Personality and Individual Differences 46 (2009) 25–29

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Social context during non-suicidal self-injury indicates suicide risk

Catherine R. Glenn, E. David Klonsky *
Department of Psychology, Stony Brook University, Stony Brook, NY 11794-2500, United States

a r t i c l e i n f o a b s t r a c t

Article history: The link between non-suicidal self-injury (NSSI) and suicide is complex. Previous research indicates that
Received 2 April 2008 self-injurers endorsing automatic/intrapersonal functions (as opposed to social/interpersonal functions)
Received in revised form 20 August 2008 for NSSI are more likely to have considered and attempted suicide. Subsequent research suggests that
Accepted 27 August 2008
those endorsing automatic/intrapersonal functions are more likely to self-injure exclusively while alone.
Available online 15 October 2008
Based on these findings, we hypothesized that the social context during NSSI (i.e., the extent to which one
self-injures alone versus around others) represents an easily measurable and theoretically meaningful
Keywords:
marker for suicide risk among those who self-injure. Participants were 205 young adults who had per-
Self-injury
Deliberate self-harm
formed one or more NSSI behaviors and completed several clinical measures. In general, self-injurers
Suicide scored higher on measures of suicidality and suicide risk factors (i.e., depression, anxiety, borderline per-
Social factors sonality disorder symptomatology) than a non-injuring control sample (n = 596). In addition, self-injurers
who engage in NSSI alone were more likely to report a history of suicide ideation, plans, and attempts
compared to other self-injurers. Endorsement of automatic/intrapersonal functions only partially
explained the relationship between the social context during NSSI and suicidality. Consistent with the
study hypothesis, social context during NSSI appears to be a marker for suicide risk in individuals who
engage in NSSI.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction et al., 2003). The elevated rates of NSSI in adolescents and young
adults, in addition to the behavior’s associations with negative
Non-suicidal self-injury (NSSI) is the deliberate, self-inflicted mental health outcomes, have stimulated research on risk and pro-
destruction of body tissue without suicidal intent for purposes tective factors, the behavior’s functions, and methods for interven-
not socially sanctioned. As many as 14 different types of NSSI have tion (Klonsky, 2007; Klonsky & Glenn, 2008; Klonsky & Glenn, in
been identified but the most common forms include skin-cutting, press; Muehlenkamp, 2006; Nock & Prinstein, 2004, 2005).
burning, and scratching (Ross & Heath, 2002; Whitlock, Eckenrode, While research has documented consistent associations be-
& Silverman, 2006). Rates of NSSI are estimated at 4% in the general tween NSSI and certain clinical correlates (e.g., depression, anxiety,
adult population and 20% in adult patient populations (Briere & Gil, and BPD), a key issue to be resolved is the relationship between
1998; Klonsky, Oltmanns, & Turkheimer, 2003). However, rates of NSSI and suicidal behavior. By definition, NSSI differs from suicidal
NSSI appear to be disproportionately high in adolescents and behavior in terms of motivation (individuals who engage in NSSI
young adults (Ross & Heath, 2002; Whitlock et al., 2006): Approx- want to continue life while those who engage in suicide attempts
imately 8% of children ages 12–14 (Hilt, Nock, Lloyd-Richardson, want to end life), and research has also uncovered key differences
& Prinstein, 2008), 14–15% of adolescents (Laye-Gindhu & Schon- in medical severity (NSSI less often requires medical attention and
ert-Reichl, 2005; Ross & Heath, 2002), and 14–17% of college stu- is more superficial in its tissue damage than attempted suicide)
dents (Favazza, DeRosear, & Conterio, 1989; Whitlock et al., (Brown, Comtois, & Linehan, 2002; Favazza & Conterio, 1989;
2006) report having self-injured. In adolescent inpatient samples, Muehlenkamp & Gutierrez, 2004). At the same time, NSSI is a doc-
rates of NSSI appear to be 80% or higher (Nock & Prinstein, umented risk factor for suicidal behavior; elevated rates of suicidal
2004). NSSI has become a growing public health concern due to ideation and behavior are consistently reported among self-injur-
its strong association with a number of serious clinical variables, ing populations (Nock, Joiner, Gordon, Lloyd-Richardson, &
including depression, anxiety, borderline personality disorder, Prinstein, 2006; Whitlock et al., 2006).
and suicidality (Andover, Pepper, Ryabchenko, Orrico, & Gibb, One hypothesis regarding the relationship between NSSI and
2005; Hawton, Rodham, Evans, & Weatherall, 2002; Klonsky suicide is that self-injurers habituate to hurting themselves over
time, and thereby become more capable of carrying out suicide at-
* Corresponding author. Tel.: +1 631 632 7801; fax: +1 631 632 7876.
tempts (Joiner, 2002). Consistent with this theory, self-injurers
E-mail address: [email protected] (E.D. Klonsky). with a longer history of NSSI and who engage in more NSSI

