Molinari V, Kunik M E, Mulsant B, Rifai A H
Houston Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas, USA.
Am J Geriatr Psychiatry. 1998 Spring;6(2):136-44.
Two psychiatrists simultaneously administered the Personality Assessment Form (PAF) to 20 depressed geropsychiatric inpatients and to 20 of their relatives (informants). A social worker also completed PAF ratings, and a consensus conference independently assigned an Axis II diagnosis. For patient interviews, categorical and dimensional interrater reliability indices for cluster and total personality disorders (PDs) were generally marginal. For informant interviews, categorical interrater reliability indices for cluster and total PDs were also marginal, but dimensional reliability was high. Consistent with previous research, agreement between categorical scores yielded by the four assessment methods was poor, but agreement between dimensional scores was generally better. It appears that patients, family members, and staff maintain different perspectives on patients' personality that, if overlapping, may provide useful diagnostic data.
两位精神科医生同时对20名老年精神科抑郁症住院患者及其20名亲属（ informant）进行了人格评估表（PAF）测评。一名社会工作者也完成了PAF评分，并且通过共识会议独立给出了轴II诊断。对于患者访谈，关于人格障碍（PDs）聚类和总体的类别及维度评分者间信度指标总体上处于临界状态。对于 informant访谈，关于PDs聚类和总体的类别评分者间信度指标同样处于临界状态，但维度信度较高。与之前的研究一致，四种评估方法得出的类别分数之间的一致性较差，但维度分数之间的一致性总体上更好。看来患者、家庭成员和工作人员对患者的人格持有不同观点，这些观点即便有重叠，也可能提供有用的诊断数据。
