Borderline Symptom List (BSL-23)

1. Abstract

The Borderline Symptom List (BSL-23) is a standardized, self-report psychometric instrument designed to assess the severity, frequency, and phenomenological burden of subjective symptoms characteristically experienced by individuals diagnosed with Borderline Personality Disorder (BPD). Developed as an abbreviated, psychometrically robust variant of the comprehensive 95-item Borderline Symptom List (BSL-95), the BSL-23 captures core domains of borderline psychopathology—including extreme affective instability, profound identity disturbance, chronic feelings of emptiness, self-depreciation, vulnerability to stress-induced dissociation, interpersonal hypersensitivity, and self-injurious urges—within a pragmatic 23-item operational architecture. Items are appraised on a 5-point Likert scale anchored from 0 (“not at all”) to 4 (“very strong”), reflecting the respondent’s clinical state over the preceding week. The inventory demonstrates exemplary psychometric architecture across international inpatient, outpatient, and community populations: internal consistency coefficients routinely exceed conventional benchmarks (Cronbach’s alpha α = .93 to .97), accompanied by exceptional convergent associations with the full BSL-95 parent scale (r ≥ .95). Factor analytic investigations confirm a dominant general factor reflecting overall borderline symptom distress, supporting unidimensional scoring. Longitudinal validation reveals pronounced sensitivity to therapeutic change, establishing the scale as an indispensable, low-burden metric for tracking recovery trajectories, evaluating evidence-based modalities such as Dialectical Behavior Therapy (DBT), and standardizing symptom tracking in rigorous psychiatric research.

2. Keywords

Borderline Symptom List, BSL-23, Borderline Personality Disorder, Emotion Dysregulation, Dialectical Behavior Therapy, Psychometrics, Self-Report Symptom Severity, Clinical Assessment, Treatment Outcome Monitoring, Affective Instability

3. Authors

The Borderline Symptom List-23 was derived and validated through collaborative clinical research spearheaded by Martin Bohus, M.D., and colleagues affiliated with the Central Institute of Mental Health (Zentralinstitut für Seelische Gesundheit, ZI) in Mannheim, Germany, and the Department of Psychosomatic Medicine and Psychotherapy at Heidelberg University. The principal development team includes:

• Martin Bohus, M.D. — Chair of Psychosomatic Medicine and Psychotherapy, Central Institute of Mental Health, Faculty of Medicine Mannheim, Heidelberg University, Germany.
• Nikolaus Kleindienst, Ph.D. — Senior Biostatistician and Methodologist, Central Institute of Mental Health, Mannheim, Germany.
• Monika F. Limberger, Dipl.-Psych. — Clinical Psychologist and Researcher, Central Institute of Mental Health, Mannheim, Germany.
• Rolf-Dieter Stieglitz, Ph.D. — Professor of Clinical Psychology and Psychiatry, Department of Psychology, University of Basel, Switzerland.
• Martina Domsalla, Dipl.-Psych. — Department of Psychosomatic Medicine, Central Institute of Mental Health, Mannheim, Germany.
• Alexander L. Chapman, Ph.D., R.Psych. — Professor of Psychology, Department of Psychology, Simon Fraser University, Burnaby, British Columbia, Canada.
• Melanie Steil, Dipl.-Psych. — Department of Psychosomatic Medicine, Central Institute of Mental Health, Mannheim, Germany.
• Claudia Philipsen, M.D. — Department of Psychiatry and Psychotherapy, University Hospital Freiburg, Germany.
• Markus Wolf, Ph.D. — Department of Psychotherapy and Systems Research, University of Zurich, Switzerland.

