Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Oct 30.
Published in final edited form as: Psychiatry Res. 2016 Jul 13;244:179–184. doi: 10.1016/j.psychres.2016.07.025

Nonsuicidal Self-Injury Disorder: Does Criterion B Add Diagnostic Utility?

Amy M Brausch a,*, Jennifer J Muehlenkamp b, Jason J Washburn c,d
PMCID: PMC5026934  NIHMSID: NIHMS807215  PMID: 27479110

Abstract

The current paper evaluates Criterion B for Non-Suicidal Self-Injury Disorder, which states that non-suicidal self-injury (NSSI) must occur for at least one function. The majority of individuals who engage in NSSI report at least one function, so it is unclear if Criterion B provides diagnostic utility in individuals who already meet Criterion A (i.e., NSSI occurring on 5 or more days in the past 12 months). This paper compared individuals meeting Criterion A (threshold group) to those with 1-4 acts of NSSI in the past year (subthreshold group) in two different samples. The first sample included 217 undergraduate students, and the second sample included 1082 individuals from a behavioral health hospital, all with past-year NSSI. The majority of both samples reported at least one function of NSSI (99%). For the undergraduate sample, the number of and level of endorsement of functions were similar across threshold and subthreshold groups. For the behavioral health sample, the threshold group endorsed significantly more functions for NSSI and greater endorsement of affect regulation, self-punishment, and toughness compared to the subthreshold group. While some differences were found between NSSI groups, overall endorsement of functions for NSSI appears to be a universal characteristic regardless of NSSI frequency.

Keywords: deliberate self-harm, self-harm behavior, DSM-5, functions, criteria

1. Introduction

Nonsuicidal self-injury (NSSI), or the intentional injury of body tissue without suicidal intent, has recently been designated as a condition for further study within the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5; American Psychiatric Association [APA], 2013); establishing a clear need for additional research on the disorder, particularly in evaluating the proposed criteria. While the proposed criteria have received initial empirical support (e.g., Glenn and Klonsky, 2013), they showed poor inter-rater reliability in the DSM-5 field trials (Regier et al., 2013). Recent studies have raised concerns about the clinical utility of criterion A and B because few differences have been identified between individuals who meet diagnostic criteria for NSSI disorder and those with a history of NSSI but don't meet the disorder criteria in both community and clinical samples (e.g., Andover, 2014; Muehlenkamp and Brausch, in press; Washburn et al., 2015). These studies indicate a strong need to further investigate the validity and utility of the different DSM-5 criterion sets. The purpose of the current study was to examine whether differences in NSSI functions (criterion B) exist between individuals who report 5 or more acts of NSSI within the past 12 months (meeting criterion A; referred to as “threshold”) compared to those with fewer than 5 acts of NSSI in the past year (not meeting criterion A; referred to as “subthreshold”); thus, examining the utility of criterion B as a distinguishing factor in NSSI clinical severity.

Criterion B specifies that self-injurious behavior must be performed for at least one of the following reasons: 1) relieve negative thoughts or feelings, 2) resolve an interpersonal problem, or 3) cause a positive feeling or emotion. Studies that were published well before the DSM-5 (APA, 2013) criteria indicate that almost all individuals who engage in NSSI report at least one function of the behavior, which is most commonly some form of affect regulation (Klonsky, 2007). Initial investigations of the DSM-5 criterion B have generally revealed that identifying specific functions has limited clinical utility, with one study finding that 87.7% of a sample of patients with current NSSI reported at least one function of NSSI and almost 25% reported all three functions assessed (Washburn et al., 2015). Almost all participants (98%) in an adolescent inpatient sample who met Criterion A reported affect regulation as a function of their NSSI, followed by marking distress (89%), self-punishment (88%), and anti-dissociation (88%), again indicating that functions of NSSI may not be adding diagnostic utility to NSSI Disorder (Glenn and Klonsky, 2013). Additional evidence for the universality of NSSI functions come from studies that find high rates (85% - 100%) of function endorsement in a variety of samples (In-Albon et al., 2013; Turner et al., 2012; Zetterqvist et al., 2013). Other studies have reported comparable results showing that participants who meet criterion A of NSSI disorder are not meaningfully distinguished from those who do not meet criterion A when it comes to the functions endorsed (e.g., Barrocas et al., 2012; Glenn and Klonsky, 2013), or assessments of distress and impairment (Andover, 2014). Very few studies provide descriptive information regarding the mean number of functions endorsed, but one study reported an average of 4.76 functions endorsed in a high school sample (Lloyd-Richardson et al., 2007).

