Abstract
Dysfunctions of metacognitive aspects of metamemory and metaconcentration are common in substance use. No questionnaire tool has been validated among substance users to assess these metacognitive abilities. Therefore, this study investigated the psychometric properties of a brief metamemory and metaconcentration scale (BMMS) in adults with substance use. Using a cross-section design, participants (n = 395, age = 18-45 years, 22.7 +/- 2.4 kg/m(2)) were purposively selected from houses earmarked by simple random sampling in Mizan, south-west Ethiopia. Interviewer-administered BMMS, the severity of dependence on khat (SDS-khat), and a socio-demographics tool were employed. A two-factor model with correlated error terms was found valid based on adequate model fit and measurement invariance across gender groups. No major ceiling or floor effect was found in the BMMS scores. Good internal consistency (Cronbach's alpha = 0.83) and adequate internal homogeneity (moderate to strong item-total ISI score correlations; r >= 0.46) were found. Khat dependent and non-dependents differed significantly across BMMS total score, factor scores and all the item scores except three. The BMMS has adequate psychometric validity in the population with substance use.