NARCISSISTIC PERSONALITY DISORDER: PATTERNS, PROCESSES, AND INDICATORS OF CHANGE IN LONG-TERM PSYCHOTHERAPY

Elsa Ronningstam, Igor Weinberg
Journal of Personality Disorders
Department of Psychiatry, Harvard Medical School, Boston, Massachusetts

Table of Contents

Overall Summary

Study Background and Main Findings

This study aimed to address the long-standing challenge of facilitating change in individuals with Narcissistic Personality Disorder (NPD), a condition often resistant to treatment. Moving beyond simply asking if change is possible, the researchers sought to understand how positive change unfolds by identifying specific patterns, processes, and indicators associated with improvement during long-term psychotherapy. The research fills a gap, as systematic studies exploring the nuances of the change process itself in NPD treatment have been lacking.

To investigate this, the authors employed a qualitative methodology, analyzing detailed case reports provided by eight experienced therapists (average 22 years experience). These reports documented the treatment of eight patients (aged 20-58) initially diagnosed with NPD according to standard DSM-5 criteria, who underwent individual psychotherapy (often supplemented with other treatments like group or family therapy) for an average of 3.5 years (range 2.5-5 years) and were judged by their therapists to have shown significant improvement. Therapists retroactively assessed diagnostic criteria at the start and end of treatment, with the authors independently confirming the initial diagnoses based on the reports.

The findings were notably positive for this selected group. All eight patients demonstrated significant improvements in personality and life functioning, achieving remission from the NPD diagnosis based on DSM-5 criteria (average criteria met decreased from 7.75 to 2.31). Functionally, average Global Assessment of Functioning (GAF) scores rose substantially from 40 (indicating major impairment) to 70 (generally functioning well). All patients became engaged in work or education (compared to none initially), achieved financial independence (compared to only one initially), and most (7 out of 8) established stable relationships (compared to one initially). The change process was typically gradual, often catalyzed by life events, and frequently began with enhanced commitment to work/study and increased reflective ability (the capacity to understand mental states), followed later by improvements in relationships and self-esteem regulation. Key patient factors identified as indicators and potential contributors to change included motivation, reflective ability, emotion regulation, sense of agency (feeling in control and responsible), and interpersonal engagement.

The study concludes that meaningful change and remission are attainable for individuals with NPD through long-term, often multimodal, psychotherapy. It emphasizes that change is a complex, gradual, and multifactorial process influenced by the interplay of therapy, patient characteristics (like motivation and developing internal capacities), and significant life events. Improvements in external functioning (work, relationships) and internal capacities (reflection, agency, emotion regulation) serve as important markers, suggesting potential targets for intervention and monitoring in clinical practice, while acknowledging that core narcissistic vulnerabilities require sustained therapeutic attention within this dynamic context.

Research Impact and Future Directions

This qualitative study provides valuable, albeit preliminary, insights into the complex process of change for individuals diagnosed with Narcissistic Personality Disorder (NPD) undergoing long-term psychotherapy. Its primary contribution lies in offering a hopeful perspective, demonstrating that significant functional improvement and even diagnostic remission are achievable outcomes within this often challenging-to-treat population. The identification of specific patterns (e.g., gradual change, sequence starting with external functioning/reflection), indicators (e.g., agency, emotion regulation), and influencing factors (e.g., motivation, life events, multimodal treatment) offers potential avenues for future research and clinical focus.

However, the study's conclusions must be interpreted with significant caution due to major methodological limitations. The reliance on a small sample (N=8) of successful cases, documented through retrospective therapist reports without independent verification or systematic qualitative analysis procedures described, severely limits generalizability and introduces potential bias. We observe an association between long-term therapy and positive outcomes in this select group, but cannot definitively conclude causation or assume these findings apply broadly. The absence of the patient perspective is a critical omission, leaving a gap in understanding the subjective experience of change.

Practically, the findings suggest clinicians might monitor improvements in work/study and relationships as potential early indicators of change, alongside developing internal capacities like reflective ability and a sense of agency. The study underscores the likely necessity of long-term commitment and potentially integrated (multimodal) treatment approaches. Future research should prioritize prospective designs with larger, more diverse samples, incorporate patient self-report and independent assessments, employ rigorous, clearly defined qualitative analysis methods, and investigate factors contributing to treatment failure or dropout to provide a more complete picture.