Thus, it appears that the number and types of functions specified in criterion B may not add incremental validity to differentiating individuals who do and do not meet a clinically relevant threshold for NSSI disorder. However, the existing studies on NSSI disorder criteria have predominantly focused on evaluating frequency and recency criteria only, or have evaluated the criterion set as a whole (e.g., Andover, 2014; Washburn et al., 2015; Zetterqvist et al., 2013), reporting on criterion B descriptively. Studies have not intentionally examined the clinical utility of criterion B, and the functions assessed have varied across studies, with some using existing measures of NSSI functions and others using novel items that were written to assess only the three functions outlined in the DSM criteria (Glenn and Klonsky, 2013; In-Albon et al., 2013; Selby et al., 2012; Washburn et al., 2015). The inconsistent methodologies and lack of a planned evaluation of criterion B's usefulness to diagnostically differentiating NSSI groups compromises the conclusions that can be drawn about the clinical utility of criterion B at this point. The current study aims to fill this gap and is the first known study to specifically evaluate the utility of Criterion B in both community and clinical samples of self-injurers, while using a comprehensive measure of NSSI functions.

Rationale and Hypotheses

A handful of studies examining the newly proposed DSM-5 (APA, 2013) diagnostic criteria for NSSI disorder have commented about the clinical and diagnostic utility of criterion B, with many noting that overwhelming majorities of their sample report at least one function of NSSI (Glenn and Klonsky, 2013; Washburn et al., 2015; Zetterqvist et al., 2013). Furthermore, endorsement of criterion B symptoms does not appear to associate with psychopathology or impairment above and beyond common characteristics of NSSI (Washburn et al., 2015). The current study examined whether Criterion B adds clinical and diagnostic utility by differentiating individuals who engage in NSSI at clinically significant levels (criterion A) compared to those who do not, within two unique samples. The first study included a large community sample of young adults with recent NSSI, and the second study included adolescent and adult inpatients admitted to a behavioral health hospital program specifically for NSSI. The research question driving this study was: Are individuals who meet NSSI disorder criterion A differentiated by the number and relevance of functions for NSSI compared to those with NSSI who do not meet criterion A of NSSI disorder?

Methods

2.1 Subjects, Study 1

A total of 2,950 subjects completed the study. Subjects were predominantly female (71.2%), heterosexual (84.4%), full-time students (94.7%), with a mean age of 20.41 years (SD = 3.03). Participants predominantly identified with Caucasian (87.4%) or Black/African-American race/ethnicity (3.8%), with small representations of Asian (2.3%), Hispanic/Latino (2%), and Hmong (1.6%). All demographics are reflective of the university from which the sample was drawn. Within the sample, 24.5% (n = 724) endorsed ever engaging in NSSI, with 30% (n=217) of those subjects having done so within the past 12 months. Using DSM-5 criterion A to categorize subjects, 3.5% (n=105) of the total sample reported 5 or more acts of NSSI in the past 12 months (14.5% of the sample reporting any NSSI history) and made up the “threshold” community sample; 3.9% (n=115) of the total sample reported 1-4 acts of NSSI in the past 12 months (15.9% of the sample reporting any NSSI history) and made up the “subthreshold” community sample.

2.2 Procedure

Subjects were recruited from students enrolled at two different universities; the first is a comprehensive liberal arts university located within the Midwestern United States that has an average campus enrollment of 10,000 students. A random sample of 5,000 student emails was identified by the University Registrar and an email invitation to participate in the current study was sent to those addresses. A brief description of the study was provided within the email invitation along with a link to the anonymous, online survey that was housed on Qualtrics, an academic online survey platform. IP tracking functions were disabled for this study to ensure anonymity. Informed consent was presented on the first screen and subjects only accessed the survey questions after indicating informed consent. After completing demographic items, subjects completed the questionnaires in a randomized order. Upon completion, subjects saw a debriefing screen that also housed a link to a unique survey where subjects could enter contact information to be included in a drawing for 1 of 50, $20 gift cards. The second university is a regional comprehensive university in the South-central region of the United States with an average enrollment of 20,000 students. Subjects were recruited from introductory and upper-level psychology courses through an on-line study board. Subjects completed the study questionnaires in small group settings on-campus and received course credit for their participation. Trained research assistants assessed subject's self-report responses and provided individual debriefing and referral as needed if subjects were determined to be at risk for suicidal behavior. The study received full approval from the Institutional Review Board at both universities.