4. Purpose

The primary clinical and empirical imperative behind the engineering of the Borderline Symptom List-23 was to resolve a persistent logistical challenge in psychiatric measurement: balancing assessment depth with administrative brevity. Borderline Personality Disorder is an exceptionally debilitating psychiatric syndrome defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) and the International Classification of Diseases (ICD-11) by pervasive instability in interpersonal relationships, self-image, and affects, alongside marked impulsivity. While structured clinical interviews such as the Structured Clinical Interview for DSM-IV/5 Axis II Disorders (SCID-II/SCID-5-PD) or the Diagnostic Interview for Borderlines-Revised (DIB-R) represent the diagnostic reference standards, their lengthy administration makes them unsuitable for frequent, repeated symptom monitoring.

In response to this diagnostic hurdle, Bohus and colleagues initially introduced the 95-item Borderline Symptom List (BSL-95) in 2007. Anchored in both formal DSM criteria and the subjective reports of borderline patients, the BSL-95 demonstrated unmatched granularity and content validity. Nevertheless, administering a 95-item scale in high-frequency monitoring regimens—such as weekly psychotherapy check-ins, day-to-day ecological momentary assessments, or ongoing treatment trials—imposed prohibitive respondent burden. Patients afflicted with acute borderline pathology frequently suffer from profound emotional distress, attentional fragmentation, cognitive fatigue, and low frustration tolerance, which can elevate attrition and non-compliance when facing lengthy inventories.

The BSL-23 was systematically crafted to eliminate these logistical barriers without sacrificing psychometric integrity. Its operational purposes encompass:

• Tracking Longitudinal Treatment Trajectories: The scale enables clinicians and researchers to capture subtle fluctuations in borderline symptom distress across brief intervals, making it optimal for measuring pre-to-post gains in structured modalities such as Dialectical Behavior Therapy, Mentalization-Based Treatment (MBT), Transference-Focused Psychotherapy (TFP), and Schema Therapy.
• Minimizing Patient and Administrative Burden: Requiring an average completion time of merely three to five minutes, the BSL-23 substantially attenuates cognitive fatigue and enhances compliance across acute psychiatric inpatient units, crisis intervention settings, and outpatient clinics.
• Evaluating Severity Rather Than Categorical Diagnosis: The BSL-23 does not classify, categorize, or assign a formal psychiatric diagnosis of BPD; instead, it quantifies current symptom severity along a continuous dimensional spectrum, directly aligning with contemporary dimensional models of personality pathology (such as the Alternative Model for Personality Disorders in DSM-5 Section III).
• Standardizing Research Benchmarks: Serving as a continuous dependent variable in randomized controlled trials (RCTs), the BSL-23 allows for precise meta-analytic comparisons of therapeutic efficacy across diverse clinical trials internationally.

5. Psychological Construct

The psychological construct operationalized by the BSL-23 is the phenomenological symptom severity of Borderline Personality Disorder. Unlike instruments designed to measure static personality traits or externalized behavioral dysregulation alone, the BSL-23 targets the immediate subjective suffering, distress, and intrapsychic turmoil intrinsic to the lived borderline experience. Rather than treating borderline pathology as an arbitrary constellation of detached behaviors, the scale conceptualizes it as an interconnected network of cognitive, affective, self-referential, and relational disturbances.

Affective Instability and Hyperarousal

Affective dysregulation forms the foundational substrate of the construct. Individuals with elevated BSL-23 scores report rapid, uncontrolled emotional shifts, enduring states of painful tension, and pervasive internal anxiety. The construct evaluates the patient’s acute perception that emotional states are intolerable, overwhelming, and impossible to calm down autonomously. Unlike standard depressive or generalized anxiety scales, the BSL-23 captures the specific quality of borderline emotional pain—often described as a burning, all-consuming internal agony accompanied by intense irritability and panic.

Identity Alteration and Chronic Emptiness

A cardinal dimension embedded within the BSL-23 construct is profound identity diffusion and self-alienation. The scale examines subjective manifestations of inner void—a visceral sensation that one lacks a core self, has no purpose, or does not truly exist. This includes pervasive self-loathing, characterized by persistent cognitions that one is fundamentally toxic, defective, contemptible, or an irremediable burden to others. Patients endorse severe feelings of shame and alienation from their own physical bodies and autobiographical narratives.

