Next Article in Journal
The COMT rs4680 Met Allele Is Associated with High Impulsivity, Hyperactivity and Inattention in Patients with Opioid Use Disorder
Previous Article in Journal
Cognitive-Emotional Teacher Burnout Syndrome: A Comprehensive Behavioral Data Analysis of Risk Factors and Resilience Patterns During Educational Crisis
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Psychiatry International
Volume 7
Issue 1
10.3390/psychiatryint7010027
psychiatryint-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Is Borderline Personality Disorder a Precursor of Schizoaffective Psychosis? A Twenty-Year Retrospective Study of More than 400 Patients from a Psychiatric Hospital

by
Joana Henriques-Calado
1,*,
Martin M. Schumacher
2 and
João Gama-Marques
1,3,4
1
Centro de Investigação em Ciência Psicológica (CICPSI), Faculdade de Psicologia, Universidade de Lisboa, Alameda da Universidade, 1649-013 Lisboa, Portugal
2
Independent Researcher, 4450 Sissach, Switzerland
3
Consulta de Esquizofrenia Resistente (CER), Unidade Local de Saúde São José (ULSSJ), Centro Clínico Académico de Lisboa (CCAL), Hospital Júlio de Matos (HJM), 1749-002 Lisboa, Portugal
4
Clínica Universitária de Psiquiatra e Psicologia Médica (CUPPM), Faculdade de Medicina, Universidade de Lisboa (FMUL), Centro Académico de Medicina de Lisboa (CAML), 1649-028 Lisboa, Portugal
*
Author to whom correspondence should be addressed.
Psychiatry Int. 2026, 7(1), 27; https://doi.org/10.3390/psychiatryint7010027
Submission received: 18 October 2025 / Revised: 30 December 2025 / Accepted: 29 January 2026 / Published: 2 February 2026
Versions Notes

Abstract

Background: Both borderline personality disorder (BPD) and schizoaffective disorder (SAD), as well as their potential connection, remain controversial diagnoses. To explore whether BPD may be part of the spectrum of SAD, we conducted a longitudinal study of a large clinical cohort of patients with BPD. Methods: We assessed the diagnostic trajectories of 402 patients with BPD in a 20-year retrospective study based on electronic clinical records from a psychiatric hospital using ICD-9 diagnoses. Data were descriptively examined on concurrent and sequential diagnoses in patients with BPD. For the classification of SAD, a proxy diagnosis was used. Results: The study population showed a high prevalence of affective disorders and a high frequency of concurrent diagnoses of affective–BPD. Together, stable BPD, stable affective disorder sequences and transitions from affective disorders to BPD represented 79% of all longitudinal trajectories. Conclusion: These findings should be considered exploratory and do not allow confirmation or refutation of the hypothesis that BPD serves as a precursor, prodrome, or component within the spectrum of SAD.

1. Introduction

The diagnostic constructs of borderline personality disorder (BPD) and schizoaffective disorder (SAD) have long been the subject of debate, similar to schizophrenia itself [1,2]. These debates reflect enduring challenges in psychiatric nosology, particularly in defining boundaries between affective, psychotic, and personality-related psychopathology.

1.1. Borderline Personality Disorder

Historically, researchers have reasoned whether BPD represents a variant of schizophrenia or an affective disorder [3]. Some proposals have placed BPD within the bipolar spectrum [4], conceptualized it as a form of bipolar II disorder [5], or suggested alternative designations such as fluxithymia [6] or borderpolar [7]. These conceptualizations are primarily grounded in phenomenological overlap and longitudinal instability rather than in consensus nosological criteria. Current diagnostic systems acknowledge that BPD may include transient, stress-related psychotic symptoms, non-affective micro psychotic episodes, which are typically brief, circumscribed, and non-persistent, and differ from the sustained mood episodes that characterize bipolar II disorder [4,5,7]. Consequently, BPD continues to be classified as a personality disorder distinct from mood and psychotic disorders in contemporary classification systems.
Other authors have proposed that BPD may belong to the schizophrenia spectrum [8]. This view has been based on the presence of psychotic-like symptoms and disturbances in self-experience, although BPD is generally characterized by greater affective reactivity, interpersonal instability, and less persistent psychosis than schizophrenia. Together, these divergent conceptualizations illustrate the heterogeneity of BPD and the absence of consensus on its placement within existing diagnostic spectra.

1.2. Schizoaffective Disorder

Although SAD remains classified as a distinct diagnostic entity in ICD-11 and DSM-5, it continues to generate conceptual debate [9,10,11]. Some authors have questioned its validity and reliability, while others have supported its position as an intermediate disorder between schizophrenia and affective disorders [12,13,14,15,16,17,18]. This intermediate positioning is reflected in dimensional and spectrum-based models that conceptualize SAD as combining the core features of psychotic and affective syndromes. Kendler’s trichotomy [16] represents an influential framework to understand these relationships, and empirical studies have documented diagnostic instability over time, with patients receiving alternating diagnoses of schizophrenia, bipolar disorder, and SAD in different clinical assessments.

1.3. Borderline Personality Disorder and Schizoaffective Disorder

The potential relationship between BPD and SAD has been intermittently discussed in the literature. Conceptually, it has been suggested that BPD and SAD may share certain clinical features without necessarily belonging to the same diagnostic spectrum. More than four decades ago, BPD was proposed as an attenuated form of SAD [19], although subsequent empirical support for this hypothesis has remained limited [20,21,22]. More recent authors have suggested that BPD may attract [23] or mimic [24] SAD diagnoses, raising the question of whether BPD could precede, overlap with, or be misclassified as SAD in routine clinical practice.
Aiken [25] noted that the early drafts of DSM-III referred to BPD as cycloid personality disorder, a term that evokes similarities with cyclothymic disorder. This historical observation highlights longstanding nosological challenges related to cycloid psychoses, which have traditionally resisted clear classification within bipolar [26,27] or schizophrenia spectra [28,29]. These constructs underscore the persistent difficulty in delineating boundaries between affective instability, psychotic symptoms, and personality pathology.
Despite these conceptual discussions, empirical data on the longitudinal diagnostic relationships between BPD and SAD remain scarce. Specifically, it is unclear whether BPD is followed over time by diagnostic patterns consistent with schizoaffective disorder, or whether observed associations reflect primarily diagnostic instability, overlap with affective disorders, or clinical coding practices. To address this gap, the present study conducted a retrospective longitudinal analysis of diagnostic trajectories in patients with BPD, focusing on concurrent, sequential, and trajectory-based diagnoses recorded over a twenty-year period. Using electronic clinical records, we examined whether diagnostic sequences involving affective and schizophrenia spectrum diagnoses occurred with sufficient frequency to support the hypothesis that BPD may be part of a schizoaffective spectrum. We hypothesized that, if BPD functions as a precursor or component of such a spectrum, diagnostic trajectories compatible with schizoaffective patterns would be observable over time [30,31,32,33,34,35,36,37]. The primary objective of the study was to examine diagnostic comorbidity and longitudinal diagnostic trajectories, with specific attention to the occurrence and possible diagnostic evolution of SAD in patients with BPD.

