Resolving the Latent Structure of Borderline Personality Disorder: Predicting Symptoms in Daily Life
Restricted (Penn State Only)
- Author:
- Johnson, Benjamin Norman
- Graduate Program:
- Psychology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 15, 2020
- Committee Members:
- Kenneth Levy, Chair & Dissertation Advisor
Michael Hallquist, Major Field Member
Kristin Buss (she/her), Major Field Member
Peter Molenaar, Outside Unit & Field Member
Kristin Buss (She/Her), Program Head/Chair - Keywords:
- personality disorder
phenotype
classification
assessment
diagnosis
individualized treatment
ecological momentary assessment
ambulatory assessment
mixture modeling
latent variable modeling
borderline personality disorder - Abstract:
- Objective. BPD is associated with high rates of emergency room visits and costly healthcare service utilization, affecting 10-20% of psychiatric outpatients and 20-40% of psychiatric inpatients. BPD also contributes to impaired social and occupational functioning and significant suicide risk, with 1 in 10 individuals with BPD completing suicide. Recent research has aimed to better understand the hybrid dimensional-categorical latent structure of BPD in order to improve assessment and potentially enhance treatment effectiveness via personalized care for individuals among the BPD phenotype. However, no research has simultaneously included: a) a sufficiently large patient sample; b) ecologically sound validation of results; and c) use of appropriate statistical techniques. Method. The present study aims to both confirm and validate the hybrid dimensional-categorical structure of BPD in an outpatient sample (N = 376) completing semi-structured psychodiagnostic evaluations. We utilized a model comparison approach, comparing purely dimensional, purely categorical, and hybrid dimensional-categorical models of BPD and then validated the optimal model using symptom data assessed from a subset of patients (n = 57) assessed on smartphones over 21-days in everyday life. Results. Results suggested a hybrid factor mixture model best fit the data, indicating a strong single dimension of BPD and simultaneously a discrete division between “phenotypic” individuals with elevated BPD symptoms (particularly impulsivity, anger, self-harm, and chaotic relationships) and individuals low in BPD symptoms. The dimensional component of this model, rather than the categorical component, significantly predicted daily manifestations of symptoms, including negative affect, affective lability, and self-harm urges. Furthermore, dimensional model parameters outperformed both severity scores as assessed by the International Personality Disorder Examination and both clinician-given and interview-based BPD diagnosis. Conclusion. Our study extends recent hybrid modeling research exploring the latent structure of BPD and provides further evidence that BPD consists of both dimensional and categorical aspects. The dimensionality underlying BPD is a strong candidate for predicting displays of symptoms in everyday life, though further exploration of the BPD phenotype (which is not adequately captured by the DSM polythetic diagnostic system) may also have important implications for personalized treatment. We encourage researchers to utilize hybrid latent variable models to better understand and capture taxonic aspects of BPD and other disorders that exist along a continuum of severity. We also provide suggestions for clinicians to enhance the utility of diagnostic scoring algorithms built on the DSM framework that ignore heterogeneity in the ways in which BPD items function, both in terms of how much they reflect the underlying BPD construct and in terms of how well they identify individuals comprising the BPD phenotype.