0191-8869/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.paid.2008.08.020
26 C.R. Glenn, E.D. Klonsky / Personality and Individual Differences 46 (2009) 25–29

methods are particularly likely to have attempted suicide (Nock the ISAS to be a reliable and valid measure of NSSI frequency and
et al., 2006). However, these NSSI variables (i.e., history and meth- functions in a large sample of young adults (Klonsky & Olino,
od) accounted for only a small portion of the variance in suicidal 2008). The first section of the ISAS assesses the lifetime frequency
behavior. It would therefore be useful to identify other character- of 12 different NSSI behaviors performed ‘‘intentionally (i.e., on
istics of NSSI that put self-injurers at increased risk for suicide. purpose) and without suicidal intent” (i.e., banging/hitting self, bit-
Another possibility is that risk for suicide might vary based on ing, burning, carving, cutting, wound picking, needle-sticking,
the function of NSSI. At least two superordinate functions of NSSI pinching, hair pulling, rubbing skin against rough surfaces, severe
are documented in the literature: (1) automatic/intrapersonal scratching, and swallowing chemicals). In addition, the question-
functions (e.g., affect regulation, self-punishment) and (2) social/ naire assesses descriptive features of NSSI including the age of on-
interpersonal functions (e.g., interpersonal influence, peer-bond- set, experience of physical pain, time from the urge to self-injure
ing) (Klonsky & Olino, 2008; Nock & Prinstein, 2004). Initial re- until the NSSI act, and the tendency to self-injure alone (i.e., AL-
search suggests that different functions of NSSI may be NSSI). AL-NSSI was assessed with the question: ‘‘When you
associated with different levels of suicide risk. In one study, self- self-harm, are you alone?” Participants selected among three re-
injurers endorsing automatic/intrapersonal functions of NSSI were sponse options: ‘‘YES” = always alone during NSSI, ‘‘Sometimes” =
more likely to have made a recent suicide attempt (Nock & sometimes alone during NSSI, or ‘‘NO” = never alone during NSSI.
Prinstein, 2005). A subsequent study used a latent class analysis The second section of the ISAS measures the functions of non-
to identify clinically distinct subgroups of self-injurers, and found suicidal self-injury. The ISAS assesses 13 functions of NSSI that
a subgroup characterized by high suicidality, endorsement of auto- have been proposed in the empirical and theoretical mental health
matic/intrapersonal functions, and a tendency to self-injure alone literature (Klonsky, 2007). The 13 functions of NSSI fall into two
(Klonsky & Olino, 2008). Because reinforcement associated with superordinate factors: (1) intrapersonal functions (i.e., affect regu-
automatic/intrapersonal functions is self-focused (e.g., relieving lation, anti-dissociation, anti-suicide, marking distress, and self-
one’s negative emotions, directing anger at oneself), it follows that punishment) and (2) interpersonal functions (i.e., autonomy, inter-
individuals endorsing automatic/intrapersonal functions would personal boundaries, interpersonal influence, peer bonding, re-
most often self-injure alone. Indeed, Klonsky and Glenn (in press) venge, self-care, sensation seeking, and toughness).
found evidence that self-injurers endorsing automatic/intraper- Youth Risk Behaviors Survey (YRBS). Suicidality was assessed
sonal functions more often self-injured while alone. However, no using suicide questions from the Centers for Disease Control and
study has directly examined the relationship among automatic/ Prevention’s 1999 Youth Risk Behavior Survey (YRBS; Kann,
intrapersonal functions, tendency to self-injure alone, and 2001). Research has confirmed the test-retest reliability of the
suicidality. YRBS in an ethnically diverse sample of adolescents (Brener
Based on findings by Nock and Prinstein (2005) and Klonsky et al., 2002). A single suicidality score was created from these ques-
and Olino (2008), we believe the tendency to self-injure alone tions such that ‘0’ indicates no history of suicidality, ‘1’ indicates a
(henceforth AL-NSSI) may be an easily measurable and theoreti- history of suicidal ideation, ‘2’ indicates a history of suicidal plans,
cally meaningful marker for suicide risk among those who self-in- ‘3’ indicates a history of one suicide attempt, and ‘4’ indicates a his-
jure. Specifically, we hypothesize that those who self-injure tory of multiple suicide attempts.
exclusively while alone will report more suicidal thoughts and The McLean Screening Instrument for Borderline Personality Disor-
behaviors than those who occasionally or frequently self-injure der (MSI-BPD). Borderline personality disorder (BPD) was measured
with or around others. We further hypothesize that endorsement in this sample using the McLean Screening Instrument for Border-
of automatic/intrapersonal functions will partially explain the rela- line Personality Disorder (MSI-BPD), a 10-item self-report measure
tionship of AL-NSSI and suicidality given its association with both of BPD features (Zanarini et al., 2003). When compared to a vali-
these variables. dated structured interview, sensitivity and specificity of the MSI-
BPD were both above .90 in young adults (Zanarini et al., 2003).
Depression Anxiety Stress Scales (DASS-21). Depression and anxi-
2. Method ety were assessed using the Depression Anxiety Stress Scale (DASS-
21; Henry & Crawford, 2005). The DASS-21 is a shortened version
2.1. Participants and procedure of the original 42-item scale (Lovibond & Lovibond, 1995). Re-
search has confirmed the construct validity of the DASS-21 in
801 college students from lower-level psychology classes were non-clinical samples (Henry & Crawford, 2005).
screened for a history of 12 NSSI behaviors. Approximately twenty-
six percent of the current sample endorsed lifetime non-suicidal
self-injury, which is consistent with rates of NSSI found in previous 3. Results
research on college samples (between 17 and 35%; Gratz, 2001;
Whitlock et al., 2006). Participants were 205 students (57% female) 3.1. History of NSSI
who endorsed having engaged in at least one form of NSSI. The
mean age of the sample was 18.5 years (SD = 1.2) and the racial The most common NSSI behaviors in the sample were banging/
composition of the sample was 42% Caucasian, 39% Asian, 6% Afri- hitting self (61% of the sample), pulling hair (47%), pinching (42%),
can American, 6% Hispanic, and 7% other ethnicity. All participants and cutting (40%). 82% of participants used multiple methods of
gave informed consent and completed a battery of self-report mea- NSSI. The average age of onset of NSSI was 13 years old, and
sures for course credit. Only participants endorsing NSSI completed approximately 62% of the sample had self-injured within the past
portions of the ISAS (see below) assessing the functions and social year.
context of NSSI. Additional details about participants and the pro-
cedure are described in Klonsky and Olino (2008). 3.2. Clinical variables

