Reliability and validity of the Japanese version of the World Health Organization‐Five Well‐Being Index in the context of detecting depression in diabetic patients

The present study had two aims. The first was to evaluate the reliability and the validity of the Japanese version of the World Health Organization (WHO)-Five Well-Being Index (WHO-5-J) as a brief well-being scale. The second was to examine the discriminatory validity of this test as a screening tool for current depressive episodes in diabetic patients. A sample of 129 diabetic patients completed the WHO-5-J. Of these, 65 were also interviewed by psychiatrists to assess whether they had any current depressive episodes according to DSM-IV. The internal consistency was evaluated using Cronbach's alpha, the Loevinger coefficient of homogeneity, and factor analysis. The external concurrent validity was evaluated by correlations with the external scales potentially related to subjective well-being. Discriminatory validity was evaluated using receiver operating characteristic (ROC) analysis. Cronbach's alpha and the Loevinger coefficient were estimated to be 0.89 and 0.65, respectively. A factor analysis identified only one factor. The WHO-5-J was significantly correlated with a number of major diabetic complications, depression, anxiety, and subjective quality of life. ROC analysis showed that the WHO-5-J can be used to detect a current depressive episode (area under curve: 0.92; 95% confidence interval: 0.85-0.98). A cut-off of <13 yielded the best sensitivity/specificity trade-off: sensitivity, 100%; specificity, 78%. The WHO-5-J was thus found to have a sufficient reliability and validity, indicating that it is a useful instrument for detecting current depressive episodes in diabetic patients.