Factor structure of the Brief Psychiatric Rating Scale – Expanded (BPRS-E) in a large hospitalized sample

doi:10.1016/j.jpsychires.2017.05.011

Journal of Psychiatric Research

Volume 93, October 2017, Pages 79-86

https://doi.org/10.1016/j.jpsychires.2017.05.011 Get rights and content

Abstract

Meta-analytically derived models of the BPRS-E factor structure were tested on a large (n = 33,903) hospitalized sample using confirmatory factor analysis. A modified four factor model (Positive symptoms, Negative symptoms, Affect, and Activation) containing 12 core BPRS-E symptoms based on the meta-analytic models had excellent model fit. An additional five factor, 15 core symptom model, which added a Disorganization factor consistent with the pentagonal model of schizophrenic symptoms also had support with generally good fit. These factors demonstrated acceptable reliability as measured by coefficient alpha (M = 0.77). These factors were compared across three major diagnostic classes and indicated clinically relevant differences between these groups such as depressed patients having higher Affect scores (d = 1.03), manic-mixed episode patients having higher Activation scores (d = 0.83) and schizophrenic patients having higher Positive Symptom scores (d = 0.89) providing evidence for the validity of these factors. Further exploratory factor analyses provided support for the factors of Positive symptoms, Negative symptoms, Affect, and Activation with additional smaller and less robust factors corresponding to Resistance emerging as a fifth factor and Disorganization emerging as a sixth factor.

Section snippets

Participants

The BPRS-E rating scale assessments were conducted on patients admitted to nine Texas state mental health hospitals which serve many of the most severely mentally ill admitted to psychiatric hospitals in the state. A total of 33,903 unduplicated patients had complete BPRS-E data which represents about 60% of all patients admitted during the time period of 2001–2006. Data for this study was extracted from a clinical database of those inpatient records. This study was approved by the Texas

Results

The 33 alternative models specified on average 4.5 factors using 22.5 salient items. None of the 33 models met recommended levels for a “good” fit, with the following mean indexes of fit: TLI/NNFI = 0.65, CFI = 0.70, RMSEA = 0.11, SRMR = 0.13, AIC = 97,541, BIC = 97,987.

Neither of the full models (A, B) based on Dazzi, Shafer and Lauriola's (2016) BPRS-E similarity and reproduced correlation meta-analyses had acceptable levels of fit (mean indexes of fit: TLI/NNFI = 0.75, CFI = 0.78,

Discussion

The results of the confirmatory factor analytic models tested in this study clearly supported the four factor 12 item model (D) for the BPRS-E which was similar to the four factor 15 core item model derived from BPRS-E meta-analysis (C). There was also support for a modified and expanded five factor 15 item pentagonal model (E) that added a fifth Disorganization factor which had good fit for all but one criteria. However, larger more inclusive CFA models derived from the BPRS-E meta-analysis

Author note

The first and second author contributed equally to the design, analysis and writing. The third author contributed to writing the manuscript and provided critical reviews. The authors have no conflicts of interest. The views and opinions expressed in this paper are the authors own and do not represent any official policy of any government or university. The authors would also like to note the passing of the BPRS creator John E. Overall last year (2016).

References (40)

F. Dazzi et al.
Meta-analysis of the brief psychiatric rating scale - Expanded (BPRS-E) structure and arguments for a new version
J. Psychiatr. Res.
(2016)
F. Mouaffak et al.
Validation of the French version of the Expanded Brief Psychiatric Rating Scale with anchor BPRS-E
Encephale
(2010)
I. Pacchiarotti et al.
The symptom structure of bipolar acute episodes: in search for the mixing link
J. Affect. Disord.
(2013)
G. Perugi et al.
Phenomenological subtypes of severe bipolar mixed states: a factor analytic study
Compr. Psychiatry
(2014)
E. Petkari et al.
Acute psychopathology as a predictor of global functioning in patients with ICD-10 non-affective psychosis: a prospective study in 11 European countries
Schizophr. Res.
(2011)
A. Picardi et al.
Symptom structure of acute mania: a factor study of the 24-item Brief Psychiatric Rating Scale in a national sample of patients hospitalized for a manic episode
J. Affect. Disord.
(2008)
A. Picardi et al.
Heterogeneity and symptom structure of schizophrenia
Psychiatry Res.
(2012)
M. Ruggeri et al.
Factor solution of the BPRS-expanded version in schizophrenic outpatients living in five European countries
Schizophr. Res.
(2005)
D. Velligan et al.
Brief Psychiatric Rating Scale Expanded version: how do new items affect factor structure?
Psychiatry Res.
(2005)
J. Ventura et al.
Symptom dimensions in recent-onset schizophrenia and mania: a principal components analysis of the 24-item Brief Psychiatric Rating Scale
Psychiatry Res.
(2000)