2. Methods

2.1. Data Sources and Participants

We conducted a twenty-year longitudinal retrospective analysis (2000–2019) in the largest psychiatric hospital in Lisboa, Portugal (European Union), through a systematic review of electronic clinical records. All 505 patients diagnosed with borderline personality disorder (BPD) were identified using the ICD-9 code 301.83 of the World Health Organization’s International Classification of Diseases, Ninth Revision (ICD-9). Inclusion and exclusion criteria: All 505 patients in the database who had a diagnosis of borderline personality disorder between 2000 and 2019 were initially considered. Then, two exclusion criteria were applied: (i) observation period shorter than 6 months and (ii) presence of an organic brain disorder (dementia (8), epilepsy (11), cerebral lesion (9), multiple sclerosis (2), thyroiditis (2), pseudohermaphroditism (1)), leading to the exclusion of 70 and 33 patients, respectively (103 in total). Consequently, data from 402 patients were included in the final analysis.
For these patients, we also assessed the presence and timing of the following additional psychiatric diagnoses, schizophrenia (ICD-9 295.x), affective psychosis (ICD-9 296.x), mania (ICD-9 291.x), and non-psychotic depression (ICD-9 300.4 and 311). Within ICD-9, schizoaffective disorder is classified as a subtype of schizophrenia (ICD-9 295.7). However, schizophrenia subtypes were not specified in the available dataset, which prevented direct analyses of SAD. As an operational substitute, a schizoaffective-type diagnosis was attributed to patients who received consecutive or sequential diagnoses of schizophrenia and any affective disorder. In this study, we applied a sequential definition of schizoaffective disorder based on the Marneros proposal [13,14,16] and defined SAD-compatible trajectories as diagnostic sequences that combined affective disorders and schizophrenia diagnoses. SAD is stated if an affective disorder and schizophrenia were diagnosed sequentially in the same patient.
For patients who had received only one or two diagnoses, the commonly applied biostatistical method of “last observation carried forward” (LOCF) was used, which means that the most recent diagnosis was carried forward to fill the missing second and third diagnostic entries. This procedure assumes that the diagnosis remains stable when subsequent diagnostic reassessments are not available.
The demographic variables extracted included sex, age, marital status, and years of education. The variables of clinical use included the number of psychiatric admissions, the total days of inpatient days, and the dates of the first and last contact with the hospital.

2.2. Data Analysis

Clinical data from all 402 patients with at least one diagnosis of BPD were examined over an observation period of up to 20 years, focusing on diagnostic transition frequencies and longitudinal diagnostic trajectories. During follow-up, up to five diagnoses per patient were recorded. Because only 8% of patients received more than three diagnoses (four diagnoses: 6%; five diagnoses: 2%), analyses of diagnostic trajectories were restricted to the first three diagnoses. Given the chronic clinical profiles of this cohort, for patients with only two diagnoses, the third diagnosis in the trajectory was assumed to be identical to the second, following a last-observation-carried-forward principle.
Five ICD-9 diagnostic categories were available: affective psychosis not otherwise specified (AP), borderline personality disorder (BPD), depression (D), mania (M), and schizophrenia (S). Because the AP category does not differentiate between depressive, manic, mixed, or bipolar presentations, diagnoses of AP, D, and M were grouped into a single category of affective disorders (A) to ensure analytical clarity. Consequently, most of the analyses were conducted using three psychiatric diagnostic categories: affective disorders (A), borderline personality disorder (BPD), and schizophrenia (S).
A substantial proportion of patients (25%) received multiple concurrent diagnoses, most of which were assigned on the same day, generally at first contact. All diagnoses recorded within a 31-day interval were considered concurrent. To avoid artificial inflation of comorbidity, hierarchical diagnostic rules (e.g., affective > schizophrenia > personality disorders) were not applied, as this could result in the loss or misclassification of BPD and schizophrenia diagnoses. The dataset was therefore divided into two subsets: patients with concurrent diagnoses (n = 101) and patients without concurrent diagnoses (n = 301). The concurrent diagnoses subset was used to examine diagnostic overlap and specificity, using both the original five diagnostic categories and the reduced three-category classification. The subset without concurrent diagnoses was used to analyze diagnostic transitions and longitudinal trajectories using the three-category classification. No patient received the same diagnosis more than once, either sequentially or separated by another diagnosis.
The analyses were primarily descriptive and exploratory. Diagnostic frequencies were summarized using counts and proportions, with 95% confidence intervals calculated using the Wilson method. Baseline diagnostic distributions were analyzed at the diagnosis level, as patients could contribute multiple diagnoses, whereas analyses of concurrent and sequential diagnoses, and diagnostic trajectories were conducted at the patient level. The longitudinal diagnostic change was examined using unconditional analyses of overall transition distributions and conditional analyses stratified by initial diagnosis. Global χ tests were applied to identify transitions contributing to overall significance when appropriate, with adjusted standardized residuals. All calculations and statistical analyses were conducted using Microsoft Excel (Version 8.0), (Redmond, WA, USA) and R (R Foundation for Statistical Computing, Vienna, Austria). Effects for p values ≤ 0.05 were considered statistically significant.

3. Results

3.1. Data Summary

The descriptive statistics for the full sample of 402 patients are presented in Table 1 and Supplementary Table S1. The cohort was predominantly female (71.6%), with a mean age of 31.3 years (SD = 9.9) at first contact and a mean of 3.4 psychiatric admissions (SD = 8.0). Half of the patients had an overall follow-up time exceeding seven years.
Of the 402 patients, 87 patients had only 1 diagnosis during the observation period, resulting in 933 global diagnoses (Table 1). Affective disorders accounted for 49.2% of diagnoses (95% CI [45.9, 52.5]), followed by BPD diagnoses (43.1%, 95% CI [39.9, 46.3]), while schizophrenia diagnoses were less frequent (7.7%, 95% CI [6.2, 9.6]). Original diagnoses subsumed under category A (affective disorders) varied in frequency, with depressive and affective psychosis diagnoses predominating.