Critical Analysis and Recommendations

Clear Problem Statement and Significance (written-content)
The abstract clearly states the research problem (difficulty treating NPD) and its significance (challenges to alliance/goals). This effectively frames the study's purpose and importance for readers.
Section: Abstract
Highlights Key Positive Findings and Remission (written-content)
The abstract highlights significant positive outcomes, including functional improvements and NPD remission, in all eight patients. This provides a compelling, concise overview of the study's optimistic main findings.
Section: Abstract
Strong Rationale and Gap Identification (written-content)
The introduction clearly articulates the lack of systematic studies on the processes and patterns of change in NPD treatment. This effectively identifies a critical gap in the literature, strongly justifying the need for this qualitative investigation.
Section: INTRODUCTION
Strengthen Connection Between Literature Review and Specific Study Aims (written-content)
The introduction reviews relevant literature but doesn't explicitly connect specific concepts (e.g., agency, emotion regulation) to the study's focus on identifying indicators, patterns, and processes. Making these links explicit would strengthen the logical argument for the study's specific aims and improve reader comprehension of its conceptual grounding.
Section: INTRODUCTION
Clear Diagnostic Criteria and Confirmation (written-content)
The method details the use of standardized DSM-5 criteria for NPD diagnosis, assessed retroactively by therapists and confirmed independently by authors, with quantitative data on criteria reduction (average 7.75 to 2.31). This adds diagnostic rigor and provides concrete evidence supporting the significant change observed in the selected sample.
Section: METHOD
Detail Qualitative Data Analysis Method (written-content)
The Method section fails to describe the qualitative analysis approach used to analyze the rich case report data. This is a critical omission, as it prevents readers from understanding how the reported patterns, processes, and indicators were derived, significantly limiting the study's reproducibility and the ability to evaluate the validity and reliability of the qualitative findings.
Section: METHOD
Specify Therapist Selection/Invitation Process (written-content)
The method does not specify how the eight participating therapists were selected or invited. Clarifying the recruitment process is necessary to assess potential selection bias in the therapist sample and, consequently, the generalizability of findings derived from their reports.
Section: METHOD
Quantification of Functional Improvement (written-content)
The Observations section quantifies substantial improvements in functioning using baseline vs. endpoint data (e.g., average GAF increase 40 to 70; shifts from 0% to 100% in work/study engagement; 12.5% to 100% financial independence; 12.5% to 87.5% in relationships). This provides concrete, compelling evidence supporting the claim of significant positive change in this sample.
Section: OBSERVATIONS
Identification of Change Patterns and Sequences (written-content)
The study successfully identifies and categorizes distinct patterns and sequences of change (summarized in Table 3), such as its gradual nature, the typical order of improvements (e.g., work/studies and reflection first), and key patient indicators (e.g., agency, emotion regulation). This moves beyond simple outcome reporting to offer valuable descriptive insights into the dynamics of how change unfolded in these cases.
Section: OBSERVATIONS
Explicitly Link Findings to Qualitative Analysis Method (written-content)
The Observations section reports patterns derived from qualitative data but doesn't explicitly state how these were synthesized (linking back to the missing analysis method). Clarifying the analytic process used to derive these patterns is crucial for methodological transparency and evaluating the trustworthiness of the findings.
Section: OBSERVATIONS
Effective Synthesis and Nuanced Factor Exploration (written-content)
The Discussion effectively synthesizes complex qualitative findings into coherent categories (factors promoting/challenging change) and explores factors like work/study and relationships with nuance (as both indicators and promoters). This structured synthesis aids reader comprehension of the multifaceted nature of change.
Section: DISCUSSION
Acknowledges Study Limitations (written-content)
The Discussion includes a dedicated subsection acknowledging key limitations (e.g., retrospective therapist reports, small N, lack of independent raters/patient perspective, potential confounds). This demonstrates appropriate scientific caution and aids readers in critically evaluating the findings.
Section: DISCUSSION
Elaborate on Clinical Applications and Therapeutic Strategies (written-content)
While the Discussion identifies factors associated with change, it could more explicitly discuss how therapists might leverage these insights therapeutically (e.g., specific strategies for addressing control needs or fear of change, prioritizing vocational goals). Elaborating on clinical applications would increase the research's practical utility for practitioners.
Section: DISCUSSION

Section Analysis

Abstract

Key Aspects

Strengths

Suggestions for Improvement

INTRODUCTION

Key Aspects

Strengths

Suggestions for Improvement

METHOD

Key Aspects

Strengths

Suggestions for Improvement

OBSERVATIONS

Key Aspects

Strengths

Suggestions for Improvement

Non-Text Elements

TABLE 1. Patients' Demographics, Comorbid Psychiatric Conditions, and Treatment...
Full Caption

TABLE 1. Patients' Demographics, Comorbid Psychiatric Conditions, and Treatment Modalities at Start of Treatment