2.3 Measures

2.3.1 Non-Suicidal Self-Injury

Features of NSSI were assessed with the Self-Injurious Thoughts and Behaviors Interview – short form (SITBI; Nock et al., 2007) and the Inventory of Statements About Self-Injury (ISAS; Klonsky and Glenn, 2009). Items from the SITBI and the ISAS assessing the type, frequency, and recency of NSSI behaviors engaged in were used as self-report questions in the current study (see Latimer et al., 2013). Data from one university utilized the SITBI for frequency and recency information and data from the second university utilized the ISAS. Raw frequency and recency data from both measures were coded as being “1-4 acts of NSSI” or “5 or more” and as “more than 12 months ago” or “within the past 12 months.” Subjects with NSSI history could then be coded as “meeting DSM-5 criterion A” or “not meeting criterion A” in terms of frequency and recency. The SITBI has been used in a variety of studies on NSSI and has demonstrated reliability and validity within both clinical and non-clinical samples of youth (e.g., Latimer et al., 2013; Nock et al., 2007).

The functions of NSSI were assessed using the Inventory of Statements About Self-Injury(Klonsky and Glenn, 2009) at both universities, which includes 39 items measuring 13 potential functions, or reasons why an individual engages in self-injury. Due to space limitations within the research protocol, only the four subscales representing the strongest factor loadings (Klonsky and Glenn, 2009), or theoretical salience, to the latent functions of interpersonal (i.e., interpersonal influence, peer bonding, sensation seeking, and toughness) and intrapersonal (i.e., affect regulation, self-punish, anti-suicide, marking distress) were used. These also correspond with the three DSM-5 criterion B functions listed in the symptom set. Participants rated the relevance of each functional item to their self-injury using a 3-point scale ranging from 0 “not relevant” to 2“very relevant” and each subscale can have a total score ranging from 0-6. Both the two latent-factor function structure and 13-subscale factor structure of the ISAS has been validated across a variety of samples, with the scale showing strong reliability and validity (Glenn and Klonsky, 2011; Klonsky and Glenn, 2009). Within the current sample, the internal consistencies across the eight subscales assessed were good, ranging from an alpha of .67 to .85.

2.4 Subjects, Study 2

Subjects in study 2 included 1,082 individuals who were consecutively admitted to an acute care program specifically designed to treat nonsuicidal self-injury; thus all participants in this sample had lifetime history of NSSI. About one-third of the sample (34%, n=366) were receiving inpatient treatment at the time of the study and the remaining subjects received some combination of intensive outpatient treatment and partial hospitalization. The sample was largely female (88.1%) and non-Hispanic Caucasian (78.5%). Other race/ethnic groups were represented: Hispanic (11.3%), Black (1.9%), Asian (0.6%), and American Indian (0.7%). The mean age was 17.4 years (SD = 6.67) and ranged from 12 to 57, as the hospital provides treatment to both adolescents and adults. . About half of the sample (49.8%, n=539) reported a mean of at least 5 NSSI acts in the past 12 months, and were considered to meet Criterion A for frequency and recency (the “threshold” treatment group). The other half (50.2%, n=543) reported 1-4 acts of NSSI in the past 12 months and were coded as the “subthreshold” treatment group.

2.5 Procedure, Study 2

Subjects were recruited through consecutive admissions at a behavioral health hospital in in an urban center in the Midwestern Region of the United States. Subjects completed the measures for the study at the time of admission to the hospital and/or treatment program. The self-report measures were completed as part of the routine clinical assessment at intake, and this data was combined with demographic data from medical records. According to the federally defined de-identification “Safe Harbor” standard for protected health information, all individual-level data were de-identified. The procedures for this study were reviewed and approved by the hospital and affiliated university systems’ Institutional Review Boards.

