Preliminary Psychometric Evaluation of the Kessler Psychological Distress Scale (K10) in a Mozambican Sample: Evidence on Internal Structure and Reliability

Abstract

The Kessler Psychological Distress Scale (K10) is a widely used self-report instrument for assessing psychological distress, particularly symptoms of anxiety and depression. Although the scale has been extensively validated in Western populations, evidence from sub-Saharan African contexts remains limited. The present study aimed to provide a preliminary psychometric evaluation of the K10 in a Mozambican sample, focusing on its factor structure and score reliability. A total of 224 participants, including medical students, trainees, and residents, completed the K10. Confirmatory factor analyses were conducted using the WLSMV estimator for ordinal data to test different measurement models, including a unidimensional model, a correlated two-factor model, a second-order hierarchical model, and a bifactor model. Internal consistency was assessed using Cronbach’s alpha and McDonald’s omega. The unidimensional model showed poor fit, whereas the correlated two-factor model provided the most interpretable representation of the data. The second-order model showed comparable fit but presented estimation problems, and the bifactor model did not converge. Reliability analyses indicated good internal consistency for the total score and adequate-to-good reliability for the subscales. These findings provide preliminary evidence that the K10 is a reliable instrument with a meaningful internal structure in this Mozambican sample. Further research is needed to examine external validity and to confirm these results in more diverse populations.