The authenticity of the Chinese translation of the Hospital Anxiety and Depression Scale (HAD) was tested in a sample of medical students. The Chinese version demonstrated good agreement with the English original. There was a large difference between the mean anxiety and depression subscores. Factor analysis consistently yielded three factors, suggesting the existence of a somatic factor. It is suggested that a common cut-off point for the subscales of the HAD scale is not advisable and a multidimensional model for mood disorders is more appropriate in a cross-cultural context.