Elsevier

Schizophrenia Research

Volume 201, November 2018, Pages 85-90
Schizophrenia Research

Factor structure of the positive and negative syndrome scale (PANSS) in people at ultra high risk (UHR) for psychosis

https://doi.org/10.1016/j.schres.2018.05.024Get rights and content

Abstract

Introduction

The Positive and Negative Syndrome Scale (PANSS), a comprehensive psychopathology assessment scale used in the evaluation of psychopathology in schizophrenia, is also often used in the Ultra-High-Risk (UHR) population. This paper examined the dimensional structure of the PANSS in a UHR sample.

Methods

A total of 168 individuals assessed to be at UHR for psychosis on the Comprehensive Assessment of At-Risk Mental States (CAARMS) were evaluated on the PANSS, Calgary Depression Scale for Schizophrenia (CDSS), Beck Anxiety Inventory (BAI), Brief Assessment of Cognition in Schizophrenia (BACS), and Global Assessment of Functioning (GAF). Exploratory factor analysis (EFA) of the PANSS was performed to identify the factorial structure. Convergent validity was explored with the CAARMS, CDSS, BAI and BACS.

Results

EFA of the PANSS yielded five symptom factors - Positive, Negative, Cognition/Disorganization, Anxiety/Depression, and Hostility. This 5-factor solution showed good convergent validity with the CAARMS composite score, CDSS, BAI, and BACS. Positive, Negative and Anxiety/Depression factors were associated with functioning.

Conclusion

The reported PANSS factor structure may serve to improve the understanding and measurement of clinical symptom dimensions manifested in people with UHR for future research and clinical setting.

Introduction

Research studies conducted on people at ultra-high risk (UHR) for psychosis (Yung and McGorry, 1996) often utilize assessment instruments employed in schizophrenia. One such example is the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987), a 30-item scale with three subscales including positive symptoms, negative symptoms, and general psychopathology, which comprehensively assesses symptoms in schizophrenia. The PANSS has been the most widely used symptom severity rating scale in psychosis (van van Veelen and Sommer, 2014) for the management of clinical outcomes and intervention efficacy (Citrome et al., 2011; Dragioti et al., 2016; Emsley et al., 2003; Lindenmayer et al., 1994; Marder et al., 1997).

Research on the UHR population utilizing PANSS mostly reported the scores of the three subscales (Fusar-Poli et al., 2010; Mossaheb et al., 2013; van Rijn et al., 2011). However, past research have found the three subscales to be over simplified and could not adequately represent the structure of PANSS in schizophrenia (Kay and Sevy, 1990; Wallwork et al., 2012). Subsequent factor analytic works reported a few diverse multidimensional solutions of PANSS structure in schizophrenia, ranging from four- to seven-factor models (Emsley et al., 2003; Kay and Sevy, 1990; Van den Oord et al., 2006; White et al., 1997). Majority of those studies provided support for a five factor PANSS model (Jiang et al., 2013; Lehoux et al., 2009; Lindenmayer et al., 1994; Wallwork et al., 2012), but the lack of a broad consensus on the items loaded in each domain remains a pertinent issue (Wallwork et al., 2012). These discrepancies may, in part, be attributed to the different statistical methods used, the removal of items with multiple loadings (van der Gaag et al., 2006a; van der Gaag et al., 2006b), course of illness progression (Dollfus and Petit, 1995; Dragioti et al., 2016; Drake et al., 2003), the lack of validation in an independent sample, and the influence of cultural differences on clinical presentation (Jiang et al., 2013).

To our knowledge, no study has elucidated the dimensional structure of PANSS in UHR, and its use in describing the UHR phenomenology. While individuals are categorized to be at UHR for psychosis mainly from intensity and frequency of attenuated psychotic symptoms, studies have also shown that UHR individuals manifest other psychopathology such as negative symptoms, depressive and anxiety features that are associated with functional deficits at baseline and follow-up (Fusar-Poli et al., 2014a; Lim et al., 2015). Moreover, most of these UHR individuals do not transit to full-blown psychosis (Fusar-Poli et al., 2012). In light of this, concerns have been raised to the specificity of UHR construct as a risk for psychosis only or a risk for any psychiatric disorder, calling the need to extend the conceptual operationalization of the UHR construct, from a single psychosis dimension to a general psychopathology model (Fusar-Poli et al., 2014b; McGorry and van Os, 2013; Yung et al., 2010). Instead of delineating UHR individuals for specific psychopathology, the general psychopathology model posits the co-occurrence of non-specific sub-threshold expression of psychotic and non-psychotic symptoms that deteriorate, eventually leading to a severe psychopathology (Fusar-Poli et al., 2014b). In this regard, a dimensional approach could serve useful in explaining the UHR phenomenology by serving as phenotypic markers of symptom dimension, severity, and illness progression which could aid in the management of prominent at-risk presentation of pre-nosographic conditions, as well as provide a better understanding of the biological underpinnings of UHR psychopathology.

