Diagnosis of Posttraumatic Stress Disorder in Preschool Children

Abstract

This study investigated the existing diagnostic algorithms for posttraumatic stress disorder (PTSD) to determine the most developmentally sensitive and valid approach for diagnosing this disorder in preschoolers. Participants were 130 parents of unintentionally burned children (1–6 years). Diagnostic interviews were conducted with parents to assess for PTSD in their child at 1 and 6 months postinjury and the Child Behavior Checklist for 1.5–5 was also completed. The proposed algorithm for PTSD in preschool children for the Diagnostic and Statistical Manual of Mental Disorders (5th ed.) provided the most developmentally sensitive and valid measure of PTSD. The rate of PTSD diagnosis was 25% at 1 month and 10% at 6 months. The predictive utility of Criterion A was not demonstrated. These findings provide support for the inclusion of the proposed algorithm for PTSD in preschool children.

Acknowledgments

This research was supported by the Cressbrook Committee & Royal Children's Hospital Foundation Postgraduate Scholarship and Australian Post Graduate Award. We thank Professor Michael Scheeringa, Dr. Robyne Le Brocque, and Professor Roy Kimble for their advice and support during various stages of the project. We also thank Joan Hendrikz for her time and helpful statistical advice. Finally, we acknowledge the anonymous reviewers for their helpful feedback.

Notes

Note: %TBSA = percentage total body surface area burned.

^a %TBSA data were not available for three children.

^b Data not available for 12 parents.

Note: The symptoms presented in this table include the developmental modifications to DSM–IV wording (Scheeringa et al., 2003). These modifications are noted in italics. DSM = Diagnostic and Statistical Manual of Mental Disorders; PTSD–DSM–V = proposed DSM–V PTSD algorithm; PTSD–DSM–IV = DSM–IV PTSD algorithm; PTSD-AA = alternative PTSD algorithm; PTSD–DSM–V–P = proposed PTSD algorithm for preschool children; PTSD-1/1/1 = initially proposed PTSD-AA.

^a N = 130.

^b N = 125.

Note: Means are based on the exact number of cases that are added in each threshold change and are not cumulative. As only one child had PTSD–DSM–IV at 6 months, there is no standard deviation and CBCL data were not available for this case. DSM = Diagnostic and Statistical Manual of Mental Disorders; PTSD-1/1/1 = initially proposed alternative PTSD algorithm; CBCL = child behavior checklist; PTSD = posttraumatic stress disorder; DSM-V-P = proposed PTSD algorithm for preschool children.

Note: PTSD–DSM–V–P status is based on children who met Criteria B through G. TBSA = total body surface area.

^a Sensitivity (proportion of children Criterion A correctly predicts as having PTSD).

^b Specificity (proportion of children Criterion A correctly predicts as not having PTSD).

^c Positive predictive value (proportion of children with a PTSD diagnosis who met Criterion A).

^d Negative predictive value (proportion of children without a PTSD diagnosis who did not meet Criterion A).

^e Overall efficiency (percentage of children that are correctly classified by Criterion A).