3.2. Concurrent Diagnoses

To characterize the co-occurrence of cross-sectional diagnoses, pairwise combinations of concurrent collapsed diagnoses were examined among patients with multiple diagnoses (Table 2 and Supplementary Table S2).
Among patients with concurrent collapsed diagnoses (n = 101), the most frequent pairwise combination was A–BPD, observed in 61.4% of patients (95% CI [51.7, 70.2]) (Table 2). Homogeneous A–A combinations occurred in 30.7% (95% CI [22.6, 40.2]), whereas combinations involving schizophrenia diagnoses were uncommon.

3.3. Sequential Diagnoses

Two complementary approaches were used to examine diagnostic transitions. First, unconditional analyses compared the overall transition distributions between sequential assessments, treating all transitions independently of the initial diagnosis (Table 3). Second, conditional analyses examined transition patterns stratified by a constant first diagnosis, allowing evaluation of diagnostic stability and change within each diagnostic triad (Table 4). This approach enabled differentiation between global redistribution effects and diagnosis-specific longitudinal patterns.
Sequential diagnostic transitions were evaluated in patients in a subgroup sample who did not have concurrent diagnoses (n = 301). Of these, 87 patients (28.9%) received only a single diagnosis (all BPD), 119 (39.5%) received two diagnoses, 71 (23.6%) received three diagnoses, 18 (6.0%) received four diagnoses, and 6 (2.0%) received five diagnoses. As detailed in Table 3, only the first three diagnoses were considered in the analyses of direct diagnostic transitions and trajectories. Diagnostic transition distributions differed significantly between the sequences (χ2(8) = 34.61, p < 0.001; Table 3). This effect was primarily driven by the increase in BPD diagnostic stability and reduced A–BPD transitions over time. The BPD-BPD sequences increased from 28.9% (95% CI [24.1, 34.3]) between first and second diagnoses to 46.3% (95% CI [39.7, 52.9]) between second and third diagnoses, exceeding expected frequencies. In contrast, A–BPD transitions were more frequent at the initial reassessment (32.9%, 95% CI [27.8, 38.4]) than at later assessments (21.0%, 95% CI [16.1, 27.0]). Although rare, S–S sequences occurred more often than expected in the second-to-third sequence (1.9%, 95% CI [0.7, 4.7]) and contributed to the overall distributional difference. Other transition categories showed no clinically meaningful variation across sequences. Of the nine possible transitions, only three occurred in more than 10% of cases: stable affective (A–A), stable BPD (BPD–BPD), and affective to BPD (A–BPD) diagnoses, and together they accounted for 79% of all transitions.
Transition conditional frequencies stratified by initial diagnosis are shown in Table 4. The conditional transition patterns differed significantly between the sequences (χ2(6) = 18.9, p = 0.004; Table 4). Among patients with an initial diagnosis of A, stability (A–A) increased from 28.9% (95% CI [22.3, 36.5]) to 41.8% (95% CI [31.6, 52.8]), while A–BPD transitions decreased. For initial BPD diagnoses, BPD–BPD stability increased over time (77.0% to 83.2%). Among schizophrenia diagnoses, S–S transitions, absent in the first assessment, emerged at the second-to-third assessment (25.0%, 95% CI [10.2, 49.5]), while S–BPD transitions declined. Other transitions showed no clinically meaningful change. Diagnostic stability increased over time, particularly for BPD–BPD and A-A transitions, while diagnostic changes were more frequent at the initial reassessment. Transitions involving schizophrenia diagnoses were infrequent but showed non-random patterns across sequences.
These conditional transition patterns translated into a limited number of dominant diagnostic trajectories throughout the entire three-assessment sequence.

3.4. Diagnostic Trajectories

With three diagnostic categories, 29 possible three-step diagnostic trajectories can be defined. Of these, 16 were observed in the dataset. Accordingly, three diagnostic trajectories were highly concentrated (Table 5). More than half of patients (55.5%) followed one of two trajectories (BPD-BPD-BPD or A-BPD-BPD), and 67.8% followed one of the three most frequent trajectories (BPD-BPD-BPD, A-BPD-BPD, or A-A-BPD). Pure stable trajectories accounted for 30.9% of the sample, whereas fully unstable trajectories were uncommon (6.3%).

3.5. Potential Schizoaffective-Spectrum Trajectories

The hypothesis that BPD may represent a precursor or component of a schizoaffective spectrum was examined using the available data. Because schizoaffective disorder could not be directly identified, trajectories that involved consecutive diagnoses of affective disorder and schizophrenia (A–S or S–A) were used as an operational proxy. Trajectories that began with BPD and followed by A–S or S–A were considered potentially supportive of this hypothesis.
As shown in Table 5, only three of the 301 patients followed one of these trajectories, all of whom were of type BPD–S–A type. Among these 301 patients, 113 (37.5%) had an initial BPD diagnosis, and a total of 55 schizophrenia diagnoses (18.3%) were recorded. Assuming a random distribution of schizophrenia diagnoses across the cohort, the expected number of schizophrenia diagnoses among patients with an initial BPD diagnosis would be 21, corresponding to an expected probability of 18.6%. The observed frequency of schizoaffective-type trajectories among patients with an initial BPD diagnosis was 6.2%. [95% CI approximately 2.1–16.8%], reflecting the small number of observed cases.
Assuming equal probabilities for the five possible trajectories (BPD-BPD-BPD, BPD-S-S, BPD-A-A, BPD-A-S, and BPD-S-A) as a theoretical reference model rather than an empirically derived distribution, the expected combined probability of the trajectories of BPD-S-A and BPD-A-S would be 40%. Observed frequencies among patients with an initial BPD diagnosis were 77.0% for BPD-BPD-BPD, 3.5% for BPD-S-S, 16.8% for BPD-A-A, and 2.7% for BPD-S-A; the BPD-A-S trajectory was not observed. Schizoaffective-type trajectories were therefore less frequent than stable BPD or BPD–affective trajectories. Within the constraints of the available data, these findings do not indicate an increased frequency of schizoaffective-type trajectories following an initial diagnosis of BPD.

