A Descriptive Study of Nonsuicidal Self-Injury Characteristics Among Veterans Meeting Diagnostic Criteria for Nonsuicidal Self-Injury Disorder

Abstract

Introduction:

Although nonsuicidal self-injury (NSSI) is more common among military veterans than adult civilians, little is known about NSSI disorder among veterans.

Method:

The present study explored NSSI characteristics among veterans meeting criteria for current NSSI disorder (N=41) and the relationship between NSSI methods and functions.

Results:

Participants reported a pattern of past-year NSSI engagement characterized by frequent and severe NSSI, a short interval between NSSI urges and behavior, and moderate or greater subjective distress and interference in family relationships, social relationships, and work/school. Psychiatric comorbidities were common and nearly half of participants reported a suicide attempt history. Participants used an average of four NSSI methods, the most common being wall-punching (85.4%), and endorsed an average of eight distinct functions of NSSI, predominantly for intrapersonal reasons. Regardless of the specific method used, the most common reason that veterans with NSSI disorder engaged in NSSI was to help themselves cope with negative emotions.

Conclusions:

Findings suggest certain NSSI characteristics are similar among (e.g., NSSI functions) and differ between (e.g., NSSI methods) veterans and civilians with current NSSI disorder. Further research is necessary in order to replicate and expand upon these findings with nationally representative samples and better understand NSSI functions among veterans.

Nonsuicidal self-injury (NSSI) is the deliberate destruction of one’s own body tissue without the intent to die (American Psychiatric Association [APA], 2013; Klonsky, 2007). While NSSI behaviors tend to be less lethal in nature than suicidal behaviors, they are associated with significant distress and impairment, with potential for serious injury (Doshi et al., 2005; Selby et al., 2012). In recognition of this, NSSI disorder was first proposed as a distinct diagnostic entity requiring further research in the most recent version of the DSM-5 (APA, 2013). The official proposed criteria for NSSI disorder released in 2013 contain detailed descriptions of NSSI frequency, reasons for NSSI engagement, and emotions and cognitions experienced prior to NSSI. Prior civilian research supports NSSI disorder as a unique clinical diagnosis distinct from suicidal behavior and associated with significant functional impairment and distress (Klonsky et al., 2014; Andover et al., 2005; Gratz et al., 2015; Odelius et al., 2014; Washburn et al., 2015; Zetterqvist, 2015).

Veterans represent a particularly important subgroup for NSSI disorder research due to their disproportionately high rates of NSSI compared to the general population. Recent meta-analytic work suggests lifetime NSSI prevalence among veterans is nearly three times the rate observed among their adult civilian counterparts (Gromatsky et al., 2023; Swannell et al., 2014). Additionally, NSSI has been consistently associated with suicidal thoughts and behavior among military veterans (Kimbrel et al., 2016; Villatte et al., 2015) highlighting its importance as a focus of research and clinical interventions. However, relative to civilians, there is presently a paucity of data for veterans who meet full diagnostic criteria for NSSI disorder and further investigation is necessary to better understand NSSI characteristics among this subgroup, especially reasons underlying NSSI.

Furthermore, veterans differ from civilians on a number of important attributes (e.g., demographics, combat exposure) that may contribute to differences in certain NSSI characteristics between veterans and civilians with NSSI disorder. Women and adolescent females are often overrepresented among civilian studies of NSSI (Taylor et al., 2018), whereas U.S. military veteran samples are comprised entirely of adults, most of whom are men. Among veteran women, there may be unique elements of the veteran experience that differentiate their presentation of NSSI disorder from their civilian counterparts. For example, Beagley and colleagues (2023) recently demonstrated that the traditional masculine norm of violence was strongly associated with lifetime NSSI disorder in female, sexual minority veterans, specifically. At the same time, while differences in NSSI risk may be influenced by both gender and veteran status, prior work suggests that lifetime NSSI prevalence in veterans is not moderated by gender (Gromatsky et al., 2023) and certain forms of NSSI (i.e., wall/object punching) may be more closely tied to the unique experiences of veterans themselves (Kimbrel et al., 2018).

