Comparing Methods to Model Stability and Change in Personality and its Pathology

人格障碍（PD）是一种严重的、使人衰弱的、昂贵的、通常是致命的疾病，在流行病学样本中非常普遍（约10%，Lenzenweger, 2008）。尽管PD的严重性和高昂的公共卫生成本，该领域在如何最好地量化规范人格与PD之间的关系方面缺乏共识。这种缺乏共识导致了PD的基本定义问题，包括疾病的结构、病程和风险/保护因素。大多数PD的诊断标准包括人际适应不良行为（如自恋型PD的人际剥削性）。代理和交流的人际特征已被确定为开发PD的维度模型的潜在有用。此外，人格的人际概念化对PD的联系非常有用，因为它们对心理治疗干预有直接的影响。此外，新兴研究表明，pd并不像以前认为的那样是慢性的，而是在时间上表现出相当大的变异性。

Personality disorders (PD) are serious, debilitating, costly, and oftentimes fatal disorders that are highly prevalent in epidemiological samples (~10 percent, Lenzenweger, 2008). Despite the severity and high public health cost of PD, the field suffers from a lack of consensus on how to best quantify the relationship between normative personality and PD. This lack of consensus has led to problems in basic definitional issues of PD, including the structure, course, and risk/protective factors of the disorders. The criteria of most PD diagnoses include maladaptive interpersonal behaviors (e.g., Interpersonal Exploitativeness in Narcissistic PD). The interpersonal traits of Agency and Communion have been identified as potentially useful for developing a dimensional model of PD. Further, interpersonal conceptualizations of personality are highly useful for linking with PD due to their immediate implications for psychotherapeutic interventions. Additionally, emerging research has shown that PDs are not as chronic as once thought, showing considerable variability across time. The current research aims to meet the strategic planning goals of the National Institute of Mental Health by using state-of-the-art methodology to articulate emerging conceptions of PD that account for simultaneous change in interpersonal behavior and PD over time, and non-normality in the distribution of PD. Specifically, we propose using Latent Growth Curve Models and associated multivariate extensions to examine change and distribution of PD in a multi-wave prospective longitudinal study. This research shall test the hypotheses that 1) there is long-term within-person variability in interpersonal style across time; 2) within-person variations in interpersonal style are predictive of concurrent within-person variations in PD symptoms; and 3) advanced distributional assumptions (i.e., Zero Inflated Poisson/Negative Binomial) provide a better fit for models of the relationship between PD and normative personality structure. These methods have the potential to inform classification questions in ways that have not yet been employed in any longitudinal PD research to date. Results of the proposed study have implications for the ongoing efforts to establish the appropriate definition, diagnosis, and treatment of PD. The results will inform the best approach to conceptualize classification and assessment of PD. In addition, the longitudinal nature of these studies will identify the interpersonal maintenance factors of PD, and thereby determine the best targets for interventions.