2.6 Measures, Study 2

2.6.1 Alexian Brothers Assessment of Self-Injury (ABASI; Washburn et al., 2015)

The ABASI was designed by the clinical staff at Alexian Brothers Behavioral Health Hospital to be included in routine clinical assessments to evaluate the presence of NSSI disorder as specified in DSM-5 (APA, 2013) within a population of patients who are already identified as engaging in NSSI. The measure asks respondents to note the number of days in the past year that they engaged in 21 types of potentially self-harmful behavior. For the purposes of the current study, to more carefully align with the traditional definition of NSSI used in empirical studies and the definition used in DSM-5, only cutting, burning/branding, hitting, banging, and carving were included as NSSI behaviors for Criterion A. The ABASI also includes items to assess Criterion B (three items asking about NSSI functions) and Criterion C (four items asking about preceding thoughts and emotions).

2.6.2 Inventory of Statements About Self-Injury (ISAS; Klonsky and Glenn, 2009)

Study 2 also used the ISAS to assess functions of NSSI. Similar to study 1, due to space limitations within the research protocol, only the four subscales representing the strongest factor loadings (Klonsky and Glenn, 2009) of interpersonal and intrapersonal functions were used. Additionally, each function subscale was assessed with only two items rather than the standard three items. Therefore, all ISAS function subscales from study 2 have total scores than range from 0-4, rather than 0-6. Within the sample from study 2, the internal consistencies across the eight subscales was varied. The anti-suicide and sensation seeking subscales demonstrated poor reliability (alpha = .19 and .13, respectively), and were removed from study analyses. Other subscales were adequate to good, with alpha values ranging from .57 to .76.

3. Results

3.1 Study 1

3.1.2 Differences in Functions Reported

Within the NSSI sample, complete data regarding NSSI functions was available for 157 participants. Overall, 99.4% of the NSSI sample reported at least 1 function. Using chi-square analysis, the threshold and subthreshold groups were equally likely to report zero NSSI functions, χ2(1) = 0.83, p = .36. In fact, only one participant in the entire NSSI group reported zero functions. To determine differences in number of overall functions reported, each function subscale was coded as 1 if any item on that scale was endorsed, or 0 if no item on that scale was endorsed. Functions with endorsement were then summed to create a total score for number of functions endorsed that ranged from 0-8. Across groups, the mean number of functions endorsed was 4.23 (SD = 1.58), with a median of 4 and mode of 5. A one-way ANOVA found no significant differences between the threshold (M=4.01, SD=1.59) and subthreshold (M=4.49, SD=1.58) NSSI groups on number of functions, F (1, 157) = 3.67, p = .06, η2=.023.

3.1.3 Differences in Relevancy Ratings for Functions

To test the different levels of endorsement of the eight NSSI functions between the threshold and subthreshold community groups, a MANOVA was used with NSSI group as the independent variable and the eight NSSI function subscale scores from the ISAS as the dependent variables. An adjusted p value of .006 was used in the MANOVA to reduce type-I error due to multiple dependent variables. The overall test was not significant, F (1, 192) = 1.90, p = .06, η2=.07. Only one of the eight NSSI function subscales showed significant group differences, with the subthreshold group reporting greater endorsement of the anti-suicide function than the threshold group (see Table 1 for means and standard deviations); endorsement of this function was relatively low overall. Affect regulation was the function rated as being most relevant to both groups (mean rating = 3.40, SD = 2.1), followed by self-punishment (mean = 2.67, SD = 2.30) and then marking distress (mean = 1.61, SD = 1.77). Peer bonding was rated as the least relevant (mean = 0.08, SD = 0.34), along with sensation seeking (mean = 0.55, SD = 0.96), and interpersonal influence (mean = 0.50, SD = 1.03).

Table 1.

Mean differences for function relevancy ratings for clinical and subclinical NSSI groups in Study 1 (community young adults)

NSSI Clinical Group (n=100) NSSI Subclinical Group (n=101)
Function Mean (SD) Mean (SD) F-value p-value η 2
Affect Regulation 3.26 (1.99) 3.51 (2.21) 0.73 .39 .004
Self-Punishment 2.64 (2.14) 2.70 (2.37) 0.39 .84 <.001
Anti-Suicide 0.92 (1.50) 1.69 (2.03) 9.41 .002 .045
Sensation Seeking 0.42 (0.81) 0.689 (1.09) 3.80 .053 .019
Peer Bonding 0.07 (0.36) 0.10 (0.33) 0.36 .55 .002
Interpersonal Influence 0.56 (1.22) 0.45 (0.79) 0.63 .43 .003
Toughness 1.14 (1.48) 1.15 (1.64) 0.007 .93 <.001
Marking Distress 1.59 (1.68) 1.62 (1.85) 0.02 .89 <.001
Number of Functions 4.01 (1.59) 4.49 (1.54) 3.67 .06 .023
Open in a new tab

Note. Functions are subscales from the Inventory of Statements about Self-Injury (ISAS). Each subscale scores ranges from 0-6. NSSI Clinical group refers to individuals who meet criterion A of NSSI Disorder; subclinical group reported 1-4 acts of NSSI in past 12 months.