Elucidating the characteristics of NSSI engagement among veterans with NSSI disorder is therefore necessary to identify unique risk factors, tailor interventions, and improve clinical care for this high-risk population. Certain characteristics typical of veteran NSSI that appear to be comparable with civilian NSSI include lifetime frequency of behavior, average severity, physical pain experienced, and distress (Barnette et al., 2023; Bryan & Bryan, 2014; Lear et al., 2021); whereas recent evidence suggests veterans may have a later age of NSSI onset (Barnette et al., 2023). However, the degree to which these findings about NSSI characteristics may translate to veterans with NSSI disorder is still unknown.

To date, existing literature about NSSI functions (i.e., motivations behind NSSI) reported by veterans with NSSI history is sparse (Barnette et al., 2023) and there has not been a systematic study among veterans who meet criteria for NSSI disorder. Prior civilian research suggests NSSI functions can be organized into a two-factor structure of intrapersonal and interpersonal reasons (Klonsky et al., 2015; Taylor et al., 2018; Turner et al., 2012). Intrapersonal functions of NSSI relate to exerting influence over one’s internal environment and broadly encompass affect regulation. Interpersonal functions of NSSI relate to exerting influence over one’s external (e.g., social) environment and may include communicating distress, seeking support, or influencing others. Meta-analytic findings suggest intrapersonal functions are reported by 66-81% of civilians who engage in NSSI, versus only 33-56% who endorse interpersonal functions (Taylor et al., 2018). This is supported by preliminary qualitative findings about veterans’ reasons for initiating NSSI, which suggest intrapersonal functions are most common among both civilians and veterans but endorsed at a significantly lower rate among veterans (Barnette et al., 2023). In line with this, NSSI is often used by veterans as a strategy to cope with distress (particularly anger) and may be reinforced by a sense of relief (Cassiello-Robbins et al., 2021; Dillon et al., 2021; Kimbrel et al., 2018; Sacks et al., 2008). Veterans may also engage in NSSI for interpersonal reasons, including communicating and getting help from others (Barnette et al., 2023; Cassiello-Robbins et al., 2021). While civilian studies suggest individuals meeting criteria for current NSSI disorder also exhibit greater variety in NSSI functions (Andover et al., 2014; Gratz et al., 2015; Zetterqvist, 2015), it is necessary to examine these characteristics among veterans with NSSI disorder.

Veterans may adopt one or several specific methods of NSSI to address these varied functions. Evidence is mixed regarding the number of NSSI methods veterans typically use (Bryan & Bryan, 2014; Holliday et al., 2018; Lear et al., 2021). However, a recent study suggests that veterans with lifetime NSSI history report utilizing a greater number of NSSI methods than civilians (Barnette et al., 2023). Individuals in the general population with current NSSI disorder also exhibit greater variety in NSSI methods than those with NSSI history alone (Gratz et al., 2015; Zetterqvist, 2015). Further investigation into the average number of methods used by veterans with NSSI disorder is therefore needed.

Additionally, there appears to be variability in the specific types of NSSI methods typically used by veterans and their relevance among veterans with NSSI disorder is unclear. While cutting oneself appears quite common, especially among women veterans (Bryan & Bryan, 2014; Lear et al., 2021), several behaviors less common among civilians are also often endorsed. These include burning oneself, hitting oneself, head banging, and provoking physical fights (Barnette et al., 2023; Bryan & Bryan, 2014; Halverson et al., 2022; Lear et al., 2021). Furthermore, wall/object punching appears to be a particularly prevalent form of NSSI among both veteran men and women that is associated with engagement in other forms of NSSI, relief from negative affect, and suicidal ideation (Halverson et al., 2022; Kimbrel et al., 2018). Thus, the diversity in NSSI methods endorsed by veterans to date highlights the need for further investigation into contributing factors like specific reasons behind NSSI, and to understand their relevance among veterans with more severe presentations of NSSI (i.e., NSSI disorder).

