilized and compared with the standard DSM-IV algorithms for diagnosing PTSD and acute stress disorder.
Results: At the 2- to 4-week assessment, 11.5% of the children met conditions for a diagnosis of PTSD based on the alternative algorithm criteria per parent report, and 13.9% met criteria for this diagnosis at the 6-month follow-up. These percentages were much higher than those for DSM-IV diagnoses of acute stress disorder and PTSD. Among 7- to 10-year-old subjects, the use of combined parent- and child-reported symptoms to derive a diagnosis resulted in an increased number of children in this age group who were identified with psychiatric illness relative to the use of parent report alone. Agreement between parent and child on symptoms for 1) a diagnosis of PTSD based on the alternative algorithm criteria and 2) diagnoses of DSM-IV acute stress disorder and PTSD in this age group was poor. Among 2- to 6-year-old subjects, the alternative algorithm PTSD diagnosis per parent report was a more sensitive predictor of later onset psychopathology relative to a diagnosis of DSM-IV acute stress disorder or PTSD per parent report. However, among 7- to 10-year-old subjects, a combined symptom report (from both parent and child) was optimal in predicting posttraumatic psychopathology.
Conclusions: These findings support the use of the proposed alternative algorithm for assessing PTSD in young children and suggest that the diagnosis of PTSD based on the alternative algorithm criteria is stable from the acute phase onward. When both parent- and child-reported symptoms are utilized for the assessment of PTSD among 7- to 10-year-old children, the alternative algorithm and DSM-IV criteria have broad comparable validity. However, in the absence of child-reported symptoms, the alternative algorithm criteria per parent report appears to be an optimal diagnostic measure of PTSD among children in this age group, relative to the standard DSM-IV algorithm for diagnosing the disorder.