Investigated the validity of the Arabic 30- and 12-item versions of the General Health Questionnaire (GHQ-30 and GHQ-12) in 157 16-65 yr old primary health care patients in United Arab Emirates. Ss completed the GHQ-30, GHQ-12 and a standardised clinical interview using the Clinical Interview Schedule (CIS). The receiver operating characteristic analyses of the Arabic versions of both GHQ-30 and GHQ-12 were compared and their correlation coefficients with the weighted total score of the CIS and the overall severity rate were estimated. Discriminant functional analysis was also performed to assess the discriminatory powers of both the GHQ-30 and GHQ-12. Results show the best cut-off point of the GHQ-30, that balances between sensitivity and specificity, was 31/32 with a sensitivity of 0.83 and specificity of 0.83, and that of the GHQ-12 was 12/13 with a sensitivity of 0.83 and specificity 0.80. The total discriminatory powers of the GHQ-30 and GHQ-12 were approximately 93 and 86%, respectively. Both versions of the GHQ were found to have a significant concurrent validity.