Veterans with NSSI disorder may also experience more clinical comorbidities associated with greater distress and impairment than veterans with other psychiatric disorders. For example, lifetime NSSI disorder has been associated with borderline personality disorder (Patel et al., 2021) and found to be a stronger predictor of lifetime suicide attempts than several other mental health diagnoses (Cunningham et al., 2021). These findings highlight the substantial mental health burden associated with lifetime NSSI disorder among veterans, underscoring the importance of examining their relevance among those with current NSSI disorder. Doing so may provide critical insight into the profile of NSSI characteristics (including distress and impairment) linked to significant psychiatric comorbidities, addressing a key gap in understanding the clinical needs of veterans with current NSSI disorder.

Study objectives

The present study is the first to comprehensively examine NSSI characteristics, including NSSI functions, among veterans with current NSSI disorder. We first characterized NSSI frequency, severity, methods, and functions among veterans meeting full criteria for current NSSI disorder. We also explored NSSI frequency, preoccupation with NSSI before engagement, subjective distress, and functional impairment. Based on the civilian literature, we hypothesized that participants would also endorse high NSSI severity, use of several NSSI methods, a high number of NSSI functions, and have comorbid psychiatric diagnoses. We then sought to explore the association between specific NSSI methods and functions. Current treatment recommendations for NSSI incorporate a functional analysis approach largely based on NSSI functions identified among female civilians. Thus, improved understanding of NSSI characteristics among veterans with NSSI disorder, particularly NSSI functions, will support development and adaptation of appropriate interventions for veterans.

Method

Participants

Participants included 41 veterans enrolled in a one-year longitudinal study examining the impact of NSSI on functional outcomes. Participants were recruited through clinician referrals, local flyers, and mailings to veterans receiving care in VA Posttraumatic Stress Disorder (PTSD) clinics or those who consented to being contacted for future research. Veterans aged 18 years or older were eligible. Exclusion criteria included lifetime diagnosis of bipolar disorder, psychotic spectrum disorder, or imminent risk for suicide/homicide warranting immediate intervention. Potential participants were screened by phone to ensure they met basic eligibility criteria, and final eligibility was determined at the baseline assessment. Although the larger study recruited veterans with a history of engaging in NSSI at any point in their lifetime and psychiatric controls, present analyses are limited to baseline data of participants who met full criteria for current NSSI disorder (n=41).

Diagnostic & Clinical Interviews

The Clinician Administered Nonsuicidal Self-injury Disorder Index (CANDI; Gratz et al., 2015) assessed past-year history of NSSI disorder as well as descriptive and contextual factors, including frequency, severity, impairment, age of onset, antecedents, and functions of NSSI. Participants indicated functions of their NSSI during the past year by assigning a percentage value (0-100%) to 17 possible functions of NSSI (e.g., to relieve tension, punish yourself, gain a sense of control). The CANDI has exhibited good interrater reliability (k=.83) and adequate internal consistency (α=.71) in prior research (Gratz et al., 2015). Because NSSI disorder is currently a condition for further study in the DSM-5, each CANDI interview was reviewed and discussed in a diagnostic review group (consisting of six members) led by a licensed clinical psychologist until diagnostic consensus was reached.

The Structured Clinical Interview for DSM-5 (SCID-5; First et al., 2015) assessed current and lifetime history of psychiatric disorders except for borderline personality disorder (BPD), which was assessed with the Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV; Zanarini et al., 1996). The SCID-5 has demonstrated excellent internal consistency and test-retest reliability in prior research (Shankman et al., 2018); the DIPD-IV has demonstrated adequate test-retest and inter-rater reliability (Zanarini et al., 2000). In the present study, the SCID-5 demonstrated excellent interrater reliability (Fleiss’ k = .92). Similar to the CANDI, DIPD interviews were discussed and reviewed in diagnostic review groups led by a licensed clinical psychologist until diagnostic consensus was reached.

