Affective recall bias index: development and evaluation as a measure of emotional discrepancy in a peruvian population

Introduction: Depressive symptomatology is associated with cognitive distortions that alter affective memory and retrospective emotional coherence. The objective was to develop and evaluate the performance of the Affective Recall Bias Index (ISRA) as a measure of discrepancy between retrospective and current depressive symptoms in a Peruvian population. Materials and Methods: Analytical study using secondary data from the 2022 and 2024 Demographic and Health Surveys (ENDES). The ISRA was calculated as the absolute difference between PHQ-9 scores referring to symptoms in the past 12 months and the past 14 days. Evidence of internal validity was assessed using Pearson correlations; variation by severity with ANOVA and Tukey's test; and diagnostic performance with ROC curve analysis. The association with current depression (PHQ-9 ≥ 10) was analyzed using logistic regression adjusted for sex, education, partnership status, and alcohol consumption. Results: The ISRA showed consistent correlations with depression measures in 2022 and 2024. PHQ-9 scores referring to the past 12 months were positively associated with those of the past 14 days and with the ISRA, confirming internal coherence. ROC analysis showed adequate discriminative capacity (AUC = 0.788; sensitivity = 0.801; specificity = 0.622). ANOVA showed an increasing gradient of ISRA according to depressive severity (F[3,34,298] = 3112.4). In logistic regression, the ISRA was significantly associated with current depression (OR = 1.188). Conclusions: The ISRA showed evidence of internal, discriminative, and temporal validity, suggesting its potential utility as a cognitive indicator of affective discrepancy associated with depression.