Using cases of dissociative disorder (n=50) and other DSM-IV diagnoses (n=50), a somatoform dissociation self-report questionnaire was developed and its capacity to function as a screening device for dissociative disorders was analysed. A list of 75 items was constructed which, according to clinical experience and expert judgement, could reflect instances of somatoform dissociation. Statistical analyses revealed the 20 best discriminating items. Stepwise forward logistic analysis detected five items which, as a group, provided optimal discrimination between the two groups. At an estimated prevalence rate of dissociative disorders of 10% among psychiatric patients the sensitivity would be 94%, the specificity would be 96%, the positive predictive value would be 72%, and the negative predictive value would be 99%. Cross-validation in an independent sample (n=33/42) largely corroborated the initial findings. The SDQ-5 can be used as a brief screening device for dissociative disorders.