Borderline Personality Questionnaire (BPQ)

The BPQ consists of 9 subscales corresponding to the DSM criteria for borderline personality disorder (BPD).

• Impulsivity: Measures the tendency to act rashly without consideration of consequences.
• Affective Instability: Assesses rapid, intense mood fluctuations.
• Abandonment: Evaluates fear and intolerance of being alone or rejected.
• Relationships: Measures pattern of unstable and intense interpersonal relationships.
• Self-Image: Assesses unstable sense of self and identity disturbance.
• Suicide/Self-Mutilation: Evaluates suicidal behaviors and self-harm tendencies.
• Emptiness: Measures chronic feelings of emptiness and lack of purpose.
• Intense Anger: Assesses difficulty controlling anger and frequent angry outbursts.
• Quasi-Psychotic States: Evaluates transient stress-related paranoid thoughts or dissociative symptoms.

                                                                                                                Example BPQ Items

The BPQ is a valuable tool for clinicians to screen for borderline personality traits and identify particular symptom clusters causing the client the most difficulty. The BPQ provides an impartial tool to aid diagnosis, and can aid clinicians and clients in discussing BPD symptoms with objective and structured symptom information.

The BPQ total score (range 0 to 80) reflects a greater overall presence of traits associated with borderline personality as defined by the DSM criteria.

Higher scores on each subscale indicate a stronger presence of the particular trait or symptom related to borderline personality disorder:

• Impulsivity (items: 1, 10, 26, 34, 42, 57, 64, 68, 71): Reflect a greater tendency to act spontaneously, recklessly and without fully considering the potential outcomes.
• Affective Instability (items: 2, 11, 19, 27, 35, 43, 49, 58, 65, 72): Indicate more frequent and intense mood changes (e.g., intense episodic dysphoria, irritability, or anxiety).
• Abandonment (items: 3, 12, 20, 28, 44, 50, 59, 66, 73, 78): Heightened concern about being abandoned or rejected, associated with attachment anxiety.
• Relationships (items: 4, 13, 21, 29, 36, 45, 51, 60): Intense and unstable interpersonal relationships, such as a tendency for idealisation and devaluation.
• Self-Image (items: 5, 14, 37, 46, 52, 61, 67, 70, 74): Identity disturbances such as an unstable sense of self.
• Suicide/Self-Mutilation (items: 6, 15, 22, 30, 38, 53, 75): Thoughts or behaviours related to self-harm or suicide, such as gestures, threats or a history of self-harm or suicide attempts.
• Emptiness (items: 7, 16, 23, 31, 39, 54, 62, 69, 76, 79): Persistent feelings of emptiness, sadness, or meaninglessness.
• Intense Anger (items: 8, 17, 24, 32, 40, 47, 55, 63, 77, 80): Challenges in managing anger and more frequent expressions of hostility (e.g., displays of temper, constant anger, recurrent physical fights).
• Quasi-Psychotic States (items: 9, 18, 25, 33, 41, 48, 56): Experiences of transient paranoid thoughts, hallucinations, or severe dissociative symptoms such as derealisation. Quasi-psychotic symptoms are typically triggered by stress and last from a few minutes to a few hours, rarely extending beyond a day.

A percentile is used to contextualise a respondent’s score compared to the normative community sample (Poreh et al., 2003). A percentile of 50 suggests typical (and healthy) patterns of responding, whereas a percentile of 99 indicates that the respondent scores higher than 99 percent of individuals, indicating severe symptoms consistent with borderline personality. The BPQ total scores are categorised using the following qualitative descriptors:

• Subclinical: 0 to 29, percentiles 0 to 75th
• Mild: 30 to 34, percentiles 75th to 85th
• Moderate: 35 to 42, percentiles 85th to 95th
• Severe: 43 to 80, percentiles 95th and over 

Subscale scores are presented as the percent of questions where the respondent’s response was consistent with borderline personality disorder symptoms (between 0 and 100%). The line on the subscale graph indicates the point that scores are considered to be consistent with BPD. Scores above this line are in the top quartile of responses compared the general population.

For multiple administrations, the line graph visually tracks the respondent’s total scores across sessions. A meaningful change (~ 0.5 SD) in score is defined as an increase or decrease of at least 7 or more points for the total score. This criterion is based on the Minimally Important Difference (MID) calculation. Such changes indicate meaningful improvement or reduction in symptoms, while a change of less than the specified points suggests no meaningful change in symptom severity between assessments.

Overall, individuals with higher scores on the BPQ, indicating more borderline personality traits, often experience more intense emotions such as impulsivity, anxiety, and have significant interpersonal difficulties. They are likely to have low life satisfaction, unconventional thinking, and high rates of self-harm and suicide (Gradinaru et al., 2024). Borderline personality disorder share symptoms with ADHD, eating disorders, substance use, major depressive disorder, and post-traumatic stress disorder. Therefore, considering these as differential diagnoses is important, given the BPQ does not directly assess for these conditions.