3.2 Study 2

3.2.1 Differences in Functions Reported

The same sequence of analyses was run for the sample in study 2. Within the NSSI sample, complete data regarding NSSI functions were available for 1082 participants. Overall, 98.5% of the NSSI sample reported at least one function. Across groups, the mean number of functions endorsed was 5.16 (SD = 1.78), with a median of 5 and mode of 6. A one-way ANOVA found that the threshold group reported significantly more functions of NSSI (M = 5.42, SD = 1.56) than the subthreshold (M = 4.95, SD = 1.88), F (1, 1015) = 19.16, p < .001, η2=.021. Chi-square analysis showed that the subthreshold NSSI group was slightly more likely to report zero NSSI functions (2.8%) than the threshold group (0.2%), χ2 (1) = 4.25, p = .04.

3.2.3 Differences in Relevancy Ratings of Functions

An adjusted p value of .008 was used in the MANOVA to reduce type-I error due to multiple dependent variables. Results from the MANOVA indicated an overall significant effect, F (1, 1007) = 26.78, p < .001, η2=.13. Significant differences between the threshold and subthreshold groups were found on the subscales of affect regulation, self-punishment, peer bonding, and toughness. For all subscales, the threshold group meeting DSM-5 Criterion A evidenced higher relevancy scores than subthreshold group, except for peer bonding in which case the subthreshold group reported higher relevancy scores1 (see Table 2).

Table 2.

Mean differences for function relevancy ratings for clinical and subclinical NSSI groups in Study 2 (adolescent and adult inpatient/outpatient)

NSSI Clinical Group (n=539) NSSI Subclinical Group (n=543)
Function Mean (SD) Mean (SD) F-value p-value η 2
Affect Regulation 3.49 (0.82) 2.99 (1.56) 83.52 <.001 .072
Self-Punishment 3.14 (1.1) 2.43 (1.35) 103.90 <.001 .090
Peer Bonding 0.24 (0.63) 0.34 (0.76) 8.27 .004 .008
Interpersonal Influence 0.92 (1.05) 1.02 (1.16) 2.58 .11 .002
Toughness 1.07 (1.22) 0.82 (1.11) 11.93 .001 .011
Marking Distress 1.94 (1.36) 1.79 (1.33) 4.91 .041 .004
Number of Functions 5.42 (1.56) 4.91 (1.94) 22.85 <.001 .021
Open in a new tab

Note. Functions are subscales from the Inventory of Statements about Self-Injury (ISAS). Each subscale scores ranges from 0-4. NSSI Clinical group refers to individuals who meet criterion A of NSSI Disorder; subclinical group reported 1-4 acts of NSSI in past 12 months.

4. Discussion

The main goal of the current study was to examine the utility of criterion B symptoms for NSSI Disorder in differentiating threshold from subthreshold groups of self-injurers. Previous studies indicated that identifying specific functions of NSSI behavior was not useful in distinguishing between more and less clinically severe NSSI behavior (e.g., Andover, 2014; Washburn et al., 2015). Results from the current study provide some additional evidence for the limited clinical utility of the criterion B symptom set in diagnosing NSSI Disorder. As in previous studies, we found that almost all participants across both samples reported at least one interpersonal or intrapersonal function for engaging in NSSI (99.4% and 98.5%, respectively) regardless of the frequency of their NSSI; indicating this may be a universal feature of the behavior that does not reflect the clinical severity of the behavior. It appears to be rare that an individual who engages in clinically relevant frequencies of NSSI will not endorse at least one function for the behavior (current criterion B). One possible way to improve upon the diagnostic criteria might be to specify the functional nature of NSSI within the criterion A symptoms rather than keep the functional nature of the behavior as a unique symptom criteria.