The aim of the current study is to explore the factor structure of PANSS in people with UHR. To our knowledge, no study has sought to examine a valid PANSS structure model for an accurate representation of sub-threshold clinical manifestation in UHR. The concurrent validity of PANSS and its correlations with functional outcomes would also be examined.

Section snippets

Study samples

The study sample consisted of a group of participants from the Longitudinal Youth at Risk Study (LYRIKS). Details of the study sample (Lee et al., 2013) and recruitment strategies (Mitter et al., 2014) have been described in detail previously. Briefly, LYRIKS is a prospective, observational study assessing the risk factors of psychosis (Lee et al., 2013; Lim et al., 2015). Recruitment adopted a hybrid approach where both help-seeking and non-help-seeking individuals from the community were

Descriptive and descriptive data

The demographic and the clinical characteristics of the UHR sample are shown in Table 1.

Building of PANSS factor model

Based on the fit indices (Table 2), the 5 to 7 factor models for PANSS fit the UHR sample. However, the loadings on the 7-factor model and 6-factor model were small. In the 7-factor model, one factor contained only one item (G2 Anxiety) reaching a loading of >0.4. Similarly, in the 6-factor model, one factor consisted of only two items (G2 anxiety and G4 Tension). Therefore, we selected the 5-factor model.

Discussion

In this study we derived five factors - positive, negative, cognitive/disorganization, anxiety/depression and hostility. The five factor model included 27 PANSS items with one item being loaded on two factors. Four of the derived factors (i.e., positive, negative, cognitive/disorganization, and anxiety/depression) were concordant with prodromal symptom structures in UHR samples as assessed on the Scale of Prodromal Symptoms (SOPS; Klaassen et al., 2011). The item of “motor retardation” cross

Funding source

This research was supported by the National Research Foundation Singapore under the National Medical Research Council Translational and Clinical Research Flagship Programme (Grant No.: NMRC/TCR/003/2008), Singapore Ministry of Health's National Medical Research Council under the Centre Grant Programme (Grant No.: NMRC/CG/004/2013).

Contributors

Study concept and design: Zixu Yang, Keane Lim, Max Lam, Jimmy Lee.

Acquisition of data: Zixu Yang.

Drafting of the manuscript: Zixu Yang, Keane Lim, Jimmy Lee.

Statistical analysis: Zixu Yang, Keane Lim, Max Lam.

Critical revision of the manuscript for important content: Zixu Yang, Keane Lim, Max Lam, Richard Keefe, Jimmy Lee.

Final approval of the version to be published: Zixu Yang, Keane Lim, Max Lam, Richard Keefe, Jimmy Lee.

Conflict of interest

Dr. Keefe has received investigator-initiated research funding support from the Department of Veterans Affairs, the National Institute of Mental Health, and the Singapore National Medical Research Council; received honoraria and serve as a consultant, speaker, or advisory board member for Abbvie, Acadia, Akebia, Akili, Alkermes, Astellas, Asubio, Avanir, AviNeuro/ChemRar, Axovant, Biogen, Boehringer Ingelehim, Cerecor, CoMentis, FORUM, Global Medical Education, GW Pharmaceuticals, Intracellular

Acknowledgements

The authors would like to express their sincere gratitude to Dr. Gurpreet Rekhi and Dr. Lim Sheng Foong, Caroline who trained and supervised research psychologists in conducting clinical assessments and all of the participants who participated in this study. This research was supported by the National Research Foundation Singapore under the National Medical Research Council Translational and Clinical Research Flagship Programme (Grant No.: NMRC/TCR/003/2008), Singapore Ministry of Health's

References (57)

  • J. Jiang et al.

    Validated five-factor model of positive and negative syndrome scale for schizophrenia in Chinese population

    Schizophr. Res.

    (2013)
  • R.S. Keefe et al.

    The brief assessment of cognition in schizophrenia: reliability, sensitivity, and comparison with a standard neurocognitive battery

    Schizophr. Res.

    (2004)
  • M.E. Kelley et al.

    Subscale structure for the positive and negative syndrome scale (PANSS): a proposed solution focused on clinical validity

    Psychiatry Res.

    (2013)
  • C. Lancon et al.

