The validity of the French version of the GHQ-28 and PSYDIS in a community sample of 20 year olds in Switzerland.

The 28-item GHQ (General Health Questionnaire) and the 43-item PSYDIS (Psychic Distress) were translated into French and administered to a community sample of 233 young adults. Of the two tests, the GHQ was shown to have the better correlation with the clinical assessment. For PSYDIS, specificity was 72.8, sensitivity was 64.2, but the misclassification rate rose to 29.2. The GHQ misclassified 18.5% of the respondents at the best cut-off point (5/6), with a specificity of 91.1 and a sensitivity of 49.1. The corrections C-GHQ (Goodchild and Duncan-Jones method of scoring) and simple Likert scoring improved sensitivity, although this was still unsatisfactory with males. Considering the lability of troubles in a young population, the GHQ should be applied with caution for epidemiological purposes.