Accuracy of the Diagnostic Infant and Preschool Assessment (DIPA) in a Dutch sample

Background: To prevent negative effects of early-onset psychiatric disorders on children's development, structured diagnostics are needed. However, validated diagnostic instruments (based on DSM-5) for children aged 7 years and younger are scarce. The Diagnostic Infant and Preschool Assessment (DIPA) is a diagnostic interview developed in the USA for measuring 16 psychiatric disorders in young children. The psychometric properties of the American version of the DIPA have been validated. Here we determined the accuracy of the psychometric properties of the Dutch DSM-5 based version of the DIPA for the corresponding population. Material and methods: Psychometric properties of the DSM-5 based version of the DIPA were determined based on a sample of 136 biological, foster, therapeutic foster and adoptive parents of clinically referred children and children involved in a serious accident (aged 1–7 years). In line with the American validation study, we included the following seven DIPA modules: posttraumatic stress disorder (PTSD), major depressive disorder (MDD), attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), separation anxiety disorder (SAD), generalized anxiety disorder (GAD), and obsessive-compulsive disorder (OCD). We administered the DIPA, Trauma Symptom Checklist for Young Children (TSCYC) and Child Behavior Checklist (CBCL). Analyses were conducted with continuous outcomes (number of symptoms) and categorical outcomes (diagnoses). Results: The Dutch DSM-5 based version of the DIPA showed good internal consistency and interrater reliability with both continuous and categorical variables. The concurrent validity was good; we found a good concordance between the DIPA and corresponding questionnaires on both the symptom and diagnoses level. In addition, the divergence on symptom level between the DIPA and non-corresponding questionnaires was adequate, which indicated adequate divergent validity. Due to a limited number of positive cases, we could not draw conclusions regarding its psychometric properties in the GAD and OCD modules. Conclusions: Our study shows promising initial results regarding the reliability and validity of the Dutch version of the DIPA, that is based on the DSM-5. Therefore, we recommend the use of the DIPA in research and clinical practice.