Perceptual and Dissociative Fragmentation

The construct explicitly captures the dissociative phenomena that emerge when individuals with BPD encounter severe emotional distress. Items tap into experiences of depersonalization, derealization, sensory numbing, and perceptual distortions (e.g., feeling disconnected from reality, feeling as if the world is artificial or distant, or experiencing profound mental fog). Capturing these dissociative features is vital because dissociation directly hinders cognitive processing and emotional regulation during standard psychotherapeutic interventions.

Interpersonal Vulnerability and Abandonment Anxieties

Interpersonally, the construct measures an agonizing sensitivity to perceived rejection, abandonment, and social exclusion. Elevated symptom states involve intense hypervigilance toward interpersonal cues, sudden shifts from idealization to devaluation, and a conviction of imminent isolation. The scale reflects how relational friction rapidly destabilizes self-worth and precipitates behavioral crises.

Behavioral Desperation and Impulsive Urges

Although the primary 23 items concentrate on subjective distress, the construct reflects the psychological preconditions that drive self-injurious and suicidal behaviors. The instrument measures intrusive impulses to harm oneself, acute desires to escape living, and feelings of helplessness that lead to impulsive action. A complementary, structurally separated 11-item behavioral supplement further evaluates concrete behavioral frequencies (e.g., self-harm, suicide attempts, binge eating, explosive outbursts), ensuring that intrapsychic suffering is systematically contextualized alongside overt clinical risks.

6. Theoretical Framework

The conceptual foundation of the Borderline Symptom List rests on contemporary, evidence-based theories of borderline psychopathology, most notably the Biosocial Theory formulated by Marsha M. Linehan, complemented by findings from developmental neurobiology and cognitive-affective processing models.

The Biosocial Model

Linehan’s biosocial model posits that BPD is predominantly a disorder of pervasive emotion dysregulation. According to this paradigm, borderline pathology emerges from reciprocal, transactional interactions between an innate biological vulnerability and an invalidating developmental environment. The biological predisposition is characterized by:

• High Sensitivity: An exceptionally low threshold for emotional triggers.
• High Reactivity: Intense affective responses that escalate rapidly into state hyperarousal.
• Slow Return to Baseline: Protracted periods of emotional activation, leaving the individual susceptible to cascading secondary emotional crises.

When this biologically vulnerable disposition develops within an environment that dismisses, punishes, or pathologizes emotional communication, the individual fails to acquire crucial skills for emotion regulation, distress tolerance, and impulse control. The BSL-23 directly reflects these biosocial mechanics: its items capture the subjective turmoil that occurs when baseline emotional arousal surges beyond the individual’s coping capacity.

Neurobiological Correlates of Stress and Dissociation

The theoretical framework of the BSL-23 is deeply informed by neurobiological investigations into the frontolimbic architecture of BPD. Bohus and his team integrated empirical data showing hyperreactivity in the amygdala and insula, coupled with diminished inhibitory control from prefrontal structures, including the anterior cingulate cortex (ACC) and the dorsolateral prefrontal cortex (dlPFC). Under conditions of severe psychological distress, limbic hyperactivity inhibits executive reflection and triggers stress-induced dissociation—a defense mediated by endogenous opioid and fronto-limbic disconnection circuits. The BSL-23 systematically integrates these neurobiological states by tracking stress-induced cognitive disruptions alongside raw emotional intensity.

Dimensional Psychopathology and Item Response Theory

From a psychometric modeling perspective, the BSL-23 is underpinned by modern latent trait theory. Rather than adopting the categorical approach of legacy psychiatric taxonomies, Bohus et al. conceptualized borderline symptom distress as a continuous, unipolar latent trait (θ). The transition from the 95-item instrument to the 23-item short form was guided by rigorous mathematical reduction: investigators identified the items exhibiting the highest item-total correlations, maximum factor loadings on the overarching general severity factor, and optimal discriminating parameters across various intervals of symptom severity. The resultant short form preserves the theoretical richness of the parent model while maximizing measurement efficiency.