4. Discussion

This study examined concurrent, sequential, and trajectory diagnoses among patients with BPD in a large hospital cohort, with the objective of exploring whether diagnostic patterns involving affective disorders and schizophrenia might support or contradict the hypothesis that BPD is a precursor to the spectrum of SAD. Within this cohort of patients with BPD, the concurrent diagnoses revealed a high frequency of combinations involving affective disorders and BPD, a pattern consistent with the previous literature describing a substantial overlap between BPD and bipolar spectrum conditions [3,4,5,7]. To ensure interpretive caution, these patterns are best understood as descriptive reflections of observed coding practices rather than indicators of underlying nosological relationships. These concurrent associations may reflect clinical co-occurrence or variability in diagnostic attribution across settings, underlying a diagnostic heterogeneity observed in psychiatric practice.
The prominence of concurrent affective diagnoses (Table 1 and Table 2), particularly affective psychosis, depression, and mania, further indicates that, in this dataset, BPD appeared most often in proximity to affective-spectrum categories rather than schizoaffective-sequence diagnoses. These findings align with the notion of homologous associations [38], in which different diagnostic constructs within a broad spectrum may capture overlapping manifestations of the underlying psychopathology. This variability within the spectrum has been discussed as a contributor to elevated comorbidity rates [39,40] and may arise from limitations in phenomenological differentiation [41] or categorical diagnostic thresholds that impose boundaries on dimensional phenomena [40,42]. In this context and given the documented associations between schizoaffective and affective disorders [13,16] as well as evidence of shared etiological factors [43], it remains conceivable that some schizoaffective presentations may not be distinctly coded in clinical records. Consequently, the present findings should be interpreted considering the structural constraints of the ICD-9 coding, rather than as direct evidence on the diagnostic processes.
Sequential diagnostic patterns (Table 3 and Table 4) in this study were characterized primarily by the stability of the diagnosis of BPD and, secondarily, by transitions from affective disorders to BPD. In this dataset, these transitions occurred more frequently than in reverse, and the diagnosis of BPD remained the most stable category when considered in relation to affective disorders and schizophrenia. Examination of the most common diagnostic trajectories (Table 5) also showed the predominance of stable BPD patterns and trajectories that included affective disorders and BPD. These trajectories collectively accounted for most cases, indicating that within this cohort the BPD codes tended to remain stable or appear in temporal proximity to the affective disorder codes. These observations suggest consistency in how BPD was applied in clinical records; however, the data do not allow for a distinction between clinical stability and assessment practices or their interaction. Interpretation of these patterns does not allow conclusions about causality, developmental stage, or diagnostic progression. Rather, they reflect how diagnoses were assigned and recorded over time in this hospital setting. The literature [4,5,7] notes that some BPD presentations may occur with subthreshold affective pathology and that affective and personality features can coexist. Approaches focused on affective temperaments [44] can contextualize co-occurring affective and personality disorders within broader models of mood-related vulnerability without positing specific causes, yet it appears to influence the clinical presentation and course of psychiatric disorders. A smaller subset of patients in this dataset began with schizophrenia codes, and these were associated with subsequent transitions to affective and BPD codes (Table 4). These patterns are consistent with previous observations of diagnostic instability in schizophrenia spectrum conditions [8], but do not permit further conclusions beyond the descriptive patterns observed.
The specific focus of this study was the potential relationship between BPD and patterns compatible with an SAD spectrum trajectory. Because ICD-9 subtypes of schizophrenia were not available in these records, SAD could not be identified directly. Instead, SAD compatible trajectories were operationalized [13,14,16] as sequences combining affective disorders and schizophrenia diagnoses. Within these constraints, the expected probability of such trajectories was higher than the observed frequency. In this dataset, schizoaffective-type sequences were infrequent compared to stable BPD or affective–BPD trajectories. Given the operationalized criteria and the inherent limitations of retrospective clinical data, these findings warrant cautious interpretation and should be viewed as indicators of recorded coding patterns rather than evidence against potential clinical associations. The variability in diagnostic frameworks or classification systems, including Marneros [13,14,16], DSM, and ICD, further limits the comparability of SAD-related categories.
Taken together and within the interpretative limits of this retrospective clinical study, the findings suggest that the BPD codes tended to remain stable or appear in combination with affective disorder codes, while patterns consistent with the SAD sequences occurred infrequently. The literature contains reports that describe possible relationships between BPD and schizoaffective conditions [19,23,24,30,31,37], although these are limited in number and methodological scope. Future diagnostic frameworks may benefit from refining the classification of disorders characterized by combinations of affective and psychosis-spectrum symptoms [45], with the aim of improving the consistency and specificity of BPD diagnosis. Dimensional models of psychopathology, such as the recent psychosis superspectrum [46] and a schizophrenia–schizoaffective–bipolar spectra model proposal [30,47], can provide complementary ways to interpret the heterogeneous diagnostic patterns observed in BPD.

