Elsevier

Schizophrenia Research

Volume 228, February 2021, Pages 373-381
Schizophrenia Research

The psychometric validity of the Montgomery–Åsberg Depression Rating Scale (MADRS) in recent onset schizophrenia spectrum disorders

https://doi.org/10.1016/j.schres.2020.12.015Get rights and content

Abstract

Earlier recognition and accurate assessment of depressive symptoms is important to improving outcomes in individuals with recent-onset schizophrenia spectrum disorders (termed SSD hereafter)—regardless of whether positive psychotic symptoms are present or have resolved. The Montgomery–Åsberg Depression Rating Scale (MADRS) is frequently used to assess depressive symptoms in SSD, but no study has examined the psychometric validity of MADRS scores in individuals exclusively with SSD and sub-grouped by those with and without positive psychotic symptoms. This study involved baseline data from the Psychosis Recent Onset GRoningen Survey (PROGR-S). Measures used were: MADRS, depressive and negative subscales of Positive and Negative Syndrome Scale (PANSSD, PANSSN), and Schedules for Clinical Assessment in Neuropsychiatry (SCAN). The MADRS total score had sufficient concurrent validity with PANSSD (evidence by ρ≥0.70), and insufficient divergent validity with PANSSN (evidenced by ρ ≥0.30), in the full cohort and when sub-grouped by positive psychotic symptoms. In symptom networks, divergent communities comprising either MADRS or PANSSN items were found, except the MADRS item inability to feel overlapped with PANSSN items. The most divergent MADRS items were sadness, pessimism, and suicidal thoughts. The MADRS total score had sufficient predictive validity for determining caseness for MDD based on SCAN, but the optimal cut-off differed in those with and without positive psychotic symptoms (MADRS≥18 versus MADRS≥11). The MADRS has sufficient validity for assessing depressive symptoms in SSD. Since scores might depend upon symptoms of SSD, MADRS≥11 and the presence of sadness, pessimism, or suicidal ideation might be the best indicator of MDD in SSD.

Introduction

Co-occurring depressive symptoms significantly contribute to the already large burden of recent-onset schizophrenia spectrum disorders (termed SSD hereafter; Charleson et al., 2016; Crumlish et al., 2005; Dutta et al., 2011; Rossler et al., 2005). They occur in up to 80% of individuals with such disorders (Herniman et al., 2019), and are associated with serious adverse outcomes, including suicide (Conley et al., 2007). Depressive symptoms are often first recognized after the resolution of positive psychotic symptoms (Sandhu et al., 2013), commonly known as post-psychotic depression [PDD] and defined in the International Classification of Diseases (ICD, F20.4; World Health Organization, 1992), in large part due to assumptions that they are transient in nature, or likely to follow and therefore resolve in line with positive psychotic symptoms, particularly after antipsychotic pharmacotherapy (Siris, 2000). However, there is now robust evidence indicating that depressive symptoms are more common and severe when positive psychotic symptoms are present (compared to post-psychotic phase; Herniman et al., 2019), rarely occur de novo after the resolution of positive psychotic symptoms (Upthegrove et al., 2010), and might even drive the initial development and maintenance of positive psychotic symptoms in SSD (Kuipers et al., 2006; Smith et al., 2006). Earlier recognition and accurate assessment and continuous monitoring of depressive symptoms is therefore critical to improving outcomes in individuals with SSD—regardless of whether positive psychotic symptoms are present or have resolved. This might be particularly critical for those with treatment resistant positive psychotic symptoms (Kuipers et al., 2006).

Accurate assessment and continuous monitoring of depressive symptoms is difficult in SSD due to phenomenological overlap or difficulty distinguishing between depressive symptoms and negative symptoms of SSD (Chiappelli et al., 2014). While the Calgary Depression Scale for Schizophrenia (CDSS; Addington et al., 1990) is the only instrument that was deigned to assess depressive symptoms specifically in SSD, the Montgomery–Åsberg Depression Rating Scale (Montgomery and Asberg, 1979) nonetheless remains extensively used to assess depressive symptoms in SSD (Chang et al., 2014; Chang et al., 2011; Cotton et al., 2010; Herniman et al., 2019; Lee et al., 2017; Pena et al., 2011; Zabala et al., 2017). Like most depression severity scales, the MADRS was designed for use in individuals with depressive disorders but not SSD (Montgomery and Asberg, 1979)—presenting potential validity issues. Some studies have determined the validity of MADRS scores in established/chronic SSD and schizoaffective disorder (Bernard et al., 1998; Fitzgerald et al., 2002; Lancon et al., 2000; Lee et al., 2003; Liu et al., 2009; Sarro et al., 2004), and recent-onset SSD and schizoaffective disorder (Wolthaus et al., 2000). However, to date, no study has examined the psychometric validity of MADRS scores in individuals exclusively with recent-onset SSD, and further sub-grouped by those with or without positive psychotic symptoms.