There was some discrepancy, however, in the total number of functions endorsed for the NSSI groups across samples. While the mean number of unique functions reported did not differentiate threshold from subthreshold NSSI groups within the community sample (4.01 vs. 4.49, respectively), it did for the treatment sample; although group differences were small (5.42 vs. 4.95). Examining the means across studies also shows that the treatment sample reported a greater number of NSSI functions than the community sample, again supporting the idea that clinically significant levels of NSSI appear to be associated with a greater number of functions for the behavior. These findings suggest that NSSI behavior occurring at a clinically significant level is more multifaceted in its functions than NSSI behavior that occurs at a subclinical level. Clinicians may want to attend to the number of unique functions reported for NSSI in order gauge potential severity of the behavior.

Such a conclusion is consistent with prior research documenting that as the frequency and severity of NSSI increases, the total number of functions endorsed for why one engages in the behavior also tends to increase (Saraff and Pepper, 2014). Furthermore, some studies have found that increasing numbers of functions endorsed for NSSI was significantly related to increased suicide risk (Paul et al., 2015; Whitlock et al., 2008). Reporting a greater number of functions for NSSI may be a sign that the behavior is becoming a generalized coping strategy, used across distressing situations, and may act as a marker for increasingly frequent and/or clinically severe NSSI. It may be worthwhile to consider revisions to criterion B such that instead of stating at least one function for the NSSI is present (which seems to be endorsed by almost all who engage in any act of NSSI), that a minimum number of unique functions are present. Granted, this new specification would require a clear delineation of a wider variety of functions for NSSI than is currently identified in the criterion B symptoms. Based on the current data, it would appear that expanding the three vague DSM-5 functions to include specific items such as self-punishment, sensation seeking, avoiding suicide, showing one's toughness, peer bonding, influencing the behavior/ emotions of others, and concretizing one's distress may help clinicians with the precision of their diagnostic assessment for this criterion. However, specifying a broader number of functions will be more cumbersome and require clinicians to use a standardized assessment tool that can aide in detecting and determining a standardized number of unique functions; but doing so could possibly add clinical utility to the criterion. Additional research will be needed to determine if a raw count of the functions endorsed for NSSI produces improvements in the clinical utility of this symptom, or if simply specifying the fact that the NSSI is used to achieve an intrapersonal or interpersonal goal (see Klonsky et al., 2015) as part of criterion A is enough to improve diagnostic reliability.

Another revision to consider that may increase the clinical utility for the criterion B symptoms would be to eliminate a focus on the number of functions endorsed, and instead focus predominantly on the relevance of particular functions for the NSSI. While all the functions of NSSI assessed were endorsed at comparable levels of relevance in the community sample, group differences in relevance were observed across a variety of functions for the treatment sample. The largest differences were for affect regulation and self-punishment functions, with the threshold group reporting significantly higher relevance of these functions than the subthreshold treatment group. Endorsement of interpersonal functions did not meaningfully differentiate groups across either sample and therefore, may have less clinical relevance for diagnostic purposes. Furthermore, across both study samples, intrapersonal functions were rated as being more relevant to motivating NSSI acts than were interpersonal functions. These findings suggest that engaging in NSSI for intrapersonal reasons may drive the clinical severity of the behavior. Consistent with that idea, prior research has indicated that those who report using NSSI for predominantly intrapersonal reasons show higher risk for suicide and co-occurring distress or impairment (Klonsky et al., 2015; Klonsky and Glenn, 2009), and that intrapersonally motivated NSSI is more likely to be repeated (Muehlenkamp et al., 2013). The congruence of findings from the current data and existing literature indicates that NSSI is largely motivated by needs to regulate emotions, self-punish, and generate feeling, and that these functions may better identify individuals who engage in NSSI at clinically salient rates than interpersonal functions. However, the group differences in the relevance of the intrapersonal NSSI functions between the threshold and subthreshold groups in the treatment sample had small effect sizes, and may be a product of a large sample size. Thus, identifying specific functions for NSSI disorder within criterion B may not meaningfully improve upon the clinical utility of this criterion. Additional research is needed to thoroughly examine whether endorsement of specific functions for NSSI improves upon the diagnostic validity for NSSI disorder.