2.2. Measure Means and standard deviations of all clinical variables (i.e.,
depression, anxiety, BPD symptomatology, and suicidality) for both
Inventory of Statements About Self-Injury (ISAS). The ISAS mea- the self-injuring sample (n = 205) and a non self-injuring control
sures the frequency and functions of NSSI. Recent research found sample (n = 596) are presented in Table 1. The DASS depression
 C.R. Glenn, E.D. Klonsky / Personality and Individual Differences 46 (2009) 25–29 27

Table 1 After finding evidence for a direct relationship between AL-

Means and standard deviations of clinical variables for self-injurers and non self- NSSI and suicidality, we first examined the role of automatic/
injuring controls
intrapersonal functions of NSSI in this association. As expected,
Variable Self-injurers Controls AL-NSSI was strongly related to the automatic/intrapersonal
M SD M SD functions of NSSI [t(195) = 4.26, p < 0.001]. An ANCOVA was con-
Depression 5.41 5.30 2.22 3.15
ducted to test the hypothesis that this variable would account
Anxiety 5.12 4.50 2.66 3.02 for the overall association between AL-NSSI and suicidality. After
Borderline PDa 4.34 2.41 2.40 2.21 controlling for automatic/intrapersonal functions, although the
Lifetime Suicidalityb 1.24 1.36 0.21 0.66 strength of the association decreased, the relationship between
Ideation 47.8% 10.1%
AL-NSSI and suicide remained robust [F(1, 197) = 8.55,
Plans 37.1% 6.4%
Single attempt 8.8% 1.3% p < 0.005] (see Table 2).
Multiple attempt 9.8% 0.9% Because automatic/intrapersonal functions only partially ex-
Last 12 months Suicidality 0.41 0.90 0.06 0.33 plained the relationship of AL-NSSI to suicide, we explored other
Ideation 18% 2.2%
variables that might account for the association. In particular, we
Plans 13.7% 2.3%
Single attempt 3.4% 0.2%
were interested in suicide risk factors that might correlate with
Multiple attempt 1.5% 0.0% AL-NSSI and thus explain the relationship of AL-NSSI to suicidality.
Several risk factors for suicide were measured in the present study,
Note. Depression, Anxiety, Borderline PD, and Suicidality were all significantly
higher in the self-injuring sample (ps < 0.0001).
and we first determined which of those risk factors were also re-
a
The self-harm item of the MSI-BPD was not included in BPD scores to avoid lated to AL-NSSI. AL-NSSI was related to anxiety at a trend level,
confounding the relationship between BPD and NSSI. but this relationship did not reach significance [t(196) = 1.91,
b
Suicidality measures severity of suicide on a scale from 0 to 4 with 0 = no p = 0.06]. However, AL-NSSI was significantly related to depression
suicidal behavior; 1 = history of suicidal ideation; 2 = history of suicidal plans; 3 =
[t(196) = 2.03, p < 0.05], BPD symptomatology [t(196) = 3.12,
history of a single suicide attempt; 4 = history of multiple suicide attempts.
p < .005], the frequency of cutting behavior [t(196) = 4.87,
p < 0.001], and the frequency of carving behavior [t(196) = 2.70,
p < 0.001]. Therefore, we entered these variables into a series of
and anxiety scales showed good internal consistency in both the
ANCOVAs. In each case a unique and robust relationship between
self-injuring (as = 0.88 and 0.82, respectively) and control samples
AL-NSSI and suicide remained after controlling for the variable. Full
(as = 0.82 and 0.74, respectively). In addition, MSI-BPD demon-
results are presented in Table 2.
strated good internal consistency in the self-injuring and control
To understand in greater detail the aspects of suicidality that re-
samples (as = 0.73 and 0.74, respectively). All clinical variables
late to AL-NSSI, we used chi-square analyses to separately examine
measured were significantly higher in the self-injuring sample (all
the relationship of AL-NSSI to suicidal ideation, plans, attempts,
