This page was translated automatically from Turkish. Read the original

Bu sayfa, 26. Dünya Psikiyatri Kongresi'nde sunulan çalışmanın yapay zekâ desteğiyle hazırlanmış Türkçe özetidir. Çalışmanın sahibi sunan yazarlardır; atıf için orijinal kaynağa başvurun.

The limits of standardization: test-retest reliability of psychiatric diagnostic interviews

The limits of standardization: test–retest reliability of psychiatric diagnostic interviews

The pooled test-retest reliability of standardized diagnostic interviews is moderate (κ = 0.69) and varies substantially across disorders; structural standardization alone does not guarantee consistent diagnoses.

The study aimed to estimate the test-retest reliability of standardized diagnostic interviews (SDIs) used to classify common adult psychiatric disorders, examine differences across disorders, and assess prespecified factors associated with between-study heterogeneity. MEDLINE, Embase, Emcare, PsycINFO, and Applied Social Sciences Index and Abstracts were searched from inception to September 2025 without date or language restrictions, and the reference lists of eligible articles and relevant reviews were also screened; disorders were selected based on their estimated prevalence in the general adult population, clinical importance, and frequent inclusion in SDIs. The primary outcome was the test-retest reliability of SDI-based diagnoses of adult psychiatric disorders, measured using Cohen’s kappa (κ). Fifty-seven studies were reviewed, and 46 were included in the meta-analysis (535 κ estimates; N = 8,146 participants; range of mean ages 22.0–54.3). Pooled reliability was κ = 0.69 (95% confidence interval 0.66–0.72), with substantial between-study heterogeneity (Q534 = 23578.7; p < 0.001; I² = 93%). Reliability was higher for substance use disorders than for mental disorders (κ = 0.72 and 0.65) and varied by disorder type; univariable meta-regression showed that diagnostic criteria partly explained between-study variability in substance use disorders, whereas indicators of methodological quality, such as small sample size and the retest interval, did not. The authors note that the moderate and heterogeneous reliability suggests that structural standardization alone may not be sufficient for consistent psychiatric diagnosis, highlighting the importance of also considering contextual and phenomenological information in diagnostic assessment and research practice.

Tüm WCP 2026 özetleri