The five self-harm behavior groupings measure: empirical and thematic data from a novel comprehensive self-harm assessment

Sophie I Liljedahl; Daiva Daukantaitė; Nikolaus Kleindienst; Margit Wångby-Lundh; Sofie Westling

doi:10.3389/fpsyt.2023.1147206

The five self-harm behavior groupings measure: empirical and thematic data from a novel comprehensive self-harm assessment

Front Psychiatry. 2023 May 5:14:1147206. doi: 10.3389/fpsyt.2023.1147206. eCollection 2023.

Authors

Sophie I Liljedahl^{1

2

3}, Daiva Daukantaitė⁴, Nikolaus Kleindienst⁵, Margit Wångby-Lundh⁴, Sofie Westling^{6

7}

Affiliations

¹ Department of Psychiatry for Affective Disorders, National Specialized Medical Care Unit for Severe Self-Harm Behaviour, Sahlgrenska University Hospital, Gothenburg, Sweden.
² Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
³ Finjagården Treatment Center, Finja, Sweden.
⁴ Department of Psychology, Lund University, Lund, Sweden.
⁵ Central Institute of Mental Health, Department of Psychosomatic Medicine and Psychotherapy, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany.
⁶ Office for Psychiatry and Habilitation, Psychiatric Clinic Lund, Lund, Sweden.
⁷ Department of Clinical Sciences Malmö, Psychiatry, Lund University, Lund, Sweden.

Abstract

Introduction: The Five Self-Harm Behavior Groupings Measure (5S-HM) is a novel assessment that evaluates behaviours which may go undetected by existing measures. Self-harm is formulated across directness and lethality spectra, including under-studied behaviors such as indirect self-harm, harmful self-neglect and sexual self-harm. Aims of the study were to: (1) empirically evaluate the 5S-HM; (2) to determine whether the 5S-HM generates relevant new information with respect to the forms and functions given by participants for self-harm within a clinical sample; (3) to test the utility and novel contributions of the Unified Model of Self-Harm and the 5S-HM by extension.

Methods: Data were collected from N = 199 individuals (M_age = 29.98, SD = 8.41, 86.4% female), receiving specialized evidence-based treatments for self-harm, borderline personality disorder or eating disorders. Construct validity was determined via Spearman correlations, and internal consistency was established from Cronbach's alpha. Inductive thematic analysis was used to analyze and interpret qualitative data on reasons, forms and functions participants reported in relation to self-harm following Braun and Clarke's analytic guidelines. Thematic mapping was used to summarize qualitative data.

Results: Test-retest reliability on a subsample of n = 24, tested 14 days after Time 1 was supported by a good intraclass correlation (0.68). Internal consistency (Cronbach's alpha = 0.75) was acceptable to good, as was construct validity comparing the 5S-HM total score to two validated self-harm measures (rho = 0.40, p < 0.01; rho = 0.26, p < 0.01). A thematic map depicting antecedents and consequences of self-harm over time suggests that self-harm is initiated by negative emotional states and self-intolerance. Novel findings in relation to sexual self-harm indicated that reasons for these behaviors were either to improve or worsen one's situation through being hurt by someone else.

Discussion: The empirical analyses of the 5S-HM demonstrate that it is a robust measure for use in clinical and research settings. Thematic analyses proposed explanations for why self-harm behaviors are initiated and how they are reinforced over time. Sexual self-harm in particular requires further careful study.

Keywords: 5S-HM; harmful self-neglect; indirect self-harm; self-harm; sexual self-harm; unified theoretical model of self-harm.

Grants and funding

This work was supported by the Swedish National Self-Harm Project (https://nationellasjalvskadeprojektet.se/english/), Swedish Research Council, Mats Paulsson Foundation, and Söderström-Königska Foundation.