ps < 0.0001). In particular, self-injurers were approximately eight
and multiple attempts. As indicated in Table 3, AL-NSSI was related
times more likely to seriously consider suicide and were almost
to histories of suicide ideation, plans, and attempts (ps60.01).
25 times more likely to attempt suicide than the non-injurers.
More specifically, higher rates of suicidal thoughts and behaviors
were observed in the ‘‘Always” alone group compared to the
3.3. The tendency to self-injure alone (AL-NSSI) and suicidality
‘‘Sometimes” or ‘‘Never” alone groups. AL-NSSI was related to a his-
tory of multiple suicide attempts at a trend level (p = 0.14) (see Ta-
Social context data was missing for seven participants; there-
ble 3). The non-significant p-value may be due to insufficient
fore, the analyses were conducted on the remaining 198 partici-
power in light of the small subsample of participants (20 of the
pants. In order to provide a sufficient sample size for the
198) who engaged in multiple suicide attempts.
analyses, the ‘‘Never” alone during NSSI (n = 23) and ‘‘Sometimes”
alone during NSSI (n = 67) participants were combined into one
group. Therefore, the following analyses refer to the comparison
4. Discussion
of (1) participants in the ‘‘Always” alone during NSSI groups
(n = 108) and (2) the combined ‘‘Sometimes” alone and ‘‘Never”
Findings support our hypothesis that the tendency to self-injure
alone during NSSI groups (n = 90). Consistent with our hypothesis,
exclusively while alone represents a theoretically meaningful and
analyses indicate a robust relationship between AL-NSSI and suici-
easily measurable indicator of suicide risk among those who en-
dality [t(196) = 4.29, p < 0.001]. Specifically, self-injuring alone as
gage in NSSI. Individuals who self-injure exclusively alone – that
opposed to around others was positively associated with suicidal-
is, never with others or never in the presence of others – endorse
ity. Because females were more likely to self-injure while alone [x2
substantially more suicidal thoughts and behaviors. Effects were
(1, 197) = 9.54, p < 0.005], gender was examined as a covariate in
strongest for suicidal ideation and plans, but were also significant
the relationship between AL-NSSI and suicide, but the association
for a history of suicide attempts.
remained significant (see Table 2). Ethnicity was not related to
Moreover, the relationship between AL-NSSI and suicidality
AL-NSSI [x2 (2, 198) = .19, p = 0.91] and thus not examined as a
remained significant even when accounting for previously docu-
covariate.
mented indicators of suicide risk. For instance, automatic/intra-
Table 2 personal functions of NSSI are a known correlate of both
Relationships of AL- NSSIa to suicidality controlling for several clinical covariates suicide risk and AL-NSSI (Nock & Prinstein, 2005). However, a
significant relationship between AL-NSSI and suicidality re-
Covariates df F p
mained even when controlling for endorsement of automatic/
Gender 1 13.16 0.000 intrapersonal functions. The relationship between AL-NSSI and
Automatic/ intrapersonal Functions 1 8.55 0.004
Depression 1 14.51 0.000
suicide risk also remained after controlling for gender, depres-
BPD symptomatology 1 11.16 0.001 sion, BPD, and frequency of NSSI. Thus, AL-NSSI appears to be
Frequency of cutting behavior 1 7.97 0.005 a particularly robust indicator of suicide risk among those who
Frequency of carving behavior 1 14.68 0.000 engage in NSSI.
a
‘‘AL-NSSI” refers to whether participants self-injure (1) ‘‘Always” while alone Findings from this study have useful treatment implications. In
versus (2) ‘‘Sometimes” or ‘‘Never” alone. treatment settings, patient safety is of chief importance and suicide
28 C.R. Glenn, E.D. Klonsky / Personality and Individual Differences 46 (2009) 25–29

