Clinical properties of the short mood and feelings questionnaire: Development of a free calculator based on the Brazilian high-risk cohort study.

Background: The Short Mood and Feelings Questionnaire (SMFQ) is a validated tool for assessing depressive symptoms in youth, though no specific cut-point exists for the Brazilian population. Item response theory (IRT) and interval likelihood ratios (ILRs) offer refined methods to monitor symptoms but involve complex calculations that hinder clinical implementation.

Methods: Cross-sectional data were drawn from an urban school-based sample (Brazilian High-Risk Cohort Study in 2018-2019, n = 1,905, aged 14-23, 46.6 % females). Diagnoses were based on Development and Well-Being Assessment (DAWBA) clinical ratings. SMFQ factor scores were estimated using IRT and transformed into T-scores. ROC curves evaluated diagnostic properties for internalizing- and externalizing-spectrum disorders. A calculator was developed to estimate post-test probabilities from T-scores using ILRs. Sensitivity analysis excluded MDD as a comorbid diagnosis.

Results: ROC curve analyses suggested a sum score cut-off of >6 and a T-score of >55 for detecting MDD. The SMFQ showed good accuracy for internalizing conditions (AUC >0.8) but low for attention and externalizing disorders (AUC <0.7). ILRs for internalizing conditions ranged from 0.12 (95 % CI: 0.07-0.19) to 29.98 (95 % CI: 11.99-75), with post-test probabilities exceeding pre-test probabilities for scores above the cut-off. Sensitivity analysis confirmed findings when excluding MDD. Including ILRs significantly improved predictive models over dichotomous cut-offs.

Conclusion: The application of ILRs based on IRT T-scores improved SMFQ's predictive ability for internalizing-spectrum conditions, regardless of comorbidity. A calculator can integrate these methods into clinical practice, supporting real-time data-driven decisions.