Background: Accurate assessment of adolescent psychopathology is often hindered by "informant discrepancy." While Western literature identifies these discrepancies as indicators of situational specificity, there is a critical lack of evidence from South Asian populations specifically Pakistan where cultural norms regarding emotional disclosure and adult-child hierarchies may uniquely moderate inter-rater agreement. This study quantified the level of absolute agreement and rank-order consistency among Pakistani adolescents, parents, and teachers regarding social, emotional, and behavioral risks using the BASC-3 BESS framework. Method: This study utilized a cross-sectional, multi-informant design in Islamabad and Rawalpindi (N=279), Pakistan, recruiting 169 adolescent self-reports with matched behavioral evaluations from 41 parents and 69 teachers. Inter-rater reliability was assessed using Spearman’s Rho (ρ) for consistency and Intra class Correlation Coefficients (ICC; two-way random effects, absolute agreement) to evaluate consensus on symptom severity. Results: This study’s analyses revealed a systemic lack of consensus across all dyads within the Pakistani sample. Spearman correlations were weak to moderate, with the highest consistency found between parent-teacher reports of internalizing risk (ρ = 0.40, p < 0.01). Critically, ICCs demonstrated "poor" absolute agreement across the triad (Adolescent-Teacher-Parent) for the global Behavioral and Emotional Risk Index (ICC = 0.055, p < 0.001). Conclusions: These findings suggest that adolescent psychopathology in Pakistan is highly context-dependent, manifesting differently in home versus school environments. The exceptionally low absolute agreement (ICC < 0.10) issues a rigorous scientific caution against "sole-informant" diagnostic models in South Asian clinical practice. These results highlight the necessity of prioritizing adolescent self-reports to capture internalizing risks that frequently remain invisible to adult observers. .
Rohail et al. (Sun,) studied this question.