Figure/Table Image (Page 9)
TABLE 1. Patients' Demographics, Comorbid Psychiatric Conditions, and Treatment Modalities at Start of Treatment
First Reference in Text
The patients' demographics, psychiatric conditions, treatment modalities, level of functioning at the beginning and end of treatment, including Global Assessment of Functioning (GAF) scores, vocational and educational commitment, close relationships, and financial dependency versus self-sufficiency are displayed in Tables 1 and 2.
Description
  • Table Overview: This table provides a snapshot of the characteristics of the eight individuals who participated in the study before their long-term psychotherapy began. It details their age, other mental health conditions they had alongside Narcissistic Personality Disorder (NPD), their educational background, and the types of therapy they were receiving at the study's start.
  • Age Distribution: The patients' ages ranged from 20 to 58 years old, with the average age being approximately 31 years (30.9). This indicates a mix of younger and middle-aged adults.
  • Comorbid Psychiatric Conditions: The table lists several other psychiatric conditions, known as comorbidities, present in the patients. 'Comorbidity' simply means having more than one medical condition at the same time. The most common were Borderline Personality Disorder (BPD, affecting 4 patients) - a condition often characterized by unstable relationships, self-image, and emotions - and Depressive disorder (also affecting 4 patients). Other conditions included Substance use disorder (3 patients), Eating disorder (3 patients), Bipolar II disorder (1 patient) - marked by depressive and hypomanic episodes, Antisocial Personality Disorder (ASPD) traits (1 patient) - ASPD involves disregard for others' rights, and Obsessive-compulsive disorder (1 patient).
  • Educational Background: Regarding education, all 8 patients had completed high school. Beyond that, educational attainment varied: 2 were college graduates, 4 had attended college but dropped out, and 2 had obtained graduate school degrees.
  • Primary Individual Treatment Modalities: The table outlines the therapeutic approaches used at the start. For primary individual therapy, the most common was Psychodynamic/psychoanalytic psychotherapy (3 patients), which focuses on unconscious processes and past experiences. Other therapies included Mentalization-Based Treatment (MBT, 1 patient), which helps patients understand their own and others' mental states; Dialectical Behavior Therapy (DBT, 1 patient), often used for BPD, focusing on skills like mindfulness and emotion regulation; Transference-Focused Psychotherapy (TFP, 1 patient), another psychodynamic approach emphasizing the patient-therapist relationship; and Eclectic therapy (2 patients), which combines techniques from different therapeutic schools.
  • Multimodal Treatments: In addition to individual therapy, many patients were involved in other forms of treatment (multimodal treatment). Four patients participated in residential treatment (living in a therapeutic facility), five received family therapy, and three attended DBT group therapy sessions.
Scientific Validity
  • Relevance of Data: The table appropriately presents baseline characteristics essential for understanding the study sample. Demographics, comorbidities, and prior/concurrent treatments are crucial covariates in psychotherapy research.
  • Sample Size Limitation: The primary limitation is the small sample size (n=8). While typical for qualitative case series, this significantly limits the generalizability of any findings derived from this group to the broader population of individuals with NPD.
  • Consistency in Comorbidity Reporting: Reporting 'ASPD (traits)' for one patient versus diagnoses for others introduces slight inconsistency. Clarifying the threshold used for reporting traits versus a full disorder would enhance precision.
  • Diagnostic Ascertainment Method: The method for ascertaining diagnoses (retroactive therapist reports confirmed by authors, as per page 344) should be kept in mind when interpreting these baseline data, as it relies on recall and potentially variable diagnostic practices, though confirmation adds rigor.
  • Treatment Modality Categorization: The categorization of treatment modalities appears sound, distinguishing primary individual therapy from adjunctive multimodal treatments, which accurately reflects complex care scenarios.
Communication
  • Organization and Clarity: The table is well-organized with clear headings and categories (Age, Comorbidity, Level of education, Treatment modalities). The use of 'n' to denote the number of patients within each category is appropriate and standard.
  • Caption Accuracy: The caption accurately reflects the content presented in the table, clearly stating it covers baseline demographics, comorbidities, and treatment modalities.
  • Use of Abbreviations and Notes: The inclusion of a note defining abbreviations (BPD, ASPD) enhances clarity, although expanding this to include less common therapy acronyms (MBT, TFP) might benefit readers less familiar with specific psychotherapy modalities.
  • Ambiguity in Comorbidity Counts: It is slightly ambiguous whether the comorbidity counts represent unique patients having at least that disorder, or if patients could be counted in multiple categories, summing to more than 8. Clarifying this (e.g., stating the total number of patients with any comorbidity) could improve precision.