Strengths and Limitations

A major strength of the current study is the opportunity to analyze concurrent and sequential diagnostic trajectories over a long observational period in a clinical cohort of patients with BPD.
However, several limitations must be acknowledged. The Portuguese healthcare system used ICD-9 from 2000 to 2020, adopting ICD-10 only in 2021. As a result, the study necessarily relied on ICD-9 diagnoses, which were the only available electronic records. Reliance on ICD-9-coding clinical records represents a substantial methodological constraint, as the precision and consistency of diagnostic attribution cannot be independently verified within this dataset. We believe that schizoaffective psychosis [13] should ideally be diagnosed in a sequential manner, requiring at least one depressive or manic episode and one non-affective psychotic episode. For this reason and due to the structure of the ICD-9 coding, particularly since schizoaffective disorder was classified as a subtype of schizophrenia within ICD-9. The ICD-9 diagnosis of affective psychosis (296) was broader than expected; many clinicians used this diagnosis without specifying whether the presentation was depressive, manic, mixed, full, or partial, or whether there were psychotic symptoms. This imprecision in diagnostic specification represents a relevant limitation of clinical records. Furthermore, because the operational definition of SAD used in this study was constructed solely from clinical records transitions between affective and schizophrenia codes, it constitutes an unvalidated proxy and should be strictly interpreted as a descriptive analytic approach rather than as a diagnostic construct.
Furthermore, the BPD diagnosis was made in a variety of clinical contexts, including admissions, discharges, and routine outpatient evaluations, and no longitudinal data on symptom remission were available. Additional limitations include the possibility of selection bias, the retrospective design, and the general constraints associated with clinical records. No biomarkers, neuropsychological assessments or objective physiological measures (such as cerebrospinal fluid, electroencephalography, or neuroimaging) were available, and no blood tests or urinalysis were performed to systematically assess drug use. Similarly, no genetic or family history data were collected, limiting the ability to exclude hereditary factors. The absence of these clinical, biological, and contextual variables restricts the ability to examine symptom-level trajectories, diagnostic decision-making processes, or underlying mechanisms. Consequently, no causal, developmental, or prodromal inferences can be drawn from the present data.
The exclusive reliance on ICD-9 also prevented comparison with ICD-11 criteria [48], and the study did not examine how BPD criteria may differ between diagnostic systems (such as DSM, ICD, or CCMD) [49]. Together, these limitations emphasize that the results should be strictly interpreted as descriptive patterns derived from administrative diagnostic coding, rather than as evidence of underlying clinical pathways or nosological associations.
In conclusion, within the constraints of this retrospective clinical record dataset, the available data are insufficient to confirm or exclude the hypothesis that BPD may be related to the spectrum of schizoaffective disorders. The longitudinal diagnostic trajectories observed in this twenty-year hospital cohort indicate that stable diagnoses of BPD, as well as transitions between affective disorders and BPD, were substantially more frequent than diagnostic sequences combining affective and schizophrenia spectrum codes, as operationalized in the present study in a large psychiatric cohort. In contrast, diagnostic patterns compatible with schizoaffective trajectories were uncommon in this dataset, although the limitations inherent to ICD-9 coding and the use of an unvalidated proxy for schizoaffective disorder preclude definitive conclusions about their clinical significance or true prevalence. Consequently, these findings should be strictly interpreted as descriptive patterns derived from electronic diagnostic records, rather than as evidence supporting or refuting a specific nosological or developmental hypothesis, and no causal, prodromal, or etiological inferences can be drawn. The results highlight diagnostic instability and overlap between BPD and affective disorders in a large hospital cohort. Future research applying sequential diagnostic criteria for schizoaffective disorder may help to clarify longitudinal diagnostic trajectories, providing a clinically informative framework alongside current DSM and ICD classifications.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/psychiatryint7010027/s1, Table S1: Demographic and clinical characteristics of the study sample (N = 402). Table S2: Pairwise Combinations of Concurrent Original Diagnoses.

Author Contributions

Conceptualization, J.H.-C., M.M.S. and J.G.-M.; Methodology, M.M.S. and J.H.-C.; Formal analysis, M.M.S. and J.H.-C.; Investigation, J.H.-C. and J.G.-M.; Data curation, J.G.-M.; Writing—original draft, J.H.-C., M.M.S. and J.G.-M.; Writing—review & editing, J.H.-C., M.M.S. and J.G.-M.; Supervision, J.G.-M.; Funding acquisition, J.H.-C. and J.G.-M. All authors have read and agreed to the published version of the manuscript.

Funding

This research was partially financially by national funding from Fundação para a Ciência e a Tecnologia (FCT) [Foundation for Science and Technology] through the Centro de Investigação em Ciência Psicológica (CICPSI), Faculdade de Psicologia, Universidade de Lisboa, Portugal [Research Center for Psychological Science of the Faculty of Psychology, University of Lisbon, Portugal] (UIDB/04527/2020; UIDP/04527/2020).

Institutional Review Board Statement

Ethical review and approval were waived because this retrospective analysis was part of a quality study using anonymized data.

Informed Consent Statement

Patient consent was waived due to retrospective design of the study.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Acknowledgments