7. Validity

The Borderline Symptom List-23 has undergone extensive international psychometric validation across a diverse range of psychiatric settings, consistently demonstrating exceptional construct, convergent, discriminant, and predictive validity.

Convergent Validity

Convergent validity has been established by evaluating correlations between the BSL-23 and both its parent instrument (BSL-95) and other gold-standard indices of psychological distress and personality pathology:

• Equivalence with BSL-95: Bohus et al. (2009) demonstrated that the total score of the BSL-23 correlates near-perfectly with the 95-item parent version, with correlation coefficients consistently ranging from r = .958 to .963 across clinical validation cohorts. This confirms that the short form captures the full construct without meaningful loss of clinical information.
• Beck Depression Inventory (BDI / BDI-II): The BSL-23 exhibits strong, expected correlations with depressive severity, typically ranging between r = .70 and .84. This strong association reflects the pervasive negative affectivity, helplessness, and dysphoria shared between borderline and major depressive states, while still retaining construct distinction.
• Symptom Checklist-90-Revised (SCL-90-R) / Brief Symptom Inventory (BSI): Correlations with the Global Severity Index (GSI) of the SCL-90-R are robust (r ≈ .75 to .83), affirming the scale’s alignment with general psychiatric distress.
• State-Trait Anxiety Inventory (STAI): Strong convergent associations (r > .65) underscore the chronic emotional tension and hyperarousal experienced by patients with elevated BSL scores.

Discriminant and Known-Groups Validity

Discriminant validity reflects the scale’s ability to differentiate individuals diagnosed with BPD from both healthy controls and non-BPD clinical cohorts:

• BPD vs. Healthy Controls: In the landmark validation study by Bohus et al. (2009), patients with BPD scored dramatically higher (mean overall score ≈ 2.0 to 2.4) compared to healthy individuals (mean ≈ 0.10 to 0.35), yielding enormous effect sizes (Cohen’s d > 2.0).
• BPD vs. Other Clinical Disorders: The BSL-23 successfully separates individuals with BPD from psychiatric outpatients suffering from other conditions, such as major depressive disorder, social anxiety disorder, and schizophrenia (mean differences showing medium-to-large effect sizes, d = 0.60 to 1.10). This indicates that the BSL-23 does not simply reflect generic neurosis or diffuse distress, but specifically indexes the intense intrapsychic turmoil characteristic of borderline pathology.

Cross-Cultural and Linguistic Validity

The scale has been formally translated and validated across numerous cultural environments, uniformly demonstrating invariant measurement properties. Prominent cross-cultural investigations include:

• Spanish Adaptation: Soler et al. (2013) validated the Spanish version in 204 BPD patients and other psychiatric controls, replicating the single-factor structure, finding exceptional internal consistency (α = .94), and documenting high sensitivity to clinical improvement during Dialectical Behavior Therapy.
• French Adaptation: Nicastro et al. (2016) demonstrated equivalent psychometric integrity in a clinical sample of adult BPD outpatients (α = .96), supporting strong discriminant ability against bipolar disorder and attention-deficit/hyperactivity disorder (ADHD).
• Further formal validation studies have established psychometric equivalence for Italian, Portuguese, Turkish, Dutch, and Swedish translations.

Sensitivity to Change (Predictive and Evaluative Validity)

The ultimate test of a symptom severity list lies in its responsiveness to clinical change. Bohus et al. (2009), along with numerous independent clinical trials, have shown that BSL-23 scores drop markedly during effective inpatient and outpatient psychotherapy. Standardized response means (SRM) and effect sizes across successful DBT trials consistently range from d = 0.80 to 1.30, confirming that the tool is sensitive to genuine therapeutic improvement.