  • Clarity of Treatment Categories: The distinction between 'Prime individual treatment' and 'Multimodal treatment' is clear and effectively communicates the different therapeutic contexts.
TABLE 2. Patients' Level of Functioning at Start and End of Treatment
Figure/Table Image (Page 10)
TABLE 2. Patients' Level of Functioning at Start and End of Treatment
First Reference in Text
All eight patients showed significant improvement in external personality and life functioning as measured by their engagement in work or studies, long-term relationships, and degree of financial responsibility and independence (Table 2) (Weinberg & Ronningstam, 2023).
Description
  • Table Overview: This table compares how the eight patients were functioning in key life areas at the beginning of their long-term psychotherapy versus at the end. It looks at their overall functioning, work or study engagement, relationships, and financial independence.
  • Global Assessment of Functioning (GAF) Change: A significant change is shown in the average Global Assessment of Functioning (GAF) score. The GAF scale is a numerical rating (0-100) used by clinicians to assess a person's overall level of psychological, social, and occupational functioning; higher scores indicate better functioning. The average score increased substantially from 40 at the start (indicating major impairment in several areas like work, relationships, or mood) to 70 at the end (indicating mild symptoms or some difficulty in social/occupational functioning, but generally functioning pretty well).
  • Work Status Change: Regarding work, there was a dramatic shift. At the start, 6 patients were unemployed and 2 had never worked. By the end, 5 were employed, with 3 in professional careers, 1 in skilled work, and 1 in temporary work. None remained unemployed or had never worked.
  • Study Status Change: Similarly, in education ('Studies'), no patients were engaged in studies at the start, whereas by the end, 3 patients were attending college or graduate school.
  • Relationship Status Change: In terms of relationships, only 1 patient was married at the start, with 7 having no relationship. By the end, 6 patients were in stable long-term relationships or married, 1 was dating, and only 1 reported no relationship.
  • Financial Situation Change: Financially, 7 out of 8 patients were dependent on parents or relatives at the start, with only 1 being self-sufficient (fully or partly). At the end of treatment, all 8 patients were reported as fully or partly self-sufficient, and none were dependent.
Scientific Validity
  • Focus on Functional Outcomes: Presenting functional outcomes (work, relationships, finances, GAF) is crucial in treatment studies, as symptom reduction doesn't always equate to improved life functioning. This table addresses that important dimension.
  • Use of GAF Score: The GAF score provides a global metric, but its reliability and validity have been debated, and it's known to be subjective. Its removal from the main DSM-5 criteria section reflects these concerns, although it remains a widely understood historical measure. The substantial change reported (40 to 70) is clinically significant.
  • Magnitude of Reported Change: The improvements shown across all domains (work/study, relationships, finances) are striking, suggesting substantial positive change. However, the data are categorical counts based on retroactive therapist reports.
  • Sample Size Limitation: The small sample size (n=8) severely limits the ability to generalize these findings. While indicative of potential change within this specific group, these results cannot be assumed to apply broadly to all NPD patients in long-term therapy.
  • Assessment Method (Retroactive): The reliance on retroactive therapist assessment introduces potential recall bias and lacks the rigor of prospective, standardized assessments by independent raters.
  • Lack of Statistical Testing within Table: The table presents descriptive counts ('n='). While the reference text claims 'significant improvement', the table itself doesn't include statistical tests (e.g., McNemar tests for paired categorical data, paired t-test for GAF) to formally assess the statistical significance of these changes from start to end.
  • Ambiguity in Financial Category: The definition of 'Self-sufficient (fully or partly)' could be more precise. Does 'partly' still involve some dependence?
Communication
  • Layout and Comparability: The table effectively uses a side-by-side layout to contrast patient functioning at the start versus the end of treatment, facilitating easy comparison across different domains (Work/Studies, Relationships/Finances).
  • Categorization: Categorization of functioning into distinct areas (Work, Studies, Relationships, Finances) is logical and clearly presented.
  • GAF Score Presentation: The presentation of GAF scores provides a concise summary measure of overall functioning change. Including the note defining GAF is crucial for readers unfamiliar with this scale.
  • Detail in Employment Status: The breakdown of employment status at the end of treatment (Professional, Skilled, Temp) adds useful granularity beyond simply stating 'Employed'.
  • Clarity of Counts: The use of 'n = x' clearly indicates the number of patients in each category, which is standard and effective.
  • Caption Accuracy: The caption accurately reflects the table's content, focusing on the change in functioning levels over the course of treatment.