The current study expands the empirical investigation of the DSM-5 criteria for NSSI disorder, which is an essential step to ensuring accuracy of any newly proposed disorder. However, there are some limitations that need to be considered when interpreting the current data. One notable limitation is that all the data were self-report and relied on retrospective recall of NSSI functions, although all participants included in the study analyses reported recent (in the past 12-months) acts of NSSI. Even though we utilized a well-validated and common assessment of the functions for NSSI, the items are broader in scope than what is specified by the criterion B symptom set. Therefore, the lack of meaningful differences may be even more salient to questioning the relevance of criterion B because our assessment of functions was more inclusive than the existing DSM-5 criteria. A strength of the study was the use of two samples, community and treatment based, which is unique to the existing literature examining diagnostic validity of NSSI disorder; but the populations were rather homogeneous within each sample. This limits the generalizability of the findings to individuals who represent non-majority racial/ ethnic and gender populations.

Overall, there seems to be few clinically significant distinguishing features regarding NSSI functions that would help to clearly delineate those meeting frequency and recency criteria for NSSI disorder, which compromises the clinical utility of the diagnostic symptoms. The current data adds to a handful of existing studies that call into question the diagnostic validity of criterion B for NSSI disorder (e.g., Andover, 2014; Washburn et al., 2015). Data appears to be inconsistent across samples and settings, with some finding evidence that criterion B symptoms are diagnostically meaningful (Glenn and Klonsky, 2013). However, many studies find that the criterion B symptoms do not add incremental validity above and beyond the other DSM-5 criteria. The current study also offers data suggesting that while the criterion B symptoms accurately capture the primary functions of NSSI, they do not offer substantial utility in differentiating threshold from subthreshold groups of self-injurers. Of note is the overwhelming majority of individuals in both samples that report at least one function of NSSI. A handful of significant differences were noted between groups in the current data, but they are small in magnitude and appear to lack clinical significance. Thus, we also call into question the clinical utility of criterion B. We recommend that additional research be conducted to empirically determine the best specification of the functions for NSSI that contribute to diagnostic validity, or reconsider the need for criterion B as currently written within the diagnostic profile for the disorder.

Highlights.

  • Criterion B for Non-suicidal Self-injury Disorder was evaluated.

  • Functions of NSSI were examined in a clinical and non-clinical sample.

  • Almost all individuals (99%) with past-year NSSI reported functions.

  • There were few clinically distinguishing features of NSSI functions.

Acknowledgments

This work was supported by the National Institutes of Health [P20GM103436].

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Due to the fact that the clinical sample included both adolescents (76%) and adults (24%), we also tested for differences by age and the interaction between age and DSM criterion group. Across DSM criterion groups, only the peer bonding function was significantly different between under 18 and over 18 groups, with under 18 reporting slightly higher scores. There was no interaction between age and DSM criteria group, and current results do not seem influenced by age.

