Reliability and validity of a Central Kurdish version of the Dizziness Handicap Inventory

Abstract

We cross-culturally adapted the Dizziness Handicap Inventory (DHI) into Central Kurdish dialect (DHI − CK) and verified its reliability and validity. A cross-sectional study was utilised to measure the impacts of vestibular disorders. Along with the DHI − CK, two comparators were introduced: the Visual Analogue Scale and the Clinical Test of Sensory Interaction and Balance. External and internal reliability were tested with intraclass correlation coefficient (ICC) and Cronbach’s alpha/composite reliability, respectively. Patients (n = 301; mean age = 44.5 ± 15.2 years; 59.8% women) presenting with vestibular symptoms for at least 30 days who were diagnosed with a vestibular disorder and healthy participants (n = 43; mean age = 42 ± 17.9 years; 62.8% women) (N = 344). The DHI − CK and its three sub-scales—Physical, Emotional, Functional—exhibited good to excellent external reliability: ICCs in the test-retest were 0.93, 0.88, 0.91, and 0.92, respectively. Cronbach’s alphas were 0.87, 0.71, 0.75, and 0.73, respectively. Convergent validity was supported by Spearman’s correlations between the DHI − CK and the comparators. The receiver operating characteristic curve analysis confirmed discriminating validity. The DHI − CK was cross-culturally validated. It is a reliable and valid tool that can be used by clinicians and researchers to quantify vestibular disorder outcomes in Kurdish - speaking populations.

Measurement Properties of the Dizziness Handicap Inventory: A Systematic Review

“…Eighteen studies reported on test–retest reliability of the DHI total score (Table 4) (4,20,22,26,27,30,34,35,43,46,47,49,51,53,54,57–59), with sample sizes varying from 14 (4) to 106 (20). Time between the test and retest ranged from “the same day” to “15 days.” We considered a time interval of >24 h “appropriate” and consequently rated the methodological quality of two studies as “inadequate” due to inappropriate (i.e., too short) time interval (4,20).…”

“…Fifteen studies reported data on construct validity (4,19,21,23,25,27,34,40,43–45,47,51,55,59) (Table 5). The methodology of the studies was rated as “adequate,” because we assumed that the measurement properties of the comparator instruments were adequate, the correlated constructs were clear, and the statistical methods appropriate.…”

“…A multiple forward and backward translation procedure was used in 13 out of 24 studies (20,22,25,26,29,32,34,36,42,47,49,53,54). Cultural or textual adaptations were made in some translated versions of the DHI (26,32,35,51,58).…”

“…No studies were found in which the primary goal was to assess the content validity of the English or translated versions of the DHI. However, the majority of studies which translated the DHI (20 out of 22) included some form of pilot testing for (one or more aspects of) the content validity, that is, relevance, comprehensiveness and/or comprehensibility (22,(25)(26)(27)29,32,34,35,37,38,42,43,(46)(47)(48)(49)51,54,57,58). We rated the methodological quality for these individual studies as ''doubtful,'' mainly because clear descriptions of the process of executing the cognitive interviews were lacking.…”

Global Competence in Canadian Teacher Candidates

“…The critical values for coefficient ω are similar to Cronbach Alpha values. The average interitem correlation should fall between 0.15 and 0.50 (Clark and Watson, 1995;Zmnako and Chalabi, 2019). Spiliotopoulou (2009) specified that the average interitem correlation should be included "within the range of 0.15-0.20 for outcome measures that measure broad characteristics (i.e., general constructs such as extraversion) and between 0.40 and 0.50 for those tapping narrower ones (i.e., specific constructs such as talkativeness."…”