Table 3
Relationship of social context during NSSI to suicide ideation, plans, and attempts

‘‘Sometimes” or ‘‘Never” alone during NSSI (n = 90) ‘‘Always” alone during NSSI (n = 108) Chi-square Test
a b
History of lifetime. . . n (%) n (%) x2 p
Suicidal ideation 27 (30.0) 69 (63.9) 22.57 0.00
Suicidal plans 20 (22.2) 55 (50.9) 17.19 0.00
Suicide attempts 10 (11.1) 27 (25) 6.23 0.01
Multiple attempts 6 (6.7) 14 (13.0) 2.14 0.14
a
Percentages out of 90 participants.
b
Percentages out of 108 participants.

risk is routinely assessed; however, assessing suicide risk is partic- cates an inability or reluctance to obtain social support, which in
ularly difficult among self-injuring patients. Although NSSI is turn confers risk for suicide.
distinct from suicide attempts based on motivation and medical
severity (Brown et al., 2002; Favazza & Conterio, 1989; Muehlenk-
Ethical Statement
amp & Gutierrez, 2004), NSSI appears similar to suicide attempts
and is a known risk factor for suicide (Whitlock et al., 2006). AL-
No part of this article has been published or submitted to any
NSSI can be easily assessed by therapists, and the current study
other journal. There are two authors on this manuscript; each
suggests that assessing the social context can provide important
one has studied the manuscript in the form submitted, agreed to
incremental information about suicide risk.
be cited as a coauthor, and has accepted the order of the author-
This study has several limitations. One limitation is the use of a
ship. The study described in this manuscript was approved by
college sample. Future studies should replicate findings in larger
the University’s Institutional Review Board.
and more diverse samples, including samples drawn from clinical
settings. In addition, features of NSSI and clinical variables were as-
Acknowledgement
sessed using retrospective self-report measures. Future research
might employ alternative methodologies, such as daily diary tech-
This work was supported in part by a grant from the American
niques that permit near real-time measurement of both the social
Foundation for Suicide Prevention and by the Office of the Vice
context during NSSI and suicidal thoughts and behaviors. In addi-
President for Research at Stony Brook University. The authors
tion, the present study used a single item rated on a three-point
would like to thank two anonymous reviewers for their feedback
Likert scale to assess AL-NSSI; this item required participants to
on earlier versions of this article.
aggregate across many NSSI episodes in making a single rating. Fu-
ture studies should more thoroughly assess AL-NSSI. For example,
References
it would be useful to have participants rate the social context for
different episodes of NSSI, as well as measure additional aspects Andover, M. S., Pepper, C. M., Ryabchenko, K. A., Orrico, E. G., & Gibb, B. E. (2005).
of the social context such as where the NSSI takes place, the type Self-mutilation and symptoms of depression, anxiety, and borderline
personality disorder. Suicide and Life-Threatening Behavior, 35, 581–591.
of persons present at the time of injury (i.e., family, friends, strang-
Brener, N. D., Kann, L., McManus, T., Kinchen, S. A., Sundberg, E. C., & Ross, J. G.
ers, or other self-injurers), and whether others are made aware (2002). Reliability of the 1999 youth risk behavior survey questionnaire. Journal
that the NSSI is taking place, even if not physically present. The of Adolescent Health, 31, 336–342.
development of a standardized measure of social context during Briere, J., & Gil, E. (1998). Self-mutilation in clinical and general population samples:
Prevalence, correlates, and functions. American Journal of Orthopsychiatry, 68,
