Good enough assessment: A comparison of brief screener and detailed interview assessments of borderline personality disorder symptoms in a clinical research setting.

Structured clinical interviews are the gold standard for identifying borderline personality disorder (BPD) in research. No self-report instrument has been validated as a replacement tool for BPD diagnosis. However, clinical interviews can be time and cost-intensive, limiting their utility as appropriate screening tools. Also, data are lacking regarding the level of training and education interviewers need to use BPD screening instruments effectively. The goal of this project was to clarify best practices for brief screening interviews to identify potential research participants with BPD. We analyzed data from research participants (N = 366) who completed both a brief phone screen and clinical interview assessments of BPD symptom experience. A subset (n = 120) also completed BPD-focused self-report questionnaires. Agreement between the full-length interview and the brief phone screen assessment was moderate (Cohen's κ = .63), and phone screens were more sensitive than specific (sensitivity = .92, specificity = .73). Fitting a logistic regression model to observations did not increase specificity without lowering sensitivity. Using a decision tree to shorten the phone screen process did not increase misclassifications, remained ∼18%; χ²(1) = 0.13, p = .72. Self-report instruments were not highly correlated with clinical interview scores and were more specific than sensitive, suggesting they are poor screening tools (r values .1-.28). Finally, screener education level did not predict agreement between assessments. Our findings suggest that full-length clinical interview assessments remain necessary for accurate identification of BPD in research contexts. However, phone screens, even if shortened and conducted by nonclinically trained personnel, can serve as a reliable first step in the screening process. (PsycInfo Database Record (c) 2026 APA, all rights reserved).