RESUMO
OBJECTIVE: The Sydney Melancholia Prototype Index (SMPI) is a scale that uses a non-conventional strategy to assess melancholia status based on prototypic symptoms and illness course variables. This study aimed to evaluate the performance of the first translation of this instrument in a non-English-speaking population. METHODS: A sample comprising 106 Brazilian outpatients with depression was evaluated simultaneously with the Brazilian version of the self-rated SMPI (SMPI-SR) and clinician-rated SMPI (SMPI-CR). The kappa coefficient and t test were used to evaluate concurrent validity vs. DSM-IV, CORE system, Hamilton Depression Rating Scale-6 item (HAM-D6), and HAM-D17 assignments to a melancholic or non-melancholic class. The prevalence of melancholia as well as sensitivity and specificity were calculated across instruments. RESULTS: The prevalence of melancholia was highest using DSM-IV criteria (56.6%). The kappa agreement between SMPI-CR and DSM-IV melancholia assignment was moderate (kappa 0.44, p ≤ 0.001). SMPI-CR-assigned melancholic patients had significantly higher CORE, HAM-D17, and HAM-D6 scores. The test-retest consistency values for the SMPI were modest at best, and somewhat superior for the CR version. CONCLUSION: The Brazilian SMPI-CR presented satisfactory psychometric properties (which were superior to those of the SMPI-SR), and therefore appears to be a useful option for identifying melancholia and studying its causes and optimal treatments.