MINI-KID: Reliability and Validity in Children and Adolescents

David V. Sheehan, MD; Kathy H. Sheehan, PhD; Doug R. Shytle, PhD; Juris Janavs, MD; Jamison E. Rogers, MD; Yvonne Bannon, RN, MSHS, CCRC; Berney Wilkinson, PhD; Saundra Stock, MD
The University of South Florida College of Medicine, Tampa (Drs Janavs, Sheehan, Shytle, Stock, and Wilkinson and Ms Bannon) and the Department of Psychiatry, Brown University, Providence, Rhode Island (Dr Rogers).

This poster presentation was supported by GlaxoSmithKline.

Background: Structured diagnostic interviews improve reliability in the assessment of psychiatric disorders in children and adolescents¹ but are often long and difficult to administer.² The Mini-International Neuropsychiatric Interview (MINI-KID) was developed to provide a short but reliable and valid structured diagnostic interview for psychopharmacology trials and epidemiology studies in this population.³

Objective: To examine the reliability and validity of the MINI-KID as a diagnostic instrument in child and adolescent outpatients and controls.

Methods: In this study, 226 subjects, aged 6–17, were administered the MINI-KID and the Schedule for Affective Disorders and Schizophrenia for School Aged Children-Present and Lifetime Version (K-SAD-PL) in a counterbalanced order by blinded raters on the same day. The interrater and retest reliability of the MINI-KID and its concordance with the parent rated MINI-KID (MINI-KID-P) were also tested. Diagnostic agreement was assessed using AUC, kappa, sensitivity, specificity, positive predictive value, negative predictive value, and efficiency at the level of syndromal diagnoses and individual disorders.

Results: MINI-KID: K-SADS-PL concordance was high (AUC = 0.81–0.96, kappa = 0.55–0.87) for syndromal diagnoses of any mood disorder, any anxiety disorder, any substance use disorder, any behavioral disorder, any eating disorder and any psychotic disorder. Sensitivity was substantial (0.64–1.0) for 15/20 individual disorders. Specificity was excellent (0.81–1.0) for 18/20 disorders and substantial (>0.73) for the remaining two disorders. Most of the classification differences between the MINI-KID and the K-SADS-PL were in the direction of the MINI-KID identifying more disorders than the K-SADS-PL. The MINI-KID identified a median of three disorders per subject compared to two on the K-SADS-PL. Overall, the MINI-KID was three times shorter to administer compared to the K-SADS-PL (33 minutes versus 104 minutes). Interrater and retest concordance was high (AUC =0.75–1.0, kappa =0.64–1.0) for all individual MINI-KID disorders except dysthymia. The concordance of the MINI-KID-P with the standard MINI-KID was better for externalizing disorders than for internalizing disorders.

Conclusions: The MINI-KID is a reliable and valid measure of child and adolescent psychopathology that can be administered in a third of time as the K-SADS-PL. The MINI-KID has the advantage of identifying disorders that may be missed on the longer interview.

References

1. Hughes CW, Rintelmann J, Mayes T, et al. Structured interview and uniform assessment improves diagnostic reliability. J Child Adolesc Psychopharmacol. 2000;10(2):119–131.
2. Hodges K. Structured interviews for assessing children. J Child Psychol Psychiatry. 1993;34(1):49–68.
3. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (MINI): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59(suppl 20):22–33.

Back      Poster