8. Reliability

The reliability of the Borderline Symptom List-23 has been confirmed across dozens of psychometric investigations, consistently satisfying the rigorous standards required for both group-level research and individual clinical decision-making.

Internal Consistency

The internal consistency of the BSL-23 is exceptionally high across all evaluated language adaptations and patient populations:

• In the original derivation and cross-validation cohorts reported by Bohus et al. (2009), Cronbach’s alpha values ranged from α = .935 to .969 across different samples (inpatient, outpatient, and diagnostic cross-validation cohorts).
• Subsequent international adaptations have mirrored these findings: the Spanish version demonstrated α = .94 (Soler et al., 2013), and the French version achieved α = .96 (Nicastro et al., 2016).
• Estimates of McDonald’s omega (ω), which avoids the assumption of tau-equivalence, similarly exceed .94, confirming that the scale functions with minimal measurement error.

Test-Retest Reliability

Evaluating test-retest reliability in borderline populations requires careful methodological timing, as BPD symptomatology fluctuates in response to environmental stressors. Over short test-retest intervals (such as 24 to 48 hours in stable clinical environments where treatment has not yet commenced), the BSL-23 exhibits high temporal stability, with intraclass correlation coefficients (ICC) and Pearson correlation coefficients consistently exceeding r = .82 to .88. Over longer assessment windows (e.g., several weeks or months), stability coefficients naturally attenuate as symptoms fluctuate or respond to therapeutic interventions, reflecting the instrument’s intended dynamic sensitivity.

Item-Total Correlations

Corrected item-total correlations for all 23 items are uniformly high, typically falling between r = .50 and .80. This demonstrates that each individual item contributes meaningfully to the overarching borderline symptom severity composite without redundancy.

9. Factor Analysis

The underlying dimensionality of the BSL-23 has been scrutinized through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Original Extraction and Derivation Analysis

The development of the BSL-23 from the parent BSL-95 was conducted by extracting items that maximize psychometric variance and content validity. In the foundational validation studies by Bohus et al. (2007, 2009):

• EFA conducted on the 23 retained items revealed a dominant, primary general factor that accounted for a substantial proportion of the total item variance. Scree plot analyses and parallel analysis unambiguously supported an essentially unidimensional factor structure.
• Factor loadings for all 23 individual items on this dominant first factor were high, ranging predominantly from .55 to .84.

Confirmatory Factor Analytic (CFA) Fit Indices

Subsequent CFA investigations across multiple international cohorts have confirmed that a single-factor model represents a clinically sound, parsimonious representation of borderline distress. While highly powered clinical samples frequently elevate the chi-square statistic (χ²) to significance due to sample size, alternative fit indices consistently satisfy established psychometric criteria:

• Comparative Fit Index (CFI): Values routinely fall between .92 and .97.
• Tucker-Lewis Index (TLI): Values consistently range between .91 and .96.
• Root Mean Square Error of Approximation (RMSEA): Values typically range from .05 to .07 (90% CI [.045, .078]), indicating acceptable to good model approximation.
• Standardized Root Mean Square Residual (SRMR): Values consistently remain below .05.

Bifactor and Multidimensional Explorations

Although the original BSL-95 featured several lower-order subscales (such as Self-Perception, Affect Regulation, Self-Harm, Interpersonal Difficulties, and Dysthymia), psychometric analyses of the BSL-23 indicate that these distinct facets collapse into a single general factor. Bifactor modeling has demonstrated that the general factor explains upwards of 80% to 85% of the common variance, confirming that computing a single composite mean score is methodologically sound and clinically representative.

