Format

Send to

Choose Destination
Psychother Psychosom Med Psychol. 2018 Aug;68(8):346-352. doi: 10.1055/s-0043-118335. Epub 2017 Sep 28.

[Psychometric Properties of the German Version of the Suicide Behaviors Questionnaire Revised (SBQ-R)].

[Article in German; Abstract available in German from the publisher]

Author information

1
Abteilung für Medizinische Psychologie und Medizinische Soziologie, Universität Leipzig.
2
Universitätsklinik für Psychoanalyse und Psychotherapie, Medizinische Universität Wien, Österreich.
3
Arbeitseinheit Klinische Psychologie und Psychotherapie, Fakultät für Psychologie, Ruhr-Universität Bochum.
4
Klinische Psychologie, MSB Medical School Berlin.
5
Institut für Medizinische Psychologie und Medizinische Soziologie, Universitätsklinikum der RWTH Aachen.

Abstract

in English, German

OBJECTIVES:

The Suicide Behaviors Questionnaire-Revised (SBQ-R) was developed by Osman et al. (2001) to economically assess different aspects of suicidality. It consists of 4 items. There are several critical points concerning item construction: (1) temporal frame (e. g. retrospective vs. prospective) and (2) response labels of item 1 and 3 (e. g. 2 alternatives with the same scoring). Information about psychometric properties of the German version of the SBQ-R were not available until now. The SBQ-R is assumed to be a unidimensional measure and is evaluated using a total score, although its dimensionality has never been tested.

METHODS:

The SBQ-R and several measures for convergent validity were assessed in a representative sample of the German general population (N=2497). Convergent validity was tested against symptoms of depression and anxiety (PHQ-4), as well as the core constructs of the Interpersonal Theory of Suicidal Behavior (IPTS) (INQ, ACSS-FAD). Unidimensionality of the SBQ-R was tested using confirmatory factor analysis and intercorrelations with the convergent measures were calculated.

RESULTS:

The initial model fit of the unidimensional solution was unsatisfactory. After inclusion of a correlated error term of item 2 and item 4 based on modification indices, the model fit was very good. Moreover the SBQ-R showed satisfactory internal consistency (Cronbach's α=0.72). With the exception of the ACSS-FAD, all correlations with convergent measures were according to the theoretically proposed expectations.

CONCLUSION:

In principle the findings on the psychometric properties justify the use of the SBQ-R, although several critical points concerning item scoring remain unresolved. Moreover, sufficient unidimensional model fit was not possible without including correlated errors. Since suicidality is a frequently assessed construct in suicide research and as an exclusion criterion in clinical studies, it would be desirable to develop an economic, psychometrically sound and compelling instrument for the assessment of the different aspects of suicidality in future.

PMID:
28958123
DOI:
10.1055/s-0043-118335
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Georg Thieme Verlag Stuttgart, New York
Loading ...
Support Center