The Schizophrenia Proneness Instrument, Child and Youth version (SPI-CY): Practicability and discriminative validity

https://doi.org/10.1016/j.schres.2013.02.014Get rights and content

Abstract

Background

Basic symptom (BS) criteria have been suggested to complement ultra-high risk (UHR) criteria in the early detection of psychosis in adults and in children and adolescents. To account for potential developmental particularities and a different clustering of BS in children and adolescents, the Schizophrenia Proneness Instrument, Child and Youth version (SPI-CY) was developed.

Aims

The SPI-CY was evaluated for its practicability and discriminative validity.

Method

The SPI-CY was administered to 3 groups of children and adolescents (mean age 16; range = 8–18; 61% male): 23 at-risk patients meeting UHR and/or BS criteria (AtRisk), 22 clinical controls (CC), and 19 children and adolescents from the general population (GPS) matched to AtRisk in age, gender, and education. We expected AtRisk to score highest on the SPI-CY, and GPS lowest.

Results

The groups differed significantly on all 4 SPI-CY subscales. Pairwise post-hoc comparisons confirmed our expectations for all subscales and, at least on a descriptive level, most items. Pairwise subscale differences indicated at least moderate group effects (r  0.37) which were largest for Adynamia (0.52  r  0.70). Adynamia also performed excellent to outstanding in ROC analyses (0.813  AUC  0.981).

Conclusion

The SPI-CY could be a helpful tool for detecting and assessing BS in the psychosis spectrum in children and adolescents, by whom it was well received. Furthermore, its subscales possess good discriminative validity. However, these results require validation in a larger sample, and the psychosis-predictive ability of the subscales in different age groups, especially the role of Adynamia, will have to be explored in longitudinal studies.

Introduction

The early detection and treatment of persons at risk of developing psychosis is a promising strategy in fighting the devastating consequences of these disorders (Fusar-Poli et al., 2012, Fusar-Poli et al., 2013). Two sets of at-risk criteria, the “ultra-high risk” (UHR) and the “basic symptom” (BS) criteria (Table 1), have been developed and validated on mainly adult samples and currently prevail in at-risk assessments (Fusar-Poli et al., 2012, Fusar-Poli et al., 2013, Schultze-Lutter et al., 2012a, Schultze-Lutter et al., 2013).

However, there is little empirical evidence that these at-risk criteria can simply be transferred to children and adolescents (CAD; Koch et al., 2010, Schimmelmann and Schultze-Lutter, 2012). Regarding the UHR criteria, there are indications that some attenuated psychotic symptoms are potentially nonspecific in adolescents; and brief limited intermittent symptoms might be difficult to clinically classify in children when observable behavioral correlates are missing. Thus, CAD mental health clinicians have reported difficulty in discriminating normal adolescent behaviors from behaviors suggestive of emerging psychosis according to UHR criteria (Welsh and Tiffin, 2012).

For BSs, i.e., subtle, subjectively experienced subclinical disturbances (Schultze-Lutter, 2009), preliminary results indicate that, as in adult populations, cognitive BSs may be promising candidates for at-risk criteria in CAD (Meng et al., 2009). However, dimensional analyses have indicated a completely different structure of BSs in CAD compared to adults (Schultze-Lutter et al., 2008, Schultze-Lutter et al., 2012a). Adynamic BSs, i.e., disturbances in drive, stress tolerance, affect, and thought energy, appeared to play a central role in adolescents, but in adults, they were not specific to psychosis, and the central position was held by cognitive BSs, i.e., disturbances in thought and speech processes (Schultze-Lutter et al., 2012a). Furthermore, potential developmental peculiarities should be considered in BS assessment, and parents' reports might facilitate assessment, particularly in children and young adolescents (Koch et al., 2010, Schultze-Lutter et al., 2012a). Therefore, the Schizophrenia Proneness Instrument, Child and Youth version (SPI-CY; Schultze-Lutter et al., 2012b) was developed to assess BSs in 8- to 18-year-olds. Although its clinical application is not restricted to the early detection of psychosis, it is currently the only instrument in this field specifically developed on and for CAD.

We explored the discriminative validity and clinical practicability of the SPI-CY. In line with earlier BS studies in adults (Klosterkötter et al., 1996, Klosterkötter et al., 2001, Schultze-Lutter et al., 2007a) and in adolescents (Resch et al., 1998, Meng et al., 2009), we expected that CAD symptomatically at risk of psychosis would exhibit not only a greater number of but also more severe BSs than a clinical control inpatient sample who, in turn, would report more frequent and severe BSs than a general population sample.