A. Zanello et al.

The Brief Psychiatric Rating Scale (version 4.0) factorial structure and its sensitivity in the treatment of outpatients with unipolar depression

Psychiatry Res.

(2013)

S.W. Anderson et al.

Predicting inpatient length of stay with the expanded version of the expanded Brief Psychiatric Rating Scale

Psychiatr. Serv.

(2004)

B. Biancosino et al.

The BPRS-E as a predictor of length of stay in a residential facility

Psychiatr. Serv.

(2005)

G.K. Burger et al.

Factor structure of the Expanded Brief Psychiatric Rating Scale

J. Clin. Psychol.

(1997)

A. Cechnicki et al.

Consistency of symptomatic dimensions of schizophrenia in a 12 year follow up study

Arch. Psychiatry Psychother.

(2010)

A. Colasanti et al.

Symptom dimensions as predictors of clinical outcome, duration of hospitalization, and aggressive behaviors in acutely hospitalized patients with psychotic exacerbation

Clin. Pract. Epidemiol. Ment. Health

(2010)

G.L. Day

Confirmatory Factor Analysis of the Brief Psychiatric Rating Scale - Expanded Version

(2003)

F. Dazzi et al.

Psychopathological assessment of risk of restraint in acute psychiatric patients

J. Nerv. Ment. Dis.

(2017)

P.M. Dingemans et al.

Component structure of the Expanded Brief Psychiatric Rating Scale (BPRS-E)

Psychopharmacology

(1995)

H.K. Horton et al.

Factor structure of the BPRS in deaf people with schizophrenia: correlates to language and thought

Cogn. Neuropsychiatry

(2011)

Cited by (38)

