Conflict of interest statement: No conflicts declared.
Criterion validity of the Mood and Feelings Questionnaire for depressive episodes in clinic and non-clinic subjects
Article first published online: 22 AUG 2006
Journal of Child Psychology and Psychiatry
Volume 47, Issue 9, pages 927–934, September 2006
How to Cite
Burleson Daviss, W., Birmaher, B., Melhem, N. A., Axelson, D. A., Michaels, S. M. and Brent, D. A. (2006), Criterion validity of the Mood and Feelings Questionnaire for depressive episodes in clinic and non-clinic subjects. Journal of Child Psychology and Psychiatry, 47: 927–934. doi: 10.1111/j.1469-7610.2006.01646.x
- Issue published online: 22 AUG 2006
- Article first published online: 22 AUG 2006
- Manuscript accepted 5 April 2006
- Pediatric depression;
Background: Previous measures of pediatric depression have shown inconsistent validity in groups with differing demographics, comorbid diagnoses, and clinic or non-clinic origins. The current study re-examines the criterion validity of child- and parent-versions of the Mood and Feelings Questionnaire (MFQ-C, MFQ-P) in a heterogeneous sample of children and adolescents from clinic and non-clinic sources.
Methods: Among 470 consecutive youth completing semi-structured interviews at a university-based child psychiatry center, total scores from the 33-item MFQ-C and 34-item MFQ-P were examined across subjects with and without mood disorders using analysis of variance, and receiver operating characteristics analysis.
Results: Mean scores of the MFQ-C and MFQ-P, respectively, differed significantly (p < .0005) across youth having major depressive episodes (MDE) (33 and 32, n = 77), mood disorders not meeting criteria for current MDE (24 and 28, n = 75), and no mood disorders (12 and 10, n = 318). In the overall sample, areas under the curve (AUC) for discriminating MDE and any mood disorder, respectively, were .85 and .83 on the MFQ-C, .86 and .90 on the MFQ-P, and .89 and .90 on the MFQ-C and MFQ-P averaged together, suggesting moderate to high criterion validity. Similar findings were noted in subgroups divided by age, sex, race, comorbid psychopathology, and clinic or non-clinic origins. AUCs of these MFQ scores compared favorably with those of the Beck's Depressive Inventory, the Child Behavior Checklist's Anxious/Depressed scale and the Children's Depressive Rating Scale–Revised by the same raters. A score of 29 on the MFQ-C (positive screen rate 21%, sensitivity 68%, specificity 88%) or 27 on the MFQ-P (positive screen rate 23%, sensitivity 61%, specificity 85%) optimally discriminated youth with MDE from the rest of the sample.
Conclusions: The MFQ-C and MFQ-P, especially used in combination, validly identify MDE or other mood disorders in youth diverse in demographic and clinical characteristics.