ADAPTATION AND PSYCHOMETRIC VALIDATION OF THE KAZAKH VERSION OF THE EDINBURGH POSTNATAL DEPRESSION SCALE

Postnatal depression (PND) is a significant public health concern affecting the health of the mother, child, and family. Despite its high prevalence, Kazakhstan lacks a validated screening tool, which complicates early detection at maternity hospitals and primary care facilities.

The aim of the study was to translate, culturally adapt, and psychometrically validate the Edinburgh Postnatal Depression Scale into Kazakh.

The study included 107 women in the early postpartum period. Translation was performed according to international standards (forward-backward translation, expert review). All participants underwent a psychiatrist's clinical interview (gold standard). Internal consistency (Cronbach's α), test-retest reliability (ICC(2,1)), classification agreement (Cohen's κ), and discriminative ability via ROC analysis were assessed.

Signs of mild PND were identified in 30 women (28.0%). The Kazakh version of Edinburgh Postnatal Depression Scale (EPDS) showed good internal consistency (α=0.86) and excellent test-retest reliability (ICC=0.99; 95% CI: 0.985-0.993). Cohen's κ at the ≥7 threshold was 0.90 (95% CI: 0.82-0.96), with a simple agreement rate of 95.3%. ROC analysis demonstrated good discriminative ability (AUC=0.88; 95% CI: 0.824-0.949; p<0.001). The optimal cut-off defined by Youden's index was ≥7, yielding 80.0% sensitivity and 75.3% specificity.

The Kazakh version of EPDS is a reliable and valid tool for PND screening. A cut-off score of ≥7 is recommended for use in primary care to identify women at risk and ensure timely referral to specialists.