Validation and clinical usefulness of the Hamilton Program for Schizophrenia Voices Questionnaire (HPSVQ) among French-speaking voice-hearers
2023, Encephale
Auditory verbal hallucinations (AVH), also called voices, are often distressing to individuals experiencing them. Valid and reliable instruments are necessary to document the hearing voices experience across cultures. The Hamilton Program for Schizophrenia Voices Questionnaire (HPSVQ) is becoming a widely used self-administered instrument for assessing characteristics, content and subjective effect of AVH. This study investigates the psychometric characteristics of the HPSVQ French version (HPSVQ FV) in a clinical sample of voice-hearers. The results showed that the HPSVQ yields a global score (HPSVQ-Global) as well as a distress (HPSVQ-Distress) and a severity (HPSVQ-Severity) sub-score having good, acceptable and questionable internal consistency respectively. Significant correlations were found between hallucination severity (BPRS 4.0), distress (PSYRATS-AH), voices acceptance (VAAS-9), anxiety and depression (HADS). However, no significant associations were observed between Suspicion and Unusual Thoughts (BPRS 4.0). At a one-week interval, the temporal stability of the three indices was excellent. Moreover, after a brief cognitive intervention, a significant reduction was observed in all indices. Taken together, the HPSVQ FV demonstrated good construct validity, reliability and sensitivity to change. These findings support the use of the HPSVQ in francophone clinical and research settings. However, the bi-factorial solution of the HPSVQ FV should be further examined in larger samples.
Entendre des voix (hallucinations acoustico-verbales) peut être une expérience subjective éprouvante. Afin de comparer le vécu des personnes issues de diverses cultures, il est nécessaire, autant en clinique qu’en recherche, de disposer d’outils d’auto-évaluation brefs, valides, fidèles et sensibles. Le Hamilton Program for Schizophrenia Voices Questionnaire (HPSVQ) remplit ces critères, mais n’a pas encore été traduit en langue française.
Cette étude a ainsi pour objectif de présenter et de valider la version française du HPSVQ (HPSVQ VF) chez des entendeurs de voix francophones.
L’échantillon initial était composé de 81 entendeurs de voix souffrant d’un trouble psychiatrique et indiqués pour une thérapie focalisée sur les voix. Le HPSVQ VF a été élaboré par la méthode traduction-rétro traduction et diverses caractéristiques psychométriques ont été examinées, notamment: la validité de construit uni-factorielle (HPSVQ-Global) et bifactorielle (HPSVQ-Détresse, HPSVQ-Sévérité) à travers une analyse factorielle confirmatoire (CFA); la validité interne (coefficient α de Cronbach); la stabilité temporelle à une semaine d’intervalle (ICC); la validité concourante (sous-échelle PSYRATS-Détresse, item Hallucination de la BPRS 4.0 et les sous-échelles Anxiété et Dépression de la HADS); la validité discriminante (items Suspicion et Pensées inhabituelles de la BPRS 4.0 et échelle d’acceptation des voix VAAS-9); ainsi que la sensibilité au changement après une brève intervention thérapeutique centrée sur les stratégies pour gérer les voix (brief-Coping Strategy Enhancement: brief-CSE).
Les résultats des CFA suggèrent que la validité de construit uni et bifactorielles sont tous deux satisfaisants (χ² > 21,00], p > 0,59); RMSEA = 0,00; CFI = 1,00; TLI [1,03–1,05]; SRMR = 0,06). La validité interne des trois indices (HPSVQ-Global, HPSVQ-Détresse et HPSVQ-Sévérité) est respectivement bonne α = 0,82), acceptable (α=0,77) et questionnable (α = 0,65). La stabilité test-retest est quant à elle excellente (ICC > 0,74). Concernant la validité concourante, les trois indices du HPSVQ corrèlent positivement et significativement avec les autres mesures ([0,30 ≤ r < 0,66]; p < 0,02), à l’exception de la corrélation entre le HPSVQ-Sévérité et le PSYRATS-Détresse (r = 0,16, p = 0,54). Pour la validité discriminante, aucune des corrélations entre les trois indices du HPSVQ et les items Suspicion et Pensées inhabituelles de la BPRS 4.0 n’est significative ([−0,12 ≤ r < 0,03], p > 0,40). Néanmoins, des corrélations négatives et significatives entre les indices du HPSVQ et la VAAS-9 sont observées ([−0,60 < r < −0,50], p < 0,001). Finalement, après participation à la brief-CSE nous observons une diminution statistiquement significative des scores du HPSVQ-Global (t = 3,00; p = 0,008), du HPSVQ-Détresse (t = 3,13, p = 0,006) et du HPSVQ-Sévérité (t = 2,50, p = 0,023).
Cette étude montre que le HPSVQ VF possède des qualités psychométriques satisfaisantes pour être utilisé dans les situations cliniques et les études francophones focalisées sur les voix. Toutefois, sa structure latente devrait être réexaminée dans un échantillon plus large.
Acceptability and effectiveness of a novel cycling training on the parkinsonian motor and psychiatric symptoms in individuals with schizophrenia: A pilot study
2021, Complementary Therapies in Medicine
Recent evidence suggests that high-speed, low-resistance stationary cycling training (termed as speedwork) alleviates motor symptoms in people with Parkinson’s disease. Similar motor symptoms commonly exist in people with schizophrenia (Sz); however, they were neglected in the previous literature.
Our objective was to evaluate if speedwork could also be used as a strategy to improve parkinsonian motor symptoms in Sz. We aimed 1) to evaluate the adherence and acceptability of speedwork in Sz, 2) to assess test-retest reliability of the motor assessments that are novel to Sz research, 3) to evaluate the effectiveness of speedwork in improving parkinsonian motor, and 4) psychiatric symptoms in Sz.
Ten Sz outpatients with concurrent parkinsonian motor symptoms completed 12 sessions (2 sessions/week) of speedwork training. Participants were evaluated on motor functioning and psychiatric symptom severity twice before (double baseline) and twice after (post-completion and 6-wk follow-up) the speedwork training.
The adherence to speedwork was high (92 %) and the results of exercise acceptability questionnaire indicate participants found various domains of exercise highly acceptable (overall average 4.49/5). There were improvements in various domains of motor symptoms including, walking speed, functional mobility, static and dynamic balance, and upper extremity motor function after the completion of training (all p < 0.025), with many of these improvements remaining at the 6-wk follow-up. Moreover, there was evidence for improvement in positive psychotic symptoms after the completion of speedwork (p < 0.025).
Speedwork training could be an acceptable and effective strategy to improve motor and psychiatric symptoms in Sz.
Meta-analytic exploration of the joint factors of the Brief Psychiatric Rating Scale – Expanded (BPRS-E) and the positive and negative symptoms scales (PANSS)
2021, Journal of Psychiatric Research