NSSI may be useful for both research and treatment purposes. 609–620.
Finally, this study was not able to clarify the specific mecha- Brown, M., Comtois, K. A., & Linehan, M. M. (2002). Reasons for suicide attempts and
nisms by which AL-NSSI confers risk for suicide and whether non-suicidal self-injury in women with borderline personality disorder. Journal
of Abnormal Psychology, 111, 198–202.
AL-NSSI is a specific risk factor or an specific severity indicator. Be- Cavanagh, J. T. O., Carson, A. J., Sharpe, M., & Lawrie, S. M. (2003). Psychological
cause NSSI typically occurs as a response to intense, negative emo- autopsy studies of suicide: A systematic review. Psychological Medicine, 33,
tions (Klonsky, 2007, in press), one possibility is that AL-NSSI 395–405.
Durkheim, E. (1966). Suicide. (Spalding, J. A., Simpson, G., Trans.). New York: Free
reflects an individual’s style for coping with these strong, negative Press. (Original work published 1897).
emotions. Those who self-injure around others may be inclined to Favazza, A. R., & Conterio, K. (1989). Female habitual self-mutilators. Acta
seek support from others when emotionally distressed, and this Psychiatrical Scandinavica, 79, 283–289.
Favazza, A. R., DeRosear, L., & Conterio, K. (1989). Self-mutilation and eating
support may act as a buffer against suicide risk. In contrast, those
disorders. Suicide and Life-Threatening Behavior, 19, 352–361.
who self-injure alone may choose to avoid others when emotion- Gratz, K. L. (2001). Measurement of deliberate self-harm: Preliminary data on the
ally distressed, or may lack sources of social support on which to Deliberate self-harm inventory. Journal of Psychopathology and Behavioral
Assessment, 23, 253–263.
lean during difficult times. One of the earliest and most prominent
Hawton, K., Rodham, K., Evans, E., & Weatherall, R. (2002). Deliberate self-harm in
theories of suicide suggest that weak social bonds and a lack of adolescents: Self-report survey in schools in England. British Medical Journal,
integration into one’s social environment confer risk for suicide 325, 1207–1211.
(Durkheim, 1966). Recent empirical research supports this the- Henry, J. D., & Crawford, J. R. (2005). The short-form version of the Depression
Anxiety Stress Scale (DASS-21): Construct validity and normative data in a large
ory—in a comprehensive review of psychological autopsy studies non-clinical sample. British Journal of Clinical Psychology, 44, 227–239.
of suicide, Cavanagh, Carson, Sharpe, and Lawrie (2003) found that Hilt, L. M., Nock, M. K., Lloyd-Richardson, E. E., & Prinstein, M. J. (2008). Longitudinal
social isolation was approximately two times as common in cases study of non-suicidal self-injury among young adolescents: Rates, correlates,
and preliminary test of an interpersonal model. Journal of Early Adolescence, 28,
of completed suicides as compared to controls. It is possible that 455–469.
self-injurers may experience social anxiety (i.e., the desire to be- Joiner, T. E. (2002). The trajectory of suicidal behavior over time. Suicide and Life-
long stifled by fear of negative evaluation) or social anhedonia Threatening Behavior, 32, 33–41.
Kann, L. (2001). The youth risk behavior surveillance system: Measuring health-risk
(i.e., a lack of interest in belonging), either of which could preclude behaviors. American Journal of Health Behavior, 25, 272–277.
them from seeking social support. Future research may examine Klonsky, E. D. (2007). The functions of deliberate self-injury: A review of the
whether the tendency to self-injure exclusively while alone indi- evidence. Clinical Psychology Review, 27, 226–239.
 C.R. Glenn, E.D. Klonsky / Personality and Individual Differences 46 (2009) 25–29 29