The Columbia Suicide Severity Rating Scale (C-SSRS; Posner et al., 2011) assessed lifetime history of suicide attempt. This semi-structured interview evaluated the presence/absence of passive (e.g., “wish to be dead”) or active suicidal ideation (e.g., “I’ve thought about killing myself”), as well as one or more suicide attempt made in a participant’s lifetime. It has demonstrated good convergent and divergent validity with other suicidal ideation and behavior scales (Posner et al., 2011). In the present study, the suicide attempt(s) variable was limited to instances of actual suicide attempts, excluding interrupted and aborted attempts. This was done to enhance the interpretability and applicability of our findings, as many studies and clinical assessments do not consistently evaluate interrupted or aborted attempts. Staff received extensive training on C-SSRS administration prior to the study, and any issues in ratings were resolved by the supervising investigator who is a licensed psychologist.

Self-Report Measures

The Deliberate Self-Harm Inventory (DSHI; Gratz, 2001) was administered to participants via self-report prior to the CANDI to identify relevant NSSI behaviors for further discussion and specific methods of NSSI engaged in. The DSHI assessed past-year frequency of engagement in 16 NSSI behaviors: cutting, burning (with cigarette, lighter, or match), carving (words, pictures, designs, other marks), severe scratching, biting, rubbing skin against rough surfaces, using chemicals on the skin, sticking self with sharp objects, rubbing glass against skin, breaking bones, banging one’s head, punching oneself, and wound picking. An additional item was added to assess wall/object punching, given the high prevalence of this form of NSSI among veterans (Kimbrel et al., 2018). A final question asked participants to identify and describe other forms of NSSI they had engaged in that were not identified in the preceding questions. The DSHI has demonstrated good test-retest reliability and convergent and divergent validity in prior research (Gratz, 2001). The internal consistency of the DSHI in the present study was good (α = .84).

Procedure

Data was collected from a larger VA-funded study approved by the Durham VA Institutional Review Board and Research and Development Committee. Additional information about study procedures can be found elsewhere (Dillon et al., 2021). After informed consent was obtained, diagnostic interviews were conducted by trained masters-level clinicians who completed extensive training. This included studying interview manuals, watching instructional training videos, rating video-recorded interviews, and ongoing participation in weekly supervision. All NSSI disorder cases were carefully reviewed and discussed in weekly diagnostic review group meetings led by a licensed clinical psychologist.

Data Analysis

Descriptive statistics were calculated for demographic and NSSI characteristics. Due to low endorsement, burning with cigarette, match, and lighter were collapsed into a “Burning” category; carving words and pictures were collapsed into a “Carving” category. Based on civilian literature (Klonsky et al., 2015; Taylor et al., 2018; Turner et al., 2012), NSSI functions were organized into a two-factor structure of intrapersonal and interpersonal reasons. Endorsement of the following twelve functions of NSSI from the CANDI were included in the intrapersonal functions factor based on prior literature: decrease/relieve negative emotion; tension relief; produce positive emotion; stop feeling numb/empty; feel something or alive; to relieve boredom; punish oneself; distract oneself from negative thoughts/feelings; stop negative thoughts; gain a sense of control; show/prove to oneself how bad things are; avoid attempting suicide. Endorsement of the following five functions of NSSI were included in the interpersonal functions factor based on prior literature: communicate to others; get help from others; influence others; get out of some responsibility; and end an argument.

Results

Demographic Characteristics

Table 1 details participant characteristics for the sample (N=41). Although participants were not recruited on the basis of gender, as is common for veteran research and reflective of the makeup of the broader veteran population, many (70.7%) were men. Mean age of participants was 46.4 years (SD = 12.7, range: 23-77) and approximately half identified as African American (56.1%). As hypothesized, nearly all of the sample met criteria for one or more mental health diagnosis. The average number of current mental health diagnoses was 5.6 (SD = 1.9), the most common being PTSD (90.2%) and major depressive disorder (MDD; 82.9%). Additionally, a majority (78.0%) of participants met lifetime criteria for BPD and nearly half (46.3%) endorsed a lifetime history of suicide attempt(s).

Table 1.

