CONTACT US: Contact info
|
Citation |
von Glischinski M, Teismann T, Prinz S, Gebauer JE, Hirschfeld G. Clin. Psychol. Psychother. 2016; 23(6): 543-549. |
|
Affiliation |
Osnabrück University of Applied Sciences, Osnabrück, Germany. |
|
Copyright |
(Copyright © 2016, John Wiley and Sons) |
|
DOI |
|
PMID |
26857479 |
|
Abstract |
Suicide is a major cause of death in adulthood and specifically in patients suffering from mental illnesses. The Depressive Symptom Inventory Suicidality Subscale (DSI-SS) is widely used to detect and prevent suicidal ideation. The aim of the present study was to determine optimal cut points for the DSI-SS in different populations. We analysed the data of one population-based sample (n = 532), one outpatient sample (n = 180) and one inpatient sample (n = 244). Internal consistency, convergent validity and optimal cut points according to receiver operating characteristics were calculated. In all samples, we found excellent item-total correlations and internal consistencies for the DSI-SS. Zero-order correlations between the DSI-SS and theoretically related constructs showed positive correlation coefficients, ranging from 0.50 to 0.67. The DSI-SS differentiated well between patients with and without suicide attempts in the population-based sample, but less so in the inpatient sample and only marginally in the outpatient sample. A bootstrapping analysis showed some variability in the cut points that emerged as optimal, but there was no overlap between the different samples. The specific cut points that we identified may be used to improve the diagnostic utility of the DSI-SS and the chance to detect suicidal ideation. Copyright © 2016 John Wiley & Sons, Ltd. KEY PRACTITIONER MESSAGE: We developed cut points for the Depressive Symptom Inventory Suicidality Subscale, to improve the early and valid detection of suicidal ideation by this measure. The cut points that were identified as optimal varied between the samples. The cut points differentiated well in a non-clinical sample, but less well in outpatient and inpatient samples. Language: en |