The authors wish to express their gratitude to research assistants, Filipa Cameirinha and Miguel Benrós for their help with the database coding.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Lake, C.R. Disorders of thought are severe mood disorders: The selective attention defect in mania challenges the Kraepelinian dichotomy: A review. Schizophr. Bull. 2008, 34, 109–117. [Google Scholar] [CrossRef]
  2. Van Os, J. ‘Salience syndrome’ replaces ‘schizophrenia’ in DSM-V and ICD-11: Psychiatry’s evidence-based entry into the 21st century? Acta Psychiatr. Scand. 2009, 120, 363–372. [Google Scholar] [CrossRef]
  3. McGlashan, T.H. The borderline syndrome. II. Is it a variant of schizophrenia or affective disorder? Arch. Gen. Psychiatry 1983, 40, 1319–1323. [Google Scholar] [CrossRef] [PubMed]
  4. Parker, G.; Bayes, A.; Spoelma, M.J. Why might bipolar disorder and borderline personality disorder be bonded? J. Psychiatr. Res. 2022, 150, 214–218. [Google Scholar] [CrossRef] [PubMed]
  5. Akiskal, H.S. The prevalent clinical spectrum of bipolar disorders: Beyond DSM-IV. J. Clin. Psychopharmacol. 1996, 16, 4S–14S. [Google Scholar] [CrossRef] [PubMed]
  6. Tyrer, P. Why borderline personality disorder is neither borderline nor a personality disorder. Personal. Ment. Health 2009, 3, 86–95. [Google Scholar] [CrossRef]
  7. Zimmerman, M. Borderpolar: Patients with borderline personality disorder and bipolar disorder. Psychiatr. Times 2019, 36, 17–18. [Google Scholar]
  8. Zandersen, M.; Parnas, J. Borderline personality disorder or a disorder within the schizophrenia spectrum? A psychopathological study. World Psychiatry 2019, 18, 109–110. [Google Scholar] [CrossRef]
  9. Maier, W. Do schizoaffective disorders exist at all? Acta Psychiatr. Scand. 2006, 113, 369–371. [Google Scholar] [CrossRef]
  10. Lake, C.R.; Hurwitz, N. Schizoaffective disorders are psychotic mood disorders; there are no schizoaffective disorders. Psychiatry Res. 2006, 143, 255–287. [Google Scholar] [CrossRef]
  11. Malhi, G.S.; Bell, E. Schizoaffective disorder: Time to refine our thinking not the criteria? Schizophr. Res. 2019, 208, 34–35. [Google Scholar] [CrossRef] [PubMed]
  12. Angst, J.; Marneros, A. Bipolarity from ancient to modern times: Conception, birth and rebirth. J. Affect. Disord. 2001, 67, 3–19. [Google Scholar] [CrossRef] [PubMed]
  13. Marneros, A. The schizoaffective phenomenon: The state of the art. Acta Psychiatr. Scand. 2003, 418, 29–33. [Google Scholar] [CrossRef] [PubMed]
  14. Marneros, A. Do schizoaffective disorders exist at all? Acta Psychiatr. Scand. 2007, 115, 162–163. [Google Scholar] [CrossRef]
  15. Peralta, V.; Cuesta, M.J. Exploring the borders of the schizoaffective spectrum: A categorical and dimensional approach. J. Affect. Disord. 2008, 108, 71–86. [Google Scholar] [CrossRef][Green Version]
  16. Marneros, A.; Tsuang, M.T. Affective and Schizoaffective Disorders: Similarities and Differences; Springer: Berlin, Germany, 2011. [Google Scholar]
  17. Mancuso, S.G.; Morgan, V.A.; Mitchell, P.B.; Berk, M.; Young, A.; Castle, D.J. A comparison of schizophrenia, schizoaffective disorder, and bipolar disorder: Results from the second Australian national psychosis survey. J. Affect. Disord. 2015, 172, 30–37. [Google Scholar] [CrossRef]
  18. Wood, A.J.; Carroll, A.R.; Shinn, A.K.; Ongur, D.; Lewandowski, K.E. Diagnostic stability of primary psychotic disorders in a research sample. Front. Psychiatry 2021, 28, 734272. [Google Scholar] [CrossRef]
  19. Stone, M.H. Assessing vulnerability to schizophrenia or manic-depression in borderline states. Schizophr. Bull. 1979, 5, 105–110. [Google Scholar] [CrossRef][Green Version]
  20. Bell, M.; Billington, R.; Cicchetti, D.; Gibbons, J. Do object relations deficits distinguish BPD from other diagnostic groups? J. Clin. Psychol. 1988, 44, 511–516. [Google Scholar] [CrossRef]
  21. Oldham, J.M.; Skodol, A.E. Personality disorders in the public sector. Hosp. Community Psychiatry. 1991, 42, 481–487. [Google Scholar] [CrossRef]
  22. Clarke, M.; Hafner, R.J.; Holme, G. Borderline personality disorder: A challenge for mental health services. Aust. N. Z. J. Psychiatry 1995, 29, 409–414. [Google Scholar] [CrossRef] [PubMed]
  23. Kingdon, D.; Taylor, L.; Ma, K.; Kinoshita, Y. Changing name: Changing prospects for psychosis. Epidemiol. Psychiatr. Sci. 2013, 22, 297–301. [Google Scholar] [CrossRef] [PubMed]
  24. Krauss, A.; Bernard, J.; Okusaga, O.O. Challenges and considerations in treating negative and cognitive symptoms of schizophrenia spectrum disorders. Fed. Pract. 2022, 39, 448–453. [Google Scholar] [CrossRef] [PubMed]
  25. Aiken, C. Separated at birth: Bipolar and borderline personality disorders. Psychiatric Times, 16 May 2019. [Google Scholar]
  26. Madalena, J.C. As psicoses atípicas [The atypical psychoses]. Rev. Psiquiat. 1981, 24, S1–S93. [Google Scholar]
  27. Pfuhlmann, B.; Jabs, B.; Althaus, G.; Schmidtke, A.; Bartsch, A.; Stöber, G.; Beckmann, H.; Franzek, E. Cycloid psychoses are not part of a bipolar affective spectrum: Results of a controlled family study. J. Affect. Disord. 2004, 83, 11–19. [Google Scholar] [CrossRef]
  28. Van de Kerkhof, N.W.A.; Fekkes, D.; van der Heijden, F.M.M.A.; Hoogendijk, W.J.; Stöber, G.; Egger, J.I.; Verhoeven, W.M. Cycloid psychoses in the psychosis spectrum: Evidence for biochemical differences with schizophrenia. Neuropsychiatr. Dis. Treat. 2016, 12, 1927–1933. [Google Scholar] [CrossRef]
  29. Foucher, J.R.; Gawlik, M.; Roth, J.N.; de Crespin de Billy, C.; Jeanjean, L.C.; Obrecht, A.; Mainberger, O.; Clauss, J.M.E.; Elowe, J.; Weibel, S.; et al. Wernicke-Kleist-Leonhard phenotypes of endogenous psychoses: A review of their validity. Dialogues Clin. Neurosci. 2020, 22, 37–49. [Google Scholar] [CrossRef]