Demographic Characteristics

	Veterans with Current NSSI Disorder (n=41)
	M	SD
Age	46.4	12.7
Years of Education	14.5	2.3
# Current Diagnoses	5.6	1.9
	N	%
Male Sex	29	70.7
African American	22	53.7
White	18	43.9
Service-Connected	34	85
Marital Status
Married	19	46.3
Divorced/Separated	14	34.1
Widowed	0	0
Single, Never Married	5	12.2
Living together	3	7.3
Psychiatric Diagnoses
PTSD (Lifetime)	39	95.1
PTSD (Current)	37	90.2
MDD (Lifetime)	39	95.1
MDD (Current)	34	82.9
BPD (Lifetime)	32	78
AUD (Lifetime)	28	70
AUD (Current)	9	23.1
SUD (Lifetime)	11	26.8
SUD (Current)	6	14.6
Lifetime Suicide Attempt	19	46.3

Note: NSSI=Nonsuicidal self-injury. PTSD=Post-traumatic Stress Disorder. MDD=Major Depressive Disorder. GAD=Generalized Anxiety Disorder. AUD=Alcohol Use Disorder. SUD=Substance Use Disorder. BPD=Borderline Personality Disorder.

Diagnoses derived from Structured Clinical Interview for DSM-5 (SCID-5) and Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV). Lifetime history of suicide attempt(s) was identified using the Columbia-Suicide Severity Rating Scale (C-SSRS).

NSSI Characteristics

Tables 2 and 3 describe NSSI characteristics of the sample. Average age of NSSI onset was 23.7 years (SD = 9.11) and ranged from 5 to 46 years. As expected, participants evidenced physically damaging, frequent, and moderate to severe NSSI resulting in clinically significant distress and impairment. Specifically, all participants reported tissue damage from NSSI, with an average of 54.2 days (SD = 100.9; i.e., 15% of the days in the past year) and 55.4 instances (SD = 115.2) of NSSI engagement in the past year. The interval between NSSI urge and engagement was generally short, with nearly all participants (87.8%) reporting acting on urges within an hour (43.9% within a minute). Additionally, nearly half (46.4%) of participants reported severe or extreme subjective distress associated with NSSI, about a third reported moderate but manageable distress (36.6%), and only 19.5% reported no NSSI-related distress. A majority of participants reported moderate or greater NSSI-related interference in family relationships (73.2%), social relationships (68.3%), and work/school (71.8%). The severity of NSSI was judged to be moderate among 61% of participants and either severe (34.1%) or extreme (4.9%) in more than a third of participants.

Table 2.

Nonsuicidal Self-Injury (NSSI) Characteristics

	Veterans with Current NSSI Disorder (n=41)
	N	%
Tissue Damage	41	100
Urge/Act Interval
<1 Minute	18	43.9
1 minute-1 hour	18	43.9
1-4 hours	4	9.8
4 hours-1 day	1	2.4
Subjective Distress
None	8	19.5
Mild/minimal	0	0
Moderate, manageable	15	36.6
Severe, considerable	14	34.1
Extreme, incapacitated	4	9.8
Interference: Family
None	9	22
Mild	2	4.9
Moderate	16	39
Severe	12	29.3
Extreme	2	4.9
Interference: Social
None	11	26.8
Mild	2	4.9
Moderate	16	39
Severe	7	17.1
Extreme	5	12.2
Interference: Work/School
None	11	28.2
Mild	0	0
Moderate	16	41
Severe	8	20.5
Extreme	4	10.3
NSSI Severity
None	0	0
Mild	1	2.4
Moderate	25	61
Severe	13	31.7
Extreme	2	4.9

Note: NSSI=Nonsuicidal self-injury. Data pertaining to past-year NSSI characteristics was collected by the Clinician-Administered Nonsuicidal Self-injury Disorder Index (CANDI).

Table 3.