Previous meta-analyses of the BPRS-E and PANSS were combined to analyze the joint factors across 77 factor analyses from 68 studies of 32,896 individuals. The BPRS-E and PANSS share the 18 original BPRS items and each has additional unique items (6 BPRS-E and 12 PANSS). Based on the combined data across the BPRS-E and PANSS four separate sets of factor analyses were conducted using sample weighting (weighted and unweighted) and matrix type (similarity coefficients and reproduced correlations). Five factors: Positive Symptoms, Affect, Negative Symptoms, Disorganization and Activation-Mania were consistent across all solutions. The factors Positive Symptoms and Affect were defined primarily by the original 18 BPRS items. The factors Negative Symptoms and Disorganization were largely defined by the PANSS additional items. The Activation-Mania factor was largely defined by the additional BPRS-E items. The results were similar to those found in previous meta-analyses but the current factors were substantially more comprehensive and better defined because they included all the items from both the BPRS-E and the PANSS. Future research should consider using all the items from both the PANSS and BPRS-E as the unique items from each scale improve the assessment and measurement of these factors.
Relation between self-perceived stress, psychopathological symptoms and the stress hormone prolactin in emerging psychosis
2021, Journal of Psychiatric Research
Citation Excerpt :
For teenagers (i.e., 14–17 years), the cutoff was 354 mU/l in boys and 492 mU/l in girls. The Brief Psychiatric Rating Scale Expanded (BPRS-E; Lukoff et al., 1986; Ventura et al., 1993) was used to assess Activation, Positive Symptoms, Negative Symptoms, Affect and Disorganization as defined by the five factor model of Shafer et al. (2017). All data were analyzed using the R environment for statistical computing (R Core Team, 2019).
Psychosocial stress and the stress hormone prolactin are assumed to play an important role in the pathogenesis of schizophrenia and related psychoses, and have been frequently observed to be increased in antipsychotic-naïve patients with a clinical high risk for psychosis (CHR-P) or first episode of psychosis (FEP). The aim of this study was to further elucidate the relationships between self-perceived stress, psychopathological symptoms and prolactin levels in these patients.
In this cross-sectional study, 45 healthy controls, 31 CHR-P patients and 87 FEP patients were recruited from two different study centers. Prolactin was measured under standardized conditions between 8 and 10 am. All patients were antipsychotic-naïve and not taking any prolactin influencing medication. Self-perceived stress during the last month was measured with the perceived stress scale (PSS-10) immediately before blood taking.
Both CHR-P and FEP patients showed significantly higher levels of self-perceived stress and prolactin than controls. Hyperprolactinemia (i.e. prolactin levels above the reference range) was observed in 26% of CHR-P and 45% of FEP patients. Self-perceived stress was significantly positively associated with affective symptoms, but not with other symptoms. There was no significant association between self-perceived stress and prolactin levels.
Our results confirm that CHR-P and FEP patients have higher stress levels than healthy controls and frequently have hyperprolactinemia, independent of antipsychotic medication. However, although it is well established that prolactin increases in response to stress, our results do not support the notion that increased prolactin levels in these patients are due to stress.
Suicide behavior is associated with childhood emotion dysregulation but not trait impulsivity in first episode psychosis
2020, Psychiatry Research
Historically, research on suicide behavior has not included those experiencing first episode psychosis (FEP), hindering prevention efforts for this population. Emotion dysregulation and impulsivity represent two mechanisms that contribute to suicide, but these have not been examined in FEP. We hypothesize that the combination of trait impulsivity and childhood emotion dysregulation are associated with suicide behavior (SB) and ideation (SI) in those experiencing FEP. Participants were recruited from an Early Psychosis Program (N=80, ages 12-32, 65% male). Clinician ratings of symptoms and history of SI and SB were obtained at baseline. Participants also completed self-report measures of childhood emotion dysregulation and trait impulsivity. Regression analyses examined whether childhood emotion dysregulation and trait impulsivity individually or in combination were associated with SI and SB, and the severity of SI and SB. Childhood emotion dysregulation was significantly associated with a history of SB and its severity, but not SI. Attention impulsivity was associated with the severity of SI. However, other impulsivity types, and interactions were not associated with a history of SI or SB. This suggests childhood emotion dysregulation is a potential target for prevention of SB in FEP, while trait impulsivity may be less important in this effort.
Global emotion dysregulation and maladaptive cognitive emotion regulation strategies mediate the effects of severe trauma on depressive and positive symptoms in early non-affective psychosis
2020, Schizophrenia Research
Citation Excerpt :
Each BPRS-E item is rated on a 7-point scale, ranging from 1 (not present) to 7 (extremely severe). The present study formulated the positive (items 9, 10, 11), negative (items 16, 17, 18), and depressive symptom (items 3, 4, 5) factors based on the most recent and rigorous confirmatory factor analysis study on the BPRS-E (Shafer et al., 2017). The BPRS-E had adequate internal consistencies for positive (α = 0.72; MIC = 0.46), negative (α = 0.89; MIC = 0.77), and depressive (α = 0.70; MIC = 0.49) symptoms in the present study.
Previous research has shown that childhood trauma contributes to the onset and maintenance of psychosis. However, few studies have accounted for the effects of lifetime trauma and post-traumatic stress disorder (PTSD), and none have examined the mediating role of emotion dysregulation in symptom maintenance after severe trauma. The purpose of this study is to determine whether maladaptive cognitive emotion regulation strategies (CERS) and global emotion dysregulation mediate the effects of probable PTSD on depressive symptoms, and whether this pathway extends to influence positive symptoms in patients with early non-affective psychotic disorders.
A total of 150 outpatients with early non-affective psychosis were assessed for trauma exposure, DSM-5 PTSD symptoms, CERS, global emotion dysregulation, and current depressive and positive symptoms. Parallel and serial mediation analyses based on ordinary least squares regressions were used to test the hypothesized models.
Mediation analyses controlling for gender, psychiatric comorbidities, antipsychotic medication dosage, duration of untreated psychosis (DUP), family history of mental illness, and cumulative trauma revealed that maladaptive CERS (rumination, catastrophic thinking, and self-blame) and global emotion dysregulation mediated the effects of probable PTSD on depressive symptoms (R² = 41%), while maladaptive CERS, global emotion dysregulation, and depressive symptoms mediated the effects of probable PTSD on positive symptoms (R² = 30%).
Our results demonstrate the indirect effects of maladaptive CERS and global emotion dysregulation on maintaining depressive and positive symptoms. Emotion dysregulation may be a potential transdiagnostic treatment target to alleviate depressive and positive symptoms in traumatized patients with early non-affective psychosis.