  30. Gama Marques, J.; Ouakinin, S. Schizophrenia-schizoaffective-bipolar spectra: An epistemological perspective. CNS Spectrums 2019, 28, 197–201. [Google Scholar] [CrossRef]
  31. Henriques-Calado, J.; Gonçalves, B.; Marques, C.; Paulino, M.; Gama Marques, J.; Grácio, J.; Pires, R. In light of the DSM-5 dimensional model of personality: Borderline personality disorder at the crossroads with the bipolar spectrum. J. Affect. Disord. 2021, 294, 897–907. [Google Scholar] [CrossRef]
  32. Maj, M.; Perris, C. An approach to the diagnosis and classification of schizoaffective disorders for research purposes. Acta Psychiatr. Scand. 1985, 72, 405–413. [Google Scholar] [CrossRef] [PubMed]
  33. Maj, M.; Perris, C. Patterns of course in patients with a cross-sectional diagnosis of schizoaffective disorder. J. Affect. Disord. 1990, 20, 71–77. [Google Scholar] [CrossRef] [PubMed]
  34. Marneros, A.; Deister, A.; Rohde, A. Stability of diagnoses in affective, schizoaffective and schizophrenic disorders: Cross-sectional versus longitudinal diagnosis. Eur. Arch. Psychiatry Clin. Neurosci. 1991, 241, 187–192. [Google Scholar] [CrossRef] [PubMed]
  35. Marneros, A. Schizoaffective disorder. Korean J. Schizophr. Res. 2012, 15, 5–12. [Google Scholar] [CrossRef]
  36. Brenner, R.; Campbell, K.; Konakondla, K.; Madhusoodanan, S.; Brenner, S. Late onset schizoaffective disorder. Consultant 2014, 54, 487–488. [Google Scholar]
  37. Cohen, C.I. Editorial comment: Schizoaffective disorder in later life: “Beware the Jabberwock”. Am. J. Geriatr. Psychiatry 2017, 25, 951–952. [Google Scholar] [CrossRef]
  38. Angst, J.; Sellaro, R.; Ries Merikangas, K. Multimorbidity of psychiatric disorders as an indicator of clinical severity. Eur. Arch. Psychiatry Clin. Neurosci. 2002, 252, 147–154. [Google Scholar] [CrossRef]
  39. Vella, G.; Aragona, M.; Alliani, D. The complexity of psychiatric comorbidity: A conceptual and methodological discussion. Psychopathology 2000, 33, 25–30. [Google Scholar] [CrossRef]
  40. First, M.B. Mutually exclusive versus co-occurring diagnostic categories: The challenge of diagnostic comorbidity. Psychopathology 2005, 38, 206–210. [Google Scholar] [CrossRef]
  41. Parnas, J. A disappearing heritage: The clinical core of schizophrenia. Schizophr. Bull. 2011, 37, 1121–1130. [Google Scholar] [CrossRef]
  42. Nordgaard, J.; Nielsen, K.M.; Rasmussen, A.R.; Henriksen, M.G. Psychiatric comorbidity: A concept in need of a theory. Psychol. Med. 2023, 53, 5902–5908. [Google Scholar] [CrossRef]
  43. Laursen, T.M.; Agerbo, E.; Pedersen, C.B. Bipolar disorder, schizoaffective disorder, and schizophrenia overlap: A new comorbidity index. J. Clin. Psychiatry 2009, 70, 1432–1438. [Google Scholar] [CrossRef] [PubMed]
  44. Favaretto, E.; Bedani, F.; Brancati, G.E.; De Berardis, D.; Giovannini, S.; Scarcella, L.; Martiadis, V.; Martini, A.; Pampaloni, I.; Perugi, G.; et al. Synthesising 30 years of clinical experience and scientific insight on affective temperaments in psychiatric disorders: State of the art. J. Affect. Disord. 2024, 362, 406–415. [Google Scholar] [CrossRef]
  45. Stein, D.J.; Lund, C.; Nesse, R.M. Classification systems in psychiatry: Diagnosis and global mental health in the era of DSM-5 and ICD-11. Curr. Opin. Psychiatry 2013, 26, 493–497. [Google Scholar] [CrossRef] [PubMed]
  46. Kotov, R.; Jonas, K.G.; Carpenter, W.T.; Dretsch, M.N.; Eaton, N.R.; Forbes, M.K.; Forbush, K.T.; Hobbs, K.; Reininghaus, U.; Slade, T.; et al. HiTOP Utility Workgroup. Validity and utility of Hierarchical Taxonomy of Psychopathology (HiTOP): I. Psychosis superspectrum. World Psychiatry 2020, 19, 151–172. [Google Scholar] [CrossRef] [PubMed]
  47. Gama-Marques, J. Diagnostic change from bipolar disorder to schizophrenia and schizophrenia to bipolar disorder makes no sense: The new diagnosis should be sequential schizoaffective disorder. Schizophr. Bull. 2025, 51, 1179. [Google Scholar] [CrossRef]
  48. Peterson, D.L.; Webb, C.A.; Keeley, J.W.; Gaebel, W.; Zielasek, J.; Rebello, T.J.; Robles, R.; Matsumoto, C.; Kogan, C.S.; Kulygina, M.; et al. The reliability and clinical utility of ICD-11 schizoaffective disorder: A field trial. Schizophr. Res. 2019, 208, 235–241. [Google Scholar] [CrossRef]
  49. Lai, C.M.; Leung, F.; You, J.; Cheung, F. Are DSM-IV-TR borderline personality disorder, ICD-10 emotionally unstable personality disorder, and CCMD-III impulsive personality disorder analogous diagnostic categories across psychiatric nomenclatures? J. Pers. Disord. 2012, 26, 551–567. [Google Scholar] [CrossRef]
Table 1. Distribution of original and lumped baseline diagnoses in patients with multiple diagnoses (N = 402).
Table 1. Distribution of original and lumped baseline diagnoses in patients with multiple diagnoses (N = 402).
Diagnosesn (%)95% CI
Original diagnoses
AP175 (18.8)[16.4–21.4]
BPD402 (43.1)[39.9–46.3]
D230 (24.7)[22.0–27.6]
M54 (5.8)[4.5–7.5]
S72 (7.7)[6.2–9.6]
Lumped diagnoses
A (AP + D + M)459 (49.2)[45.9–52.5]
BPD402 (43.1)[39.9–46.3]
S72 (7.7)[6.2–9.6]
Note. The percentages are based on the total number of baseline diagnoses recorded rather than the number of patients, as individuals could contribute more than one diagnosis. Confidence intervals (95% CI) for proportions were calculated using the Wilson method. Original diagnoses AP, D, and M collapsed into category A (affective disorders) for lumped analyses. A = affective disorders; AP = affective psychosis; BPD = borderline personality disorder; D = depression; M = mania; S = schizophrenia.
Table 2. Pairwise combinations of concurrent collapsed diagnoses.
Table 2. Pairwise combinations of concurrent collapsed diagnoses.
Pairwise
Combinations		n (%)95% CI
A–A31 (30.7)[22.6–40.2]
A–BPD62 (61.4)[51.7–70.2]
A–S4 (4.0)[1.6–9.8]
BPD–S2 (2.0)[0.6–6.9]