Nonsuicidal Self-Injury (NSSI) Methods and Functions

	Veterans with Current NSSI Disorder (n=41)
Overall	M [Md]	SD
# of NSSI Days	54.2 [19]	100.9
# of NSSI Instances	55.4 [19]	115.2
# of NSSI Methods	4.4	2.3
# of NSSI Functions	8.0	3.3
# of Intrapersonal Functions	7.2	2.8
# of Interpersonal Functions	0.9	1.2
Specific Methods	N	%
Wall/object punching	35	85.4
Punching self	20	48.8
Severe Scratching	19	46.3
Banging head	19	46.3
Biting self	16	39.0
Sticking self	15	36.6
Wound healing	13	31.7
Cutting	12	29.3
Other	9	23.7
Burning	7	17.1
Carving	4	10.0
Rubbing sandpaper	2	4.9
Rubbing glass	1	2.4
Specific Function	N	%
Any Intrapersonal Function	41	100
Decrease negative emotion	40	97.6
Relieve tension	35	85.4
Stop negative thoughts	31	75.6
Distract self	29	70.7
Stop numbness	26	63.4
Gain control	24	58.5
Punish self	23	56.1
Feel something/alive	23	56.1
Avoid suicide attempt	21	51.2
Produce positive emotion	15	36.6
Prove to self how bad things are	14	34.1
Relieve boredom	13	31.7
Any Interpersonal Function	15	36.6
Communicate to others	13	31.7
End argument	12	29.3
Get help	5	12.2
Get out of responsibilities	3	7.3
Influence others	2	4.9

Notes: NSSI=Nonsuicidal self-injury.

Data pertaining to NSSI characteristics was collected by the Clinician-Administered Nonsuicidal Self-injury Disorder Index (CANDI) and Deliberate Self-Harm Inventory (DSHI). Data assessing how often each function for NSSI was experienced in the past year (ranging from 0 to 100%) were transformed as individual binary (yes/no) variables. Burning with cigarette and with match/lighter were collapsed into a “Burning” category. Carving words and carving pictures were collapsed into a “Carving” category. Dripping acid on skin, using caustic agents, and breaking bones are not listed as they were assessed but not endorsed by any participant.

Endorsement of the following 12 functions for NSSI were included in the Intrapersonal Functions factor: To decrease/relieve negative emotion; tension relief; produce positive emotion; to stop feeling numb/empty; to feel something or alive; to relieve boredom; to punish oneself; to distract oneself from negative thoughts/feelings; to stop negative thoughts; to gain a sense of control; to show/prove to oneself how bad things are; to avoid attempting suicide.

Endorsement of the following 5 functions for NSSI were included in the Interpersonal Functions factor: to communicate to others; to get help from others; to influence others; to get out of some responsibility; to end an argument.

Some categories do not contain complete information for some participants due to missing data.

NSSI Methods

Veterans with current NSSI disorder endorsed using an average of 4.4 (SD = 2.27) different NSSI methods in the past year. The most common NSSI methods used were wall/object punching (85.4%), self-punching (48.8%), severe scratching (46.3%), head banging (46.3%), biting (39%), sticking with sharp objects (36.6%), preventing wounds from healing (31.7%), cutting (29.3%), and burning (17.1%).

NSSI Functions

Veterans with NSSI disorder endorsed an average of eight different functions contributing to NSSI in the past year (Table 3). All participants endorsed engaging in NSSI for intrapersonal reasons, and only a third (36.6%) for interpersonal reasons. The most common reasons for engaging in NSSI included decreasing negative emotion (97.6%), relieving tension (85.4%), stopping negative thoughts (75.6%), distracting themselves (70.7%), stopping numbness (63.4%), gaining control (58.5%), and punishing themselves (56.1%).

Relationship between NSSI Methods and Functions

To visualize associations between NSSI functions and methods, a heat map was generated whereby the endorsement rate of each NSSI function (columns) was reported for each NSSI method (rows). Darker shades of red represent a higher rate of function endorsement and darker shades of blue represent a lower rate of function endorsement (Figure 1). A pattern emerged such that intrapersonal functions related to decreasing negative affect (i.e., decreasing negative emotions, stopping negative thoughts, distracting oneself, relieving tension, stopping numbness) were highly endorsed across all NSSI methods (overall endorsement rate ≥80%); interpersonal functions (i.e., ending arguments, communicating to others, getting help, getting out of responsibility, influencing others) were rarely endorsed across all NSSI methods (overall endorsement rate ≤20%). These findings are highly consistent with prior civilian research (e.g., Klonsky, 2011) and suggest that, regardless of method used, the most common reason that veterans with NSSI disorder engage in NSSI is to alleviate negative emotions.