    Stability of the five-factor structure of the positive and negative syndrome scale (PANSS)

    Schizophr. Res.

    (2000)
  • J. Lee et al.

    The Longitudinal Youth at Risk Study (LYRIKS)–an Asian UHR perspective

    Schizophr. Res.

    (2013)
  • J. Lim et al.

    Impact of psychiatric comorbidity in individuals at ultra high risk of psychosis – findings from the Longitudinal Youth at Risk Study (LYRIKS)

    Schizophr. Res.

    (2015)
  • P. McGorry et al.

    Redeeming diagnosis in psychiatry: timing versus specificity

    Lancet

    (2013)
  • N. Mossaheb et al.

    Effect of omega-3 fatty acids for indicated prevention of young patients at risk for psychosis: when do they begin to be effective?

    Schizophr. Res.

    (2013)
  • B. Nelson et al.

    Ultrahigh risk (UHR) for psychosis criteria: are there different levels of risk for transition to psychosis?

    Schizophr. Res.

    (2011)
  • A. Raballo et al.

    The comprehensive assessment of at-risk mental states: from mapping the onset to mapping the structure

    Schizophr. Res.

    (2011)
  • G. Rutigliano et al.

    Persistence or recurrence of non-psychotic comorbid mental disorders associated with 6-year poor functional outcomes in patients at ultrahigh risk for psychosis

    J. Affect. Disord.

    (2016)
  • E.J. Van den Oord et al.

    Factor structure and external validity of the PANSS revisited

    Schizophr. Res.

    (2006)
  • E. Velthorst et al.

    Quantitative and qualitative symptomatic differences in individuals at ultra-high risk for psychosis and healthy controls

    Psychiatry Res.

    (2013)
  • R.S. Wallwork et al.

    Searching for a consensus five-factor model of the positive and negative syndrome scale for schizophrenia

    Schizophr. Res.

    (2012)
  • A.R. Yung et al.

    The psychosis threshold in ultra high risk (prodromal) research: is it valid?

    Schizophr. Res.

    (2010)
  • D. Addington et al.

    Assessing depression in schizophrenia: the Calgary depression scale

    Br. J. Psychiatry

    (1993)
  • A.T. Beck et al.

    An inventory for measuring clinical anxiety: psychometric properties

    J. Consult. Clin. Psychol.

    (1988)
  • M.W. Best et al.

    Examination of the positive and negative syndrome scale factor structure and longitudinal relationships with functioning in early psychosis

    Early Interv. Psychiatry

    (2016)
  • Cited by (11)

    • Large-scale evaluation of the Positive and Negative Syndrome Scale (PANSS) symptom architecture in schizophrenia

      2021, Asian Journal of Psychiatry
      Citation Excerpt :

      The schizophrenia sample characteristics (N = 3511 in the current EFA analysis) are reported in Table 1. Demographic details of the UHR sample were as reported by Yang et al. (2018). Fit indices (Table 2) indicated that the 4–7 PANSS factor models fitted the full data.

    • Negative symptom configuration within and outside schizophrenia spectrum disorders: results from the “Parma Early Psychosis” program.

      2020, Psychiatry Research
      Citation Excerpt :

      The assessment was conducted by specialized Pr-EP team members, who undergone regular PANSS supervision sessions and scoring workshops to ensure inter-rater reliability. Within factor analysis research in the full spectrum of psychosis, an agreed-upon comprehensive model of the PANSS negative symptom configuration has not yet been unanimously ascertained (Yang et al., 2018). An interesting item response analysis study reported that together with the 7 PANSS items traditionally included in the Marder's “Negative Symptom Factor” [NSF] (i.e. blunted affect, emotional withdrawal, poor rapport, passive/apathetic social withdrawal, lack of spontaneity/flow of conversation, Motor retardation and active social withdrawal) (Marder et al., 2011), at least one further item belonging to the PANSS “General Psychopathology” domain (i.e. disturbance of volition) contributed in a psychometrically meaningful way to the construct of a negative syndrome in psychosis research (Khan et al., 2013).

    • Brain changes associated with negative symptoms in clinical high risk for psychosis: A systematic review

      2020, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Firstly, different scales are used to determine whether someone is at-risk for psychosis. Although these scales often display substantial agreement in the identification of at-risk subjects (Fusar-Poli et al., 2016; Yang et al., 2018), they also show differences in sensitivity (Oliver et al., 2018), as well as differences in subgroup identification (Fusar-Poli et al., 2016). It is unclear how or whether these differences affected the samples examined in this review, but care is warranted when comparing studies using different criteria for identifying the sample to be included.

    View all citing articles on Scopus
    View full text