Note. The percentages are based on the total number of patients (n = 101). The patients could contribute to more than one pairwise combination. Confidence intervals (95% CI) for proportions were calculated using the Wilson method. Diagnoses were collapsed into categories: A = affective disorders; BPD = borderline personality disorder; S = schizophrenia.
Table 3. Diagnostic transition frequencies, 95% confidence intervals, and adjusted standardized residuals across sequential diagnoses.
Table 3. Diagnostic transition frequencies, 95% confidence intervals, and adjusted standardized residuals across sequential diagnoses.
Diagnostic
Transition		1st → 2nd Diagnosis2nd → 3rd Diagnosis
n (%) [95% CI]Rₐdⱼn (%) [95% CI]Rₐdⱼ
A–A44 (14.6%) [11.1–19.1]−0.3033 (15.4%) [11.2–20.9]0.40
A–BPD99 (32.9%) [27.8–38.4]2.70 *45 (21.0%) [16.1–27.0]−2.50 *
A–S9 (3.0%) [1.6–5.6]1.201 (0.5%) [0.1–2.6]−1.50
BPD–BPD87 (28.9%) [24.1–34.3]−3.10 *99 (46.3%) [39.7–52.9]3.40 *
BPD–A19 (6.3%) [4.1–9.6]−0.4018 (8.4%) [5.4–12.9]0.60
BPD–S7 (2.3%) [1.1–4.7]1.002 (0.9%) [0.3–3.3]−1.20
S–A16 (5.3%) [3.3–8.5]0.807 (3.3%) [1.6–6.6]−0.90
S–BPD20 (6.6%) [4.3–10.0]1.405 (2.3%) [1.0–5.4]−1.60
S–S0 (0.0%) [0.0–1.3]−2.00 *4 (1.9%) [0.7–4.7]2.80 *
Total301 (100%)—214 (100%)—
Note. The percentages are based on the totals of the columns. Confidence intervals (95% CI) for proportions were calculated using the Wilson method. Rₐdⱼ are adjusted standardized residuals from a chi-square test of independence comparing diagnostic transition distributions across sequences. Positive residuals indicate transitions more frequent than expected under the null hypothesis; negative residuals indicate fewer. Absolute Rₐdⱼ ≥ 1.96 indicate statistically meaningful contributions to the overall chi-square effect (p < 0.05) and are shown in bold with an asterisk (*). A = affective disorders; BPD = borderline personality disorder; S = schizophrenia.
Table 4. Conditional diagnostic transitions between sequential diagnoses, 95% confidence intervals, and adjusted standardized residuals.
Table 4. Conditional diagnostic transitions between sequential diagnoses, 95% confidence intervals, and adjusted standardized residuals.
Diagnostic
Transition		1st → 2nd Diagnosis2nd → 3rd Diagnosis
n (%) [95% CI]Rₐdⱼn (%) [95% CI]Rₐdⱼ
A–A44 (28.9%) [22.3–36.5]−2.10 *33 (41.8%) [31.6–52.8]2.30 *
A–BPD99 (65.1%) [57.2–72.3]1.9045 (57.0%) [46.0–67.4]−2.00 *
A–S9 (5.9%) [3.1–10.8]1.101 (1.3%) [0.2–6.8]−1.40
BPD–BPD87 (77.0%) [68.4–83.8]−1.8099 (83.2%) [75.3–88.9]2.00 *
BPD–A19 (16.8%) [11.1–24.7]0.2018 (15.1%) [9.8–22.6]−0.10
BPD–S7 (6.2%) [3.0–12.2]1.402 (1.7%) [0.5–6.0]−1.70
S–A16 (44.4%) [29.6–60.3]0.007 (43.8%) [23.1–66.8]−0.10
S–BPD20 (55.6%) [39.7–70.4]2.20 *5 (31.3%) [14.2–55.6]−2.30 *
S–S0 (0.0%) [0.0–9.6]−2.00 *4 (25.0%) [10.2–49.5]2.50 *
Note. The percentages are conditional on a constant first diagnosis within each triad. Confidence intervals (95% CI) were calculated using the Wilson method. Rₐdⱼ denotes adjusted standardized residuals from a chi-square test comparing transition distributions between sequences. Absolute values ≥ 1.96 indicate statistically meaningful contributions (p < 0.05) and are shown in bold with an asterisk (*). A = affective disorders; BPD = borderline personality disorder; S = schizophrenia.
Table 5. Diagnostic trajectories across three sequential assessments.
Table 5. Diagnostic trajectories across three sequential assessments.
Trajectory
Group		Trajectoryn (%)95% CI
PureA-A-A6 (2.0)[0.9–4.3]
BPD-BPD-BPD87 (28.9)[24.1–34.3]
Total93 (30.9)[26.0–36.3]
Mixed A–BPDA-A-BPD37 (12.3)[9.1–16.4]
A-BPD-A17 (5.6)[3.5–8.8]
BPD-A-A19 (6.3)[4.1–9.6]
A-BPD-BPD80 (26.6)[21.9–31.9]
Total153 (50.8)[45.2–56.4]
Mixed S-A/BPDA-A-S1 (0.3)[0.1–1.8]
A-S-A4 (1.3)[0.5–3.4]
S-A-A8 (2.7)[1.4–5.1]
BPD-S-S4 (1.3)[0.5–3.4]
S-BPD-BPD19 (6.3)[4.1–9.6]
Total36 (12.0)[8.9–16.0]
Fully UnstableA-BPD-S2 (0.7)[0.2–2.4]
A-S-BPD5 (1.7)[0.7–3.9]
BPD-S-A3 (1.0)[0.3–2.9]
S-A-BPD8 (2.7)[1.4–5.1]
S-BPD-A1 (0.3)[0.1–1.8]
Total19 (6.3)[4.1–9.6]
Note. The percentages are based on the subgroup sample who did not have concurrent diagnoses (n = 301). Confidence intervals (95% CI) were calculated using the Wilson method. A = affective disorders; BPD = borderline personality disorder; S = schizophrenia. Diagnostic trajectories were highly concentrated: the two most frequent trajectories (BPD-BPD-BPD and A-BPD-BPD) accounted for 55.5% of all cases, and the three most frequent trajectories (BPD-BPD-BPD, A-BPD-BPD, and A-A-BPD) accounted for 67.8%.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Henriques-Calado, J.; Schumacher, M.M.; Gama-Marques, J. Is Borderline Personality Disorder a Precursor of Schizoaffective Psychosis? A Twenty-Year Retrospective Study of More than 400 Patients from a Psychiatric Hospital. Psychiatry Int. 2026, 7, 27. https://doi.org/10.3390/psychiatryint7010027

AMA Style

Henriques-Calado J, Schumacher MM, Gama-Marques J. Is Borderline Personality Disorder a Precursor of Schizoaffective Psychosis? A Twenty-Year Retrospective Study of More than 400 Patients from a Psychiatric Hospital. Psychiatry International. 2026; 7(1):27. https://doi.org/10.3390/psychiatryint7010027

Chicago/Turabian Style

Henriques-Calado, Joana, Martin M. Schumacher, and João Gama-Marques. 2026. "Is Borderline Personality Disorder a Precursor of Schizoaffective Psychosis? A Twenty-Year Retrospective Study of More than 400 Patients from a Psychiatric Hospital" Psychiatry International 7, no. 1: 27. https://doi.org/10.3390/psychiatryint7010027

APA Style

Henriques-Calado, J., Schumacher, M. M., & Gama-Marques, J. (2026). Is Borderline Personality Disorder a Precursor of Schizoaffective Psychosis? A Twenty-Year Retrospective Study of More than 400 Patients from a Psychiatric Hospital. Psychiatry International, 7(1), 27. https://doi.org/10.3390/psychiatryint7010027

Article Metrics

Back to TopTop