Diagnostik der Beschwerdenvalidität von psychischen Störungen in der sozialmedizinischen Begutachtung

Aims: The major aim of this thesis is to determine the need of symptom validation in the socio-medical assessment and to construct a screening for the detection of negative response bias by mental disorders. Additionally, the effects of being instructed before completing a questionnaire and the motivation for obtaining a pension are examined in claimants for disability pension. Methods: On the basis of a systematic review German symptom validation questionnaires determined. Study I proved the suitability of the symptom validation tests for the detection of negative response bias in mental disorders. On the one hand an analogue study design is used, to detect any differences in the presentation of complaints between instructed malingerers, a control group, and inpatients with major depression or pain disorder with depressive symptoms in clinical symptom scales (see Study II). On the other hand, the analogue study design (see Study IV) is used to construct and validate the new screening for the detection of negative response bias by mental disorders. Using a randomized control group design, the effects of giving instructions in advance and the motivation obtaining a pension on symptom presentation are analyzed (see study III). Results: Instructed malingerers tend to provide extreme responses and overgeneralization in symptom scales. Based on these findings a new symptom validation test was constructed, which has good psychometric characteristics for reliability and validity. Instructions produce negative response bias in symptom scales. In addition, claimants for disability pension tend to negative response bias when they think to recover through their pension. Conclusion: Although the developed symptom validation test provides sufficient psychometric characteristics, the validation of the test should be continued in other more practically oriented in a known-group-Design, for example. The occurrence of any negative response bias should be scrutinized for possible causes since instructions and different motives can produce invalid symptoms.