The Kessler Psychological Distress Scale: Translation and Validation of an Arabic Version

Background. The Kessler Psychological Distress Scale has been widely used in rating psychological distress among general and clinical populations from different cultural backgrounds around the globe.  To our knowledge, however, researchers have not yet validated a translated version in Arabic.  This is problematic considering there are approximately 392 million people who inhabit the 22 Arabic-speaking countries in the world (World Bank, 2016).  One emerging nation, Occupied Palestinian Territories (OPT), presents a unique context for validating these measures, as OPT has witnessed military occupation, war, social unemployment, and refugee displacement for more than fifty years.  Having a validated scale to assess mental health is essential to service delivery in this part of the world.  Thus, the objective of this study was to test reliability and validity of two Arabic versions of this instrument—10-item (K10) and 6-item (K6) — among public sector employees in OPT.

Methods. As part of a larger study on life satisfaction, researchers collected data from 237 Palestinian social workers in the summer of 2016.  The survey included several mental health measures, including the K10, which were translated from English to Arabic by a language expert.  After cleaning the data, we tested internal reliability by measuring the Cronbach’s alpha coefficient.  Next, we assessed factor structure using variance-covariance matrix with maximum likelihood (ML) estimation.  CFA analyses were performed to examine three competing models: unidimensional K10 model, unidimensional K6 model and two-factor K6 model.  Fit indices and parameter estimates were reported.  Finally, criterion validity was examined by assessing correlations with Generalized Anxiety Disorder (GAD-7) and Somatic Symptoms Scale (SSS-8).

 

Results. The mean Kessler score was 21.8 (SD=6.7), which indicates mild distress.  Scale reliability analysis showed satisfactory results on both K6 and K10, respectively (α =  .81 (K6) and .88 (K10)).  The two-factor K6 model (i.e., depression and anxiety) provided a better model fit than either the unidimensional K6 or K10 models. The χ2 statistics for the two-factor K6 model was not significant, indicating good fit.  SRMR (.0244) and RMSEA (.040) were below the cutoff point of .05, suggesting satisfactory model fit.  The CFI value (.996) indicated excellent fit.  Overall, fit statistics indicated that the instrument performed very well.  Additionally, the two-factor K6 had moderate factorial correlations (r=.60, p<.001).  Moreover, K6 has high criterion validity with GAD-7 (r=.08, p<.001) and SSS-8 (r=.62, p<.001).

 

Implications. Results indicate that the translated version of the two-factor K6 scale is a valid and reliable tool for assessing mental distress among Arabic-speaking populations.  The instrument will be especially useful with Arab populations in nations facing military conflict, severe poverty, and fragile governance, conditions likely to elevate rates of depression and anxiety.  Clinical cutpoints provide a diagnostic reference to connect individuals at risk for psychological symptoms to mental health services.  In the long run, use of this translated, validated K6 scale may promote equal opportunity and justice in mental health service delivery in the Middle East.