It is necessary to specifically examine the psychometric validity of the MADRS in those exclusively with SSD and further sub-grouped by those with and without positive psychotic symptoms. The inclusion of affective psychotic disorders such as schizoaffective disorders may confound the relation between illness characteristics and depressive symptomatology per se (Cotton et al., 2012; Herniman et al., 2019). Affective psychotic disorders might also be inherently different to SSD, with potentially different illness trajectories and outcomes (Cotton et al., 2013; Jager et al., 2011). Results from previous studies including affective psychotic disorders might have therefore conflated and consequently misrepresented the performance of the MADRS in SSD. It is also important to examine the psychometric validity of MADRS scores sub-grouped by those with and without positive psychotic symptoms, as the presentation of depressive symptoms—and therefore the interpretation of MADRS items—might differ with or without positive psychotic symptoms (Nakaya et al., 1998). Additionally, since depressive symptoms might be causally related to positive psychotic symptoms (Kuipers et al., 2006; Smith et al., 2006), the fluctuating severity and sequalae of positive psychotic symptoms might also impact depressive symptoms and therefore the interpretation of MADRS scores in recent-onset SSD (Herniman et al., 2019). Thus, despite the already extensive use of the MADRS in SSD, a psychometric examination of its validity in individuals exclusively with SSD and sub-grouped by those with and without positive psychotic symptoms is warranted.

The aim of the current study was to determine the psychometric validity of MADRS scores (concurrent, divergent, and predictive) in individuals exclusively with recent onset SSD, and sub-grouped by those with and without positive psychotic symptoms.

Section snippets

Participants

This study involved analysis of baseline data derived from the Psychosis Recent Onset GRoningen Survey (PROGR-S; Liemburg et al., 2014). The eligibility criteria for PROGR-S have been described elsewhere (Liemburg et al., 2014). In brief, participants resided in the Groningen province of the Netherlands and were referred to a psychiatric institution with a suspected recent-onset psychotic episode (<2 years) that was yet to be treated, or emergency treatment was recently initiated allowing for

Participant flow

Six hundred and fifty-nine of the 718 (92%) PROGR-S participants had complete MADRS data and were therefore eligible for the current study (Liemburg et al., 2014). Of the 659, 28.4% (n=187) did not meet additional eligibility criteria and were excluded from the current analysis. Reasons for exclusion included: ≥40 years of age (n=66); no DSM-IV Axis 1 diagnosis (n=10); no psychotic disorder (n=62); organic psychotic disorder (n=1); and affective psychotic disorder (schizoaffective disorder [n

Discussion

Depressive symptoms significantly contribute to the already large burden of recent-onset schizophrenia spectrum disorders (SSD), yet are often first recognized only after the resolution of positive psychotic symptoms (Sandhu et al., 2013). Earlier recognition and accurate assessment of depressive symptoms in SSD—regardless of whether positive psychotic symptoms are present or resolved—is critical to improving outcomes (Herniman et al., 2019; Upthegrove et al., 2017). To our knowledge, this is

Role of funding source

SEH is supported by a Research Training Program Scholarship awarded by the Australian Commonwealth Government. SMC is supported by a National Health and Medical Research Council Senior Research Fellowship (APP1136344). KA is supported by a Career Development Fellowship from the NHMRC (1141207). The framework of the PROGR-S-cohort is supported by the University Centre of Psychiatry of the University Medical Centre Groningen, Lentis Mental Health Institute, Groningen, and the Rob Giel Research

CRediT authorship contribution statement

All authors have contributed to and approved the final manuscript.

Declaration of competing interest

The authors have no conflicts of interest to declare.

Acknowledgements

We would like to acknowledge and thank all PROGR-S participants and clinicians for supporting the study.

References (62)

  • M. Jager et al.

    Schizoaffective disorder—an ongoing challenge for psychiatric nosology

    European Psychiatry

    (2011)
  • S.W. Kim et al.

    Diagnostic validity of assessment scales for depression in patients with schizophrenia

    Psychiatry Res.

    (2006)
  • V.P. Kontaxakis et al.

    Comparison of four scales measuring depression in schizophrenic inpatients

    European Psychiatry

    (2000)
  • I.M. Lako et al.