Figure 1. Heatmap Demonstrating Associations Between Specific NSSI Methods and Functions among Veterans with NSSI Disorder (n=41).

NSSI=Nonsuicidal self-injury. Data pertaining to NSSI characteristics was collected by the Clinician-Administered Nonsuicidal Self-injury Disorder Index (CANDI) and Deliberate Self-Harm Inventory (DSHI).

Discussion

Several studies have examined methods and functions of NSSI among the general population (Fox et al., 2015; Swannell et al., 2014; Taylor et al. 2018) and provided support for the clinical utility of the NSSI disorder diagnosis among civilians in the decade following its proposal (Andover et al., 2014; Gratz et al., 2015; Zetterqvist, 2015). However, research on NSSI disorder among veterans, for whom lifetime NSSI prevalence is nearly triple the general population (Gromatsky et a., 2023; Swannell et al., 2014), remains comparatively limited. Veterans with NSSI disorder in the present study reported a pattern of past-year NSSI engagement characterized by frequent and severe NSSI and a short interval between NSSI urge and behavior. While most participants endorsed moderate or greater NSSI-related subjective distress and interference in all domains assessed, 19.5% denied NSSI-related distress. Based on expectations but not explicitly specified in the diagnostic criteria, participants engaged in several forms of NSSI, endorsed several specific functions of NSSI, and met criteria for multiple mental health diagnoses. Moreover, nearly half (46.3%) of veterans with current NSSI disorder reported a lifetime suicide attempt history. This is consistent with existing literature demonstrating the association between lifetime NSSI and suicide attempt history among military veterans (Cunningham et al., 2019, 2021; Gromatsky et al., 2021; Kimbrel et al., 2016).

These findings are consistent with prior research supporting the clinical utility of the NSSI disorder diagnosis among civilians (Selby et al., 2015). They also offer preliminary support for the conceptualization and relevance of NSSI disorder for veterans as a unique clinical condition marked by recurrent NSSI using various means and accompanying intrusive thoughts which are associated with distress and impairment (APA, 2013). Future research adapting existing NSSI treatment approaches for veterans is clearly necessary.

In line with expectations and prior literature examining lifetime NSSI among veterans (Barnette et al., 2023; Bryan & Bryan, 2014; Holliday et al., 2018), veterans with NSSI disorder endorsed using an average of four different NSSI methods. The most common forms of NSSI included wall/object punching, self-punching, head banging, scratching, biting, and sticking with sharp objects. Particularly noteworthy is that some of the most common methods of NSSI (i.e., cutting, burning) observed in the general population (Klonsky, 2011), college students (Whitlock et al., 2011), and among veterans with any NSSI history (e.g., Bryan & Bryan, 2014; Holliday et al., 2018; Lear et al., 2021) were among the least common methods of NSSI utilized by veterans with NSSI disorder in the present study. In contrast, wall/object punching (85.4% endorsement) was by far the most common method endorsed, consistent with certain prior research on NSSI among veterans (Barnette et al., 2023; Kimbrel et al., 2018).

Supporting hypotheses and civilian literature, veterans with NSSI disorder endorsed several (M=8.0) different reasons (i.e., functions) for engaging in NSSI. Notably, 100% of participants endorsed engaging in NSSI for intrapersonal reasons, most often to decrease negative emotion. This finding is highly consistent with civilian research on the functions of NSSI (Klonsky, 2011; Taylor et al., 2018; Zetterqvist, 2015). In contrast, but also consistent with studies among civilians (Klonsky, 2011) and preliminary studies among veterans with lifetime NSSI (Barnette et al., 2023), interpersonal functions were endorsed by a much smaller number of veterans with NSSI disorder (36.6%). Present findings support existing literature pertaining to the antecedents and consequences of NSSI among veterans, which suggests NSSI may be used as a means of coping with distress (particularly anger) and reinforced by a sense of relief (Cassiello-Robbins et al., 2021; Dillon et al., 2021; Kimbrel et al., 2018; Sacks et al., 2008).

