The aim of the present study was to evaluate the combined test-retest and interrater reliability of different psychiatric lifetime diagnoses yielded in the course of a family study in elderly patients and controls. The following interviews and questionnaires were used in combination: the Composite International Diagnostic Interview (CIDI), the Structured Interview for the Diagnosis of Dementia of the Alzheimer Type, Multi-infarct Dementia and Dementias of Other Aetiology (SI-DAM), the General Health Questionnaire (GHQ-12) and questionnaires for neurasthenia and recurrent brief depression (RBD). Depressive and dementia disorders can be diagnosed with good reliability in a family study setting with the use of these instruments. The diagnoses of phobic disorders, neurasthenia, RBD, subthreshold RBD and psychiatric caseness as indicated by GHQ-12 scores were less reliable in this setting and are therefore less suitable for use in family studies.

The study design of the international multicentric field trial on ICD-10, chapter V (F) "Mental, Behavioural and Developmental Disorders: Clinical Descriptions and Diagnostic Guidelines" (1987 draft) including some modifications for the study in German-speaking countries, is described. The various stages of the field trial, instruments used, and the data base, are presented.

The summary of the quantitative results of the ICD-10 field trial in German-speaking countries shows a adequate acceptance of ICD-10. The figures for interrater reliability for the diagnostic categories are adequate for most categories considering the comparatively low level of familiarity with the new system. The high degree of overlap between the diagnostic categories in ICD-9 and ICD-10 is an indication of the validity of ICD-10. The major source of disagreement is the relocation of several diagnostic categories to different sections.

On the basis of the results of the field trial, the most important classificatory innovations of ICD-10, together with their advantages and disadvantages are described. The initial good acceptance of the new system could be further improved by structural modifications, such as a uniform and systematic description of the individual diagnostic categories. Criticism of content was focused on affective and neurotic disorders and adult personality disorders. When it is introduced, the psychiatric chapter of ICD-10 will surpass most of the hitherto existing psychiatric classification systems in size, differentiation and international testing.

The AMDP scale for dissociation and conversion (AMDP-DK) represents an observer-rated scale for the assessment of dissociative and conversive phenomena, which closes the gap between available self-report questionnaires and time-consuming structured interviews with their respective disadvantages. The instrument comprises 30 operationalized items, which are rated according to the AMDP rules, and which are theoretically-driven divided into the subscales dissociation (15 items), conversion (9 items) and a "formal" dimension (6 items) to cover associated phenomena. In a sample of psychiatric inpatients (n = 73) the covered symptoms showed a great variety of frequency and intensity. The AMDP-DK and its subscales had good internal consistency (Cronbachs alpha between 0.76 and 0.85). A factor-analysis mostly replicated the theoretically constructed subscales. The convergent and divergent validity of the scale was good, as well as its ability to discriminate between high- and low-dissociating patients. The findings are discussed with regard to the psychometric properties and the relevance for the theories of dissociation and conversion.