TABLE 3. Patterns and Indicators of Change
Figure/Table Image (Page 12)
TABLE 3. Patterns and Indicators of Change
First Reference in Text
There were several identifiable change-related patterns and sequences (see Table 3).
Description
  • Table Overview: This table summarizes how change happened and what specific changes were observed in the eight patients undergoing long-term therapy for Narcissistic Personality Disorder (NPD). It breaks down the process into the pace of change, the order in which improvements appeared, and specific internal shifts noticed early on.
  • Pace of Change (Gradual/Sudden): Regarding the pace of change, all 8 patients showed a 'stepwise, gradual process of changes over time'. Additionally, 4 patients experienced moments of more sudden change, often linked to specific life events like family events, losses, or starting new jobs or relationships.
  • Sequence of Change: Early Improvements: The table outlines the sequence in which improvements in different areas of personality functioning were typically noticed by therapists. The earliest changes noted were often in 'Commitment to work or education' (noticed first in 5 patients) and 'Reflective and mentalizing ability' (noticed first in 5 patients). 'Reflective and mentalizing ability' refers to the capacity to think about one's own and others' thoughts, feelings, and intentions. Improvements in 'Interpersonal relations' (like friendships or close relationships) and 'Self-esteem regulation' (managing feelings of self-worth, reducing grandiosity, tolerating insecurity) were noticed first in fewer patients (3 each).
  • Sequence of Change: End of Treatment Status: By the end of treatment, improvements were noted in most patients across these areas, though not universally. Commitment to work/education was seen in all 8 patients. Improved interpersonal relations were noted in 7 patients. Enhanced reflective/mentalizing ability was noted in 6 patients. Better self-esteem regulation was observed in 5 patients, suggesting this area remained challenging for some even at termination.
  • Early Patient-Related Indicators: The table also lists key internal changes ('Patient-related indicators') observed within the first two years of treatment. 'Reflective ability, curiosity, forming narratives' (telling coherent stories about their experiences) was seen in 6 patients. A stronger 'Sense of agency, responsibility, initiatives' (feeling in control of their lives, taking ownership) was noted in 5 patients. Better 'Emotion regulation and tolerance' (managing feelings without being overwhelmed) was seen in 5 patients. Increased 'Interpersonal and social engagements' were also noted in 5 patients.
Scientific Validity
  • Qualitative Data Synthesis: The table synthesizes qualitative observations from therapist reports into quantifiable counts, providing a structured overview of perceived change patterns and indicators in a hard-to-study population.
  • Reliance on Subjective Therapist Reports: The primary limitation stems from the data source: retrospective therapist observations. This introduces potential subjectivity, recall bias, and lack of standardized assessment for the indicators (e.g., 'reflective ability', 'sense of agency'). The counts reflect therapist judgment rather than objective measurement.
  • Small Sample Size: The small sample size (n=8) significantly limits the generalizability of these observed patterns and sequences. These findings are suggestive for this cohort but cannot be assumed to represent universal pathways of change in NPD treatment.
  • Theoretical Relevance of Indicators: The indicators listed (reflective ability, sense of agency, emotion regulation) are theoretically relevant constructs in personality change, aligning with models of psychotherapeutic process.
  • Temporal Dimension: The distinction between 'First noticed' and 'Noticed at end of treatment' provides a useful, albeit potentially approximate, temporal dimension to the observed changes.
  • Confounding Treatment Factors: The heterogeneity of treatment modalities (psychodynamic, MBT, DBT, etc.) and the use of multimodal treatments make it difficult to attribute the observed patterns solely to one specific therapeutic factor; they likely reflect a combination of influences.
Communication
  • Structure and Organization: The table is logically structured, separating the nature of change (gradual/sudden), the sequence in which different functional areas improved, and specific patient-related indicators observed early in treatment.
  • Temporal Clarity: The use of distinct columns for 'First noticed' and 'Noticed at end of treatment' effectively communicates the temporal sequence of observed improvements in different areas.
  • Clarity of Indicators: The listed indicators (e.g., 'Commitment to work or education', 'Reflective and mentalizing ability') are relatively clear, although some terms ('reflective ability', 'sense of agency') rely on the reader's familiarity with psychotherapeutic concepts or the main text for full understanding.
  • Use of Counts: Presenting the findings as counts ('n' out of 8) is straightforward and appropriate for this type of qualitative summary, clearly indicating how many patients exhibited each pattern or indicator.
  • Caption Accuracy: The caption accurately describes the table's focus on patterns and indicators of change.

DISCUSSION

Key Aspects

Strengths

Suggestions for Improvement

↑ Back to Top