The infrequency with which interpersonal functions of NSSI were endorsed in the present sample calls into question the clinical utility of including the proposed B2 criterion (i.e., engaging in NSSI to resolve interpersonal difficulties) as part of NSSI disorder criterion. Despite engaging in NSSI on 15% of the days in the past year, nearly two-thirds (63.4%) of veterans with NSSI disorder reported they had not engaged in NSSI for interpersonal reasons at all during the past year. Among the 15 (36.6%) veterans who reported engaging in NSSI for an interpersonal reason during the past year, interpersonal reasons were never the dominant function. Moreover, even among those veterans who did endorse interpersonal functions, intrapersonal functions were endorsed approximately three times more frequently than interpersonal functions (7.4 vs. 2.3). Further support for the predominance of intrapersonal functions among veterans with NSSI disorder comes from examination of the heat map plotting associations between NSSI methods and functions (Figure 1). This visualization clearly illustrates that, regardless of specific method used, the most common reason veterans with NSSI disorder engage in NSSI is to help cope with negative emotions.

Ongoing VA efforts to increase availability of treatments targeting self-directed violence and emotional dysregulation contributing to NSSI are therefore encouraged, including Dialectical Behavior Therapy (DBT; Landes et al., 2017). Although the traditional format of DBT involves individual psychotherapy, skills group, phone coaching, and a case consultation team, evidence suggests DBT skills training alone can be effectively delivered modularly based on the individualized needs of each patient (Valentine et al., 2015). Meta-analyses have demonstrated that men tend to be somewhat less effective at communicating than women and are less likely to engage in self-disclosure (Anderson & Leaper, 1998; Dindia & Allen, 1992; Hyde, 2005). Additionally, adherence to traditional masculine gender role norms is associated with decreased help-seeking behavior (Seidler et al., 2016) and difficulty expressing vulnerability and emotional distress to others (Pollack & Levant, 1995). Given this context, communication difficulties may play an important role in maintaining NSSI among certain male veterans with NSSI disorder who may benefit from NSSI treatment approaches which allow for individualized treatment module selection, such as the Treatment for Self-Injurious Behaviors (T-SIB) protocol (Andover et al., 2017). This brief protocol includes an optional interpersonal communication module for individuals most likely to benefit from such training. Unfortunately, to our knowledge, no study to date has used the T-SIB protocol to treat veterans with NSSI disorder, although this would appear to be a very promising avenue for future investigation.

Limitations and Future Directions

The present study had several strengths, most notably inclusion of veterans with current NSSI disorder based on a comprehensive, state-of-the-art clinical interview (Gratz et al., 2015). Nonetheless, present findings should be interpreted within the context of several limitations. First, this study was cross-sectional in nature. Consequently, additional research is needed to better understand how NSSI functions and methods might change over time among veterans with NSSI disorder, as NSSI functions may vary over short intervals and distinct functions can co-occur (Coppersmith et al., 2021). Additional research is also necessary to better understand precipitants to specific NSSI behaviors in veterans. It may be especially valuable to examine emotional precursors to various NSSI methods and reasons for engagement. Examination of short-term fluctuations in NSSI urges and behaviors (e.g., using ecological momentary assessment) have suggested anger both precedes and predicts NSSI among veterans (Dillon et al., 2021).

An additional limitation concerns sample ascertainment, as the present sample was primarily recruited from mental health clinics. Thus, high rates of comorbid mental health diagnoses (e.g., PTSD, MDD) potentially limit generalizability of findings to veterans with NSSI disorder identified outside of mental health clinic settings. A final noteworthy limitation concerns the over-representation of male veterans in the present study. While the high percentage of male veterans reflects the makeup of the broader veteran population, recent meta-analytic findings suggest a similar rate of NSSI between male and female military personnel (Gromatsky et al., 2023). Additional research which includes equal numbers of male and female veterans with NSSI disorder is therefore necessary to examine potential sex differences.