View all citing articles on Scopus

View full text

Factor structure of the Brief Psychiatric Rating Scale – Expanded (BPRS-E) in a large hospitalized sample

Abstract

Section snippets

Participants

Results

Discussion

Author note

J. Psychiatr. Res.

Encephale

J. Affect. Disord.

Compr. Psychiatry

Schizophr. Res.

J. Affect. Disord.

Psychiatry Res.

Schizophr. Res.

Psychiatry Res.

Psychiatry Res.

Psychiatry Res.

Predicting inpatient length of stay with the expanded version of the expanded Brief Psychiatric Rating Scale

Psychiatr. Serv.

The BPRS-E as a predictor of length of stay in a residential facility

Psychiatr. Serv.

Factor structure of the Expanded Brief Psychiatric Rating Scale

J. Clin. Psychol.

Consistency of symptomatic dimensions of schizophrenia in a 12 year follow up study

Arch. Psychiatry Psychother.

Symptom dimensions as predictors of clinical outcome, duration of hospitalization, and aggressive behaviors in acutely hospitalized patients with psychotic exacerbation

Clin. Pract. Epidemiol. Ment. Health

Confirmatory Factor Analysis of the Brief Psychiatric Rating Scale - Expanded Version

Psychopathological assessment of risk of restraint in acute psychiatric patients

J. Nerv. Ment. Dis.

Component structure of the Expanded Brief Psychiatric Rating Scale (BPRS-E)

Psychopharmacology

Factor structure of the BPRS in deaf people with schizophrenia: correlates to language and thought

Cogn. Neuropsychiatry