In developing the BPQ, items were created based on the nine DSM criteria for BPD, adapting from existing measures like the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II) and generating new items as needed. Items that did not enhance internal consistency or were confusing were removed, resulting in an 80-item dichotomous scale (True/False) (Poreh et al., 2006). While the BPQ was developed with reference to the DSM-IV, it remains relevant in the later version of the manual (i.e. DSM-5-TR).

Factor analysis of the BPQ was performed on three distinct samples: Australian (Mean age = 19.32, SD = 2.91, 76% female), English (Mean age = 20.10, SD = 1.16, 58% female), and American (Mean age = 21.73, SD = 5.81, 67% female). The results consistently identified a single primary factor across studies, indicating that the questionnaire predominantly measures a unified aspect of borderline personality traits (Poreh et al., 2006). Convergent validity for the BPQ was established, with high correlations between the BPQ and the Minnesota Multiphasic Personality Inventory (MMPI-II) -Borderline Personality Disorder scale (r = .85) in the American sample, and with the Borderline Personality Scale (r = .72 and .78) in the English and Australian samples respectively.

The BPQ demonstrated satisfactory discriminant validity, with moderate correlations between the BPQ sum score and the MMPI-2 Schizotypal Disorder Scale (r = .48) and the Schizotypal Personality Questionnaire (r = .45) (Poreh et al., 2006). For criterion validity, the 12 highest BPQ scorers and 15 randomly selected lower scorers were interviewed by phone using the Diagnostic Interview for Borderline Patients (DIBP). The findings showed a significant correlation (τ = .51) between BPQ results and clinical diagnoses (Poreh et al., 2006).

A study by Chanen et al. (2008) compared four tools for screening BPD in young people: the McLean Screening Instrument for BPD (MSI-BPD), the Borderline Personality Questionnaire (BPQ), BPD items from the International Personality Disorder Examination Screening Questionnaire, and BPD items from the SCID-II Personality Questionnaire. The BPQ was found to have the best psychometric properties, with the best overall sensitivity, specificity, and reliability. It also showed the highest consistency in results over time (Chanen et al., 2008).

For the overall scale, Cronbach’s alpha was 0.89, and the test-retest correlation was 0.67 (Ceylan et al., 2017). The test-retest correlation coefficients for the BPQ subscales over a period of 28 days are as follows:

• Impulsivity: 0.50
• Affective Instability: 0.77
• Abandonment: 0.40
• Relationships: 0.68
• Self-Image: 0.72
• Suicide/Self-Mutilation: 0.48
• Emptiness: 0.73
• Intense Anger: 0.74
• Quasi-Psychotic States: 0.62 (Ceylan et al., 2017).

The following means and standard deviations are derived from combining three samples (N= 558)—with a pooled Mean age = 20.41, pooled SD = 3.72, 66% female)— and calculating the pooled means and standard deviations across these groups (Poreh et al., 2006):

• Total Score: M = 21.02, SD = 12.96
• Impulsivity: M = 2.47, SD = 2.22
• Affective Instability: M = 3.70, SD = 2.95
• Abandonment: M = 1.34, SD = 1.54
• Relationships: M = 2.56, SD = 2.49
• Self Image: M = 2.84, SD = 2.36
• Suicide/Self Mutilation: M = 1.05, SD = 1.57
• Emptiness: M = 2.28, SD = 2.26
• Intense Anger: M = 2.97, SD = 2.95
• Quasi-Psychotic States: M = 2.08, SD = 1.81

The above data was used to compute percentiles and severity descriptors for the total BPQ score by NovoPsych, which assigned each qualitative descriptor to a specific percentile range. These ranges were derived from the combined community samples analysed by Poreh et al. (2006). The total scores are categorised with the following qualitative descriptors:

• Subclinical: 0 to 29, percentiles 0 to 75th
• Mild: 30 to 34, percentiles 75th to 85th
• Moderate: 35 to 42, percentiles 85th to 95th
• Severe: 43 to 80, percentiles 95th and over

For most subscales, average scores above the 75th percentile are considered elevated. However, on the Suicide/Self-Mutilation scale, endorsement of any items is considered elevated.

Supplementary Material

The percentile tables below illustrate how total (see Table 1) and average scores (see Table 2) compare to the normative sample of individuals in the community (Poreh et al., 2006). Each score is accompanied by a corresponding percentile, indicating the percentage of individuals who scored the same or lower. For instance, a score of 4 corresponds to the 75th percentile on the Impulsivity subscale (Table 1), signifying that 75% of the sample had a score of 4 or lower. These graphs are instrumental in contextualising an individual’s total and average scores on each subscale, providing a clearer understanding of their standing relative to other individuals in the community.

Table 1

Table 2

Graph 1

The below graph (Graph 1) and table (Table 3) show the relationship between total scores and percentiles for the Borderline Personality Questionnaire. The S-shaped curve indicates that moderate scores are most common, while very low and very high scores are rarer.

Table 3

Assess your patients with the Borderline Personality Questionnaire (BPQ) using a NovoPsych account

• For Psychologists & Mental Health Clinicians
• Send Assessments to Patient’s Phone
• Receive Comprehensive Reports
• Access Over 100 Validated Assessments​
• Instant Psychometric Scoring
• Track Symptoms​
• Inform Treatment​