10. Instrument / Measurement Tool

• Instrument Name: Borderline Symptom List 23 (BSL-23).
• Instrument Type: Standardized self-report psychiatric symptom rating questionnaire.
• Format: Paper-and-pencil inventory or computerized digital administration.
• Item Count: 23 primary symptom items, accompanied by a distinct 11-item supplementary behavioral checklist.
• Target Population: Adolescents (validated down to age 14) and adults diagnosed with, or suspected of having, Borderline Personality Disorder.
• Administration Time: Approximately 3 to 5 minutes.
• Retrospective Time Frame: Evaluates symptoms and distress experienced during the past week.
• Authentic Response Scale: 5-point Likert scale:
  • 0 = Not at all
  • 1 = A little
  • 2 = Rather
  • 3 = Much
  • 4 = Very strong
• Completion Criterion: Respondents must complete at least 90% of the items (at least 21 of the 23 questions) for the protocol to be deemed valid for scoring (Mannheim, 2007).
• Scoring Algorithm:
  • Sum Score: Calculated by summing the numerical values assigned to each answered item (theoretical range: 0 to 92).
  • Mean Score: Calculated by dividing the sum score by the total number of answered items (theoretical range: 0.00 to 4.00). Using the mean score enables direct comparison with the original BSL-95.
  • Supplementary Behavioral Score: Scored independently by summing the behavioral frequency items; this score is never mixed into the primary 23-item mean score.
• Clinical Severity Benchmarks: Across European validation samples, clinical mean scores for patients with active BPD generally center around 1.8 to 2.4, while healthy control reference means fall below 0.4.

11. Permissions & Fee and Test Year

• Initial Publication Year: 2009 (original parent BSL-95 published in 2007).
• Copyright Holders: Martin Bohus, M.D., and the Central Institute of Mental Health (Zentralinstitut für Seelische Gesundheit, ZI), Mannheim, Germany.
• Access & Research Licensing: The BSL-23 is widely available for non-commercial academic research and routine clinical practice. Original scoring guidelines, normative tables, and approved translations (including English, German, Spanish, French, and other languages) are distributed through academic publications and the official repository of the Central Institute of Mental Health, Mannheim.
• Commercial Use: Commercial applications, inclusion in fee-for-service digital platforms, or pharmaceutical industry trials require formal licensing and written permission from the copyright holders.
• Contact and Repository Information: Central Institute of Mental Health, Department of Psychosomatic Medicine and Psychotherapy, J5, 68159 Mannheim, Germany. Official downloads and scoring documentation: https://www.zi-mannheim.de/.

12. References

Bohus, M., Kleindienst, N., Limberger, M. F., Stieglitz, R.-D., Domsalla, M., Chapman, A. L., Steil, M., Philipsen, A., & Wolf, M. (2009). The short version of the Borderline Symptom List (BSL-23): Development and initial data on psychometric properties. Psychopathology, 42(1), 32–39. https://doi.org/10.1159/000173701

Bohus, M., Limberger, M. F., Frank, U., Chapman, A. L., Kühler, T., & Stieglitz, R.-D. (2007). Psychometric properties of the Borderline Symptom List (BSL). Psychopathology, 40(2), 126–132. https://doi.org/10.1159/000098493

Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline personality disorder. Guilford Press.

Mannheim, P. Z. (2007). Borderline Symptom List 23 (BSL-23) analysis information. Central Institute of Mental Health. https://www.zi-mannheim.de/fileadmin/user_upload/downloads/forschung/PSM_downloads/BSL_23_english_scoring_06_2007.pdf

Nicastro, R., Prada, P., Kung, A.-L., Salamin, V., Dayer, A., Aubry, J.-M., Bohus, M., & Perroud, N. (2016). Psychometric properties of the French borderline symptom list, short form (BSL-23). Borderline Personality Disorder and Emotion Dysregulation, 3(1), 4. https://doi.org/10.1186/s40479-016-0038-0

Soler, J., Vega, D., Feliu-Soler, A., Trujols, J., Soto, Á., Elices, M., Soriano, J., Álvarez, E., Pérez, V., & Pascual, J. C. (2013). Validation of the Spanish version of the borderline symptom list, short form (BSL-23). BMC Psychiatry, 13(1), 139. https://doi.org/10.1186/1471-244X-13-139

13. Items of the Scale