    A systematic review of instruments to measure depressive symptoms in patients with schizophrenia

    J. Affect. Disord.

    (2012)
  • C. Lancon et al.

    Study of the concurrent validity of the Calgary Depression Scale for Schizophrenics (CDSS)

    J. Affect. Disord.

    (2000)
  • J. Pena et al.

    Executive functioning correctly classified diagnoses in patients with first-episode psychosis: evidence from a 2-year longitudinal study

    Schizophr. Res.

    (2011)
  • M.W. Post

    What to do with “moderate” reliability and validity coefficients?

    Arch. Phys. Med. Rehabil.

    (2016)
  • W. Rossler et al.

    Size of burden of schizophrenia and psychotic disorders

    Eur. Neuropsychopharmacol.

    (2005)
  • A. Sandhu et al.

    The subjective experience and phenomenology of depression following first episode psychosis: a qualitative study using photo-elicitation

    J. Affect. Disord.

    (2013)
  • S. Sarro et al.

    Cross-cultural adaptation and validation of the Spanish version of the Calgary Depression Scale for Schizophrenia

    Schizophr. Res.

    (2004)
  • B. Smith et al.

    Emotion and psychosis: links between depression, self-esteem, negative schematic beliefs and delusions and hallucinations

    Schizophr. Res.

    (2006)
  • R. Upthegrove et al.

    Depression in first episode psychosis: the role of subordination and shame

    Psychiatry Res.

    (2014)
  • American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders (4th Ed.). American...
  • D. Bernard et al.

    Calgary Depression Scale for Schizophrenia: a study of the validity of a French-language version in a population of schizophrenic patients

    Acta Psychiatr. Scand.

    (1998)
  • F.J. Charleson et al.

    Global epidemiology and burden of schizophrenia: findings from the global burden of disease study 2016

    Schizophr. Bull.

    (2016)
  • J. Chen et al.

    Extended Bayesian information criteria for model selection with large model spaces

    Biometrika

    (2008)
  • T. Craig et al.

    Measurement of mood and affect in schizophrenic inpatients

    Am. J. Psychiatr.

    (1985)
  • A.O. Cramer et al.

    Comorbidity: a network perspective

    Behav Brain Sci

    (2010)
  • N. Crumlish et al.

    Early insight predicts depression and attempted suicide after 4 years in first-episode schizophrenia and schizophreniform disorder

    Acta Psychiatr. Scand.

    (2005)
  • C.J. Edwards et al.

    The relationship between depressive symptoms and negative symptoms in people with non-affective psychosis: a meta-analysis

    Psychol. Med.

    (2019)
  • S. Epskamp et al.

    A tutorial on regularized partial correlation networks

    Psychol. Methods

    (2018)
  • Cited by (7)

    • Psychometric instruments for the assessment of depressive symptoms in patients with delusional disorder: A systematic review

      2022, Psychiatry Research
      Citation Excerpt :

      All of them has been considered as a valid and sensitive instrument with controversial opinions regarding its clinical psychometric criteria. Early identification and accurate evaluation of depressive symptoms has been recommended in patients suffering from recent-onset schizophrenia spectrum disorders (Herniman et al., 2021). The Montgomery-Åsberg Depression Rating Scale (MADRS) has shown sufficient validity for the assessment of depressive symptoms in schizophrenia spectrum disorders.

    • Interrelationships between depressive symptoms and positive and negative symptoms of recent onset schizophrenia spectrum disorders: A network analytical approach

      2021, Journal of Psychiatric Research
      Citation Excerpt :

      Second, participants with an affective psychotic disorder, including schizoaffective disorder, major depressive disorder (MDD) and bipolar I disorder with psychotic features, were excluded to avoid confounds between illness characteristics and depressive symptoms per se (Cotton et al., 2012; Herniman et al., 2019). Last, participants were also excluded if they did not currently have clinically-defined depression (defined as a MADRS score >10 with at least one of reported sadness, pessimistic thoughts, or suicidal thoughts scoring at least one (Herniman et al., 2020)) and positive psychotic symptom(s) (defined as at least one score of ≥3 on any PANSSP item (Herniman et al., 2020; Upthegrove et al., 2014)). The final participant group therefore comprised individuals in the psychotic phase of a recent-onset SSD disorder (including schizophrenia, schizophreniform disorder, delusional disorder, brief psychotic disorder, substance-induced psychotic disorder, and psychotic disorder NOS) with co-occurring clinically-defined depression.

    View all citing articles on Scopus
    View full text