References

  1. American Psychiatric Association . Diagnostic and statistical manual of mental disorders. 5th edition American Psychiatric Publishing; Washington, D.C.: 2013. [Google Scholar]
  2. Andover MS. Non-suicidal self-injury disorder in a community sample of adults. Psychiatry Res. 2014;219:305–310. doi: 10.1016/j.psychres.2014.06.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barrocas AL, Hankin BL, Young JF, Abela JRZ. Rates of nonsuicidal self-injury in youth: Age, sex, and behavioral methods in a community sample. Pediatrics. 2012;130:39–45. doi: 10.1542/peds.2011-2094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Glenn CR, Klonsky ED. One-year test-retest reliability of the Inventory of Statements about Self-Injury (ISAS). Assessment. 2011;18(3):375–378. doi: 10.1177/1073191111411669. [DOI] [PubMed] [Google Scholar]
  5. Glenn CR, Klonsky ED. Nonsuicidal Self-Injury Disorder: An empirical investigation in adolescent psychiatric patients. J. Clin. Child. Adolesc. Psychol. 2013;42:496–507. doi: 10.1080/15374416.2013.794699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. In-Albon T, Ruf C, Schmid M. Proposed diagnostic criteria for the DSM-5 of nonsuicidal self-injury in female adolescents: Diagnostic and clinical correlates. Psychiatry J. 2013;2013:159208. doi: 10.1155/2013/159208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Klonsky ED. The functions of deliberate self-injury: A review of the evidence. Clin. Psychol. Rev. 2007;27:226–239. doi: 10.1016/j.cpr.2006.08.002. [DOI] [PubMed] [Google Scholar]
  8. Klonsky ED, Glenn CR. Assessing the functions of non-suicidal self-injury: Psychometric properties of the Inventory of Statements for Self-Injury (ISAS). J. Psychopathol. Behav. Assess. 2009;31:215–219. doi: 10.1007/s10862-008-9107-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Klonsky ED, Glenn CR, Styer DM, Olino TM, Washburn JJ. The functions of nonsuicidal self-injury: Converging evidence for a two-factor structure. Child Adolesc. Psychiatry Ment. Health. 2015;9(44):1–9. doi: 10.1186/s13034-015-0073-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Latimer S, Meade T, Tennant A. Measuring engagement in deliberate self-harm behaviours: Psychometric evaluation of six scales. BMC Psychiatry. 2013;13(4):1–11. doi: 10.1186/1471-244X-13-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Lloyd-Richardson EE, Perrine N, Dierker L, Kelley ML. Characteristics and functions of non-suicidal self-injury in a community sample of adolescents. Psychol. Med. 2007;37:1183–1192. doi: 10.1017/S003329170700027X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Muehlenkamp JJ, Brausch AM. Reconsidering Criterion A for the diagnosis of Non-suicidal Self-Injury Disorder. J. Psychopathol. Behav. Assess. DOI 10.1007/s10862-016-9543-0. 2016 [Google Scholar]
  13. Muehlenkamp JJ, Brausch AM, Quigley K, Whitlock J. Interpersonal features and functions of nonsuicidal self-injury. Suicide Life. Threat. Behav. 2013;43:67–80. doi: 10.1111/j.1943-278X.2012.00128.x. [DOI] [PubMed] [Google Scholar]
  14. Nock MK, Holmberg EB, Photos VI, Michel BD. Self-Injurious Thoughts and Behaviors Interview: Development, reliability, and validity in an adolescent sample. Psychol. Assess. 2007;19:309–317. doi: 10.1037/1040-3590.19.3.309. [DOI] [PubMed] [Google Scholar]
  15. Paul E, Tsypes A, Eidlitz L, Ernhout C, Whitlock J. Frequency and functions of non-suicidal self-injury: Associations with suicidal thoughts and behaviors. Psychiatry Res. 2015;225:276–282. doi: 10.1016/j.psychres.2014.12.026. [DOI] [PubMed] [Google Scholar]
  16. Regier DA, Narrow WE, Clarke DE, Kraemer HC, Kuramoto J, Kuhl EA, Kupfer DJ. DSM-5 Field trials in the United States and Canada: Test-retest reliability of selected categorical diagnoses. Am. J. Psychiatry. 2013;170:59–70. doi: 10.1176/appi.ajp.2012.12070999. [DOI] [PubMed] [Google Scholar]
  17. Saraff PD, Pepper CM. Functions, lifetime frequency, and variety of methods of non-suicidal self-injury among college students. Psychiatry Res. 2014;219:298–304. doi: 10.1016/j.psychres.2014.05.044. [DOI] [PubMed] [Google Scholar]
  18. Selby EA, Bender TW, Gordon KH, Nock MK, Joiner TE., Jr Non-suicidal self-injury (NSSI) disorder: A preliminary study. Personal. Disord. 2011;3:167. doi: 10.1037/a0024405. [DOI] [PubMed] [Google Scholar]
  19. Turner BJ, Chapman AL, Layden BK. Intrapersonal and Interpersonal functions of non suicidal self-injury: Associations with emotional and social functioning. Suicide Life. Threat. Behav. 2012;42:36–55. doi: 10.1111/j.1943-278X.2011.00069.x. [DOI] [PubMed] [Google Scholar]
  20. Washburn JJ, Potthoff LM, Juzwin KR, Styer DM. Assessing DSM-5 nonsuicidal self-injury disorder in a clinical sample. Psychol. Assess. 2015;27:31–41. doi: 10.1037/pas0000021. [DOI] [PubMed] [Google Scholar]
  21. Whitlock J, Muehlenkamp JJ, Eckenrode J. Variation in nonsuicidal self-injury: Identification and features of latent classes in a college population of emerging adults. J. Clin. Child. Adolesc. Psychol. 2008;37:725–735. doi: 10.1080/15374410802359734. [DOI] [PubMed] [Google Scholar]
  22. Zetterqvist M, Lundh L-G, Dahlstrom O, Goran Svedin C. Prevalence and function of nonsuicidal self-injury (NSSI) in a community sample of adolescents, using suggested DSM-5 criteria for a potential NSSI Disorder. J. Abnorm. Child Psychol. 2013;41:759–773. doi: 10.1007/s10802-013-9712-5. [DOI] [PubMed] [Google Scholar]

RESOURCES