Section snippets

Sample

We compared 3 groups recruited in 2008 and 2009: 2 clinical samples that consisted of in- and outpatients of the Department of Child and Adolescent Psychiatry, University of Cologne, and a nonclinical general population sample. One clinical group consisted of in- and outpatients suspected to be at an increased risk of psychosis, referred to a specialized diagnostic service, and found to meet UHR criteria according to the Structured Interview for Prodromal Syndromes (SIPS 3.0; McGlashan et al.,

SPI-CY subscales

The 3 groups differed significantly in the expected direction on all subscale totals of the SPI-CY (Fig. 1): Adynamia (χ2(2) = 39.110, p < 0.001), Perception Disturbances (χ2(2) = 27.805, p < 0.001), Neuroticism (χ2(2) = 21.432, p < 0.001), and Thought and Motor Disturbances (χ2(2) = 31.464, p < 0.001). All post-hoc comparisons between AtRisk and CC and between CC and GPS showed significant differences of moderate to large effect size for all subscales (Fig. 1; 42.0  U  142.5; 0.001  p  0.01), with the highest

Discussion

The present pilot study explored the practicability and discriminative validity of the SPI-CY, developed especially for use in CAD. Three groups were compared: AtRisk, CC, and GPS. Regarding the selection bias in favor of AtRisk to score higher than the other groups, the exclusive and moderate correlation between Thought and Motor Disturbances and COPER indicates that this bias affected the results to only a limited degree.

Conclusion

This is the first pilot study on the discriminative validity and practicability of the SPI-CY. It indicates that the SPI-CY is well received by young patients and their clinicians and suggests that the instrument is a helpful tool in detecting CAD who are symptomatically at risk of psychosis. The SPI-CY might also be a helpful tool to guide clinical decisions about UHR status in cases in which differentiating between aspects of normal adolescent behaviors and behaviors suggestive of an emerging

Role of funding source

No funding was received in support of the presented study.

Contributors

LF, PW, and FSL designed the study; LF and FSL managed the literature searches, undertook the statistical analyses, and wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflict of interest.

Acknowledgment

The paper is based on the medical thesis of LF, which was supervised by PW and FSL.

References (24)

  • H. Meng et al.

    Basic symptoms in the general population and in psychotic and non-psychotic psychiatric adolescents

    Schizophr. Res.

    (2009)
  • B.G. Schimmelmann et al.

    What percentage of people in the general population satisfies the current clinical at-criteria of psychosis?

    Schizophr. Res.

    (2011)
  • B.A. Cornblatt

    The New York high risk project to the Hillside recognition and prevention (RAP) program

    Am. J. Med. Genet.

    (2002)
  • P. Fusar-Poli et al.

    Predicting psychosis: meta-analysis of transition outcomes in individuals at high clinical risk

    Arch. Gen. Psychiatry

    (2012)
  • P. Fusar-Poli et al.

    The psychosis high-risk state: a comprehensive state-of-the-art review

    JAMA Psychiatry

    (2013)
  • G. Gross et al.

    Bonner Skala für die Beurteilung von Basissymptomen (BSABS)

    (1987)
  • D.W. Hosmer et al.

    Applied Logistic Regression

    (2000)
  • J. Klosterkötter et al.

    Diagnostic validity of basic symptoms

    Eur. Arch. Psychiatry Clin. Neurosci.

    (1996)
  • J. Klosterkötter et al.

    Diagnosing schizophrenia in the initial prodromal phase

    Arch. Gen. Psychiatry

    (2001)
  • E. Koch et al.

    On the importance and detection of prodromal symptoms from perspective of child and adolescent psychiatry

    Clin. Neuropsychiatry

    (2010)
  • T.H. McGlashan et al.

    Structured Interview for Prodromal Syndromes (Version 3.0)

    (2001)
  • C.J. Morgan et al.

    Neurocognitive function and schizophrenia-proneness in individuals dependent on ketamine, on high potency cannabis (‘skunk’) or on cocaine

    Pharmacopsychiatry

    (2012)
  • Cited by (0)

    1

    Indicates joint first-authorship.

    View full text