Klonsky, E. D. (in press). The functions of self-injury in young adults who cut Muehlenkamp, J. J., & Gutierrez, P. M. (2004). An investigation of differences
themselves: Clarifying the evidence for affect regulation. Psychiatry Research. between self-injurious behavior and suicide attempts in a sample of
Klonsky, E. D., & Glenn, C. R. (2008). Resisting urges to self-injure. Behavioural and adolescents. Suicide and Life-Threatening Behavior, 34, 12–23.
Cognitive Psychotherapy, 36, 211–220. Nock, M. K., Joiner, T. E., Gordon, K. H., Lloyd-Richardson, E., & Prinstein, M. J. (2006).
Klonsky, E.D., Glenn, C.R. (in press). Assessing the functions of non-suicidal self- Non-suicidal self-injury among adolescents: Diagnostic correlates and relation
injury: Psychometric properties of the Inventory of Statements About Self- to suicide attempts. Psychiatry Research, 144, 65–72.
injury (ISAS). Journal of Psychopathology and Behavioral Assessment. Nock, M. K., & Prinstein, M. J. (2004). A functional approach to the assessment of
Klonsky, E. D., & Olino, T. M. (2008). Identifying clinically distinct subgroups of self- self-mutilative behavior. Journal of Consulting and Clinical Psychology, 72,
injurers among young adults: A latent class analysis. Journal of Consulting and 885–890.
Clinical Psychology, 76, 22–27. Nock, M. K., & Prinstein, M. J. (2005). Contextual features and behavioral functions
Klonsky, E. D., Oltmanns, T. F., & Turkheimer, E. (2003). Deliberate self-harm in a of self-mutilation among adolescents. Journal of Abnormal Psychology, 114,
non-clinical population: Prevalence and psychological correlates. American 140–146.
Journal of Psychiatry, 160, 1501–1508. Ross, S., & Heath, N. (2002). A study of the frequency of self-mutilation in a
Laye-Gindhu, A., & Schonert-Reichl, K. A. (2005). Non-suicidal self-harm among community sample of adolescents. Journal of Youth and Adolescence, 31,
community adolescents: Understanding the ‘whats’ and ‘whys’ of self-harm. 67–77.
Journal of Youth and Adolescence, 34, 447–457. Whitlock, J., Eckenrode, J., & Silverman, D. (2006). Self-injurious behaviors in a
Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the Depression Anxiety Stress college population. Pediatrics, 117, 1939–1948.
Scales (2nd ed.). Sydney, Australia: Psychology Foundation of Australia. Zanarini, M. C., Vujanovic, A. A., Parachini, E. A., Boulanger, J. L., Frankenburg, F. R., &
Muehlenkamp, J. J. (2006). Empirically supported treatments and general therapy Hennen, J. (2003). A screening measure for BPD: The McLean screening
guidelines for non-suicidal self-injury. Journal of Mental Health Counseling, 28, instrument for borderline personality disorder (MSI-BPD). Journal of
166–185. Personality Disorders, 17, 568–573.