The Karolinska Scales of Personality (KSP) were used for the description of self-reported personality characteristics in pain patients. The patients were a selected group, suffering from chronic non-malignant pain mainly of neurological origin, and had in all cases an established somatic diagnosis. In comparison with normal controls, the pain patients differed only with respect to two KSP scales, showing significantly more negative childhood experiences and less Inhibition of Aggression than the normals. In contrast, there were several and pronounced differences between the pain patients and two psychiatric groups with depressive disorders, one with and one without pain. Thus this group of chronic pain patients did not show the personality characteristics predisposing for depressive disorders. Furthermore, the KSP personality factors were found to be related to variables describing consequences of the pain, but the power of the KSP to predict the outcome of the pain treatment appeared to be limited. The KSP appears to have several advantages in comparison, e.g., with the Minnesota Multiphasic Personality Inventory, the most frequently used method for the assessment of chronic pain patients. Unlike the MMPI, the KSP reflects aspects of personality and is not primarily a measure of psychopathology. In addition, the KSP is more up-to-date and its completion is less time-consuming. The usefulness of the KSP for the assessment of chronic pain patients deserves therefore further examination.