Available online at www.sciencedirect.com
ScienceDirect Comprehensive Psychiatry 59 (2015) 1 – 7 www.elsevier.com/locate/comppsych
Nonsuicidal self-injury as a prospective predictor of suicide attempts in a clinical sample of military personnel Craig J. Bryan a, b,⁎, M. David Rudd a, c , Evelyn Wertenberger d , Stacey Young-McCaughon e , Alan Peterson e a
National Center for Veterans Studies b The University of Utah c The University of Memphis d Fort Carson e The University of Texas Health Science Center at San Antonio
Abstract Background: Nonsuicidal self-injury (NSSI) is a risk factor for suicide attempts, but it has received little attention in military populations, for whom suicide rates have doubled over the past decade. In the current study, the relationship of NSSI with future suicide attempts was prospectively examined in a sample of active duty Soldiers receiving outpatient psychiatric treatment for suicide ideation and/or a recent suicide attempt. Methods: Data were collected as part of a two-year prospective study of 152 active duty Soldiers (87% male, 71% Caucasian, mean age = 27.53) in outpatient mental health care who reported current suicide ideation and/or a suicide attempt during the month preceding intake. Suicide attempts and NSSI were assessed using the Suicide Attempt Self Injury Interview. Results: Forty percent of Soldiers with a history of nonsuicidal self-injury and 25% of Soldiers with a history of suicide attempt made a suicide attempt during the 2-year follow-up. Soldiers with a history of nonsuicidal self-injury were more than twice as likely to make a subsequent suicide attempt (hazard ratio [HR] = 2.25, P = .045). Soldiers with a history of suicide attempt were no more likely to make a subsequent suicide attempt than Soldiers without a previous suicide attempt (HR = .88, P = .787). Thirty percent of Soldiers with a history of suicide attempt had also engaged in nonsuicidal self-injury. Forty-two percent of Soldiers with histories of both nonsuicidal self-injury and suicide attempt made a subsequent suicide attempt and were more likely to make a suicide attempt during follow-up than Soldiers with a history of suicide attempt only. Number of NSSI episodes, but not number of suicide attempts, was significantly associated with increased risk for future suicide attempt. Results were unchanged when adjusting for baseline symptom severity. Limitations: Predominantly male, active duty Army sample. Conclusions: Among Soldiers in outpatient mental health care, a history of NSSI is a stronger predictor of future suicide attempts than a history of suicide attempts. Soldiers with a history of both NSSI and suicide attempt are at especially increased risk. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The suicide rate among members of the US military has more than doubled during the past decade [1]. Nonsuicidal self-injury (NSSI), which entails self-directed, deliberate behavior that results in injury or the potential for injury
⁎ Corresponding author at: National Center for Veterans Studies, 260 S. Central Campus Dr., Room 205, Salt Lake City, UT 84105. Tel.: +1 801 587 7978. E-mail address: [email protected] (C.J. Bryan). http://dx.doi.org/10.1016/j.comppsych.2014.07.009 0010-440X/© 2014 Elsevier Inc. All rights reserved.
to oneself without evidence of suicidal intent [2], is a well-established risk factor for suicide ideation and attempts [3–5], but to date it has received little empirical attention among military personnel. Prevalence rates of NSSI among military personnel are estimated to range from 4% to 14% [5,6], making them comparable to rates seen among US adults [5,7,8] and adolescents/young adults [8,9]. In psychiatric samples, NSSI is much more common, with 40%–70% of adolescent psychiatric cases reporting it in their history [3,4,8,10–12]. To date there are no published estimates of the relative prevalence of NSSI in clinical samples of military personnel.
2
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
Studies indicate that NSSI often co-occurs with, and is an especially robust risk factor for, suicide attempts. Up to 70% of adolescents in inpatient psychiatric treatment who have engaged in NSSI, for instance, have also made a suicide attempt [3,4]. Among military personnel, approximately 25% of those who have engaged in NSSI have also made a suicide attempt [6]. As a risk factor for suicide attempts, prospective studies among adolescents suggest that a history of NSSI is a stronger predictor of future suicide ideation [13] and suicide attempts [3] than previous suicide attempts. Similar research has yet to be conducted with military personnel, however. The primary aims of the current study were to prospectively examine the relationship of NSSI with suicide attempts in a clinical sample of military personnel. Two specific hypotheses were tested. First, we hypothesized that military personnel with a history of NSSI would be significantly more likely to make a suicide attempt during the 2-year follow-up period than military personnel with no history of NSSI. Second, we hypothesized that a history of NSSI would demonstrate a stronger association with future suicide attempts than a history of suicide attempts.
Army Medical Center, the University of Utah, and the University of Texas Health Science Center at San Antonio. 2.2. Assessments Suicide attempts and NSSI that occurred prior to intake and during the follow-up period were determined with the Suicide Attempt Self Injury Interview (SASII) [14], a validated clinician-administered interview that differentiates between suicide attempts and NSSI by assessing the intent, desired and expected outcome, medical severity, and other characteristics of intentional self-injury. The intensity of current (i.e., past week) suicide ideation was measured using the 19-item self-report Beck Scale for Suicide Ideation (BSSI-C) [15]. Depression symptom severity was assessed with the 21-item self-report Beck Depression Inventory [16], severity of hopelessness was assessed with the 20-item Beck Hopelessness Scale [17], and posttraumatic stress symptom severity was assessed with the 17-item PTSD Checklist-Military Version [18]. The presence of current psychiatric diagnosis was determined using the Structured Clinical Interviews for Axis I and Axis II DSM-IV disorders [19]. 2.3. Statistical analysis
2. Material and methods 2.1. Participants and procedures Participants were 176 active duty Soldiers participating in a randomized controlled trial testing a brief cognitive-behavior therapy for the reduction of suicide attempts at Fort Carson, Colorado. Participants were predominantly male (n = 153, 86.9%) and ranged in age from 19 to 44 years (M = 27.53, SD = 6.26). Self-identified race was 71.0% Caucasian, 21.6% Hispanic/Latino, 13.1% African–American, 4.5% Native American, 2.3% Pacific Islander, 1.7% Asian, and 8.0% other. Military grade distribution was 47.2% junior enlisted (E1–E4), 22.7% noncommissioned officer (E5–E6), 3.4% senior noncommissioned officer (E7–E9), and 0.6% warrant officer. Participants had served in the military for a mean (SD) of 5.70 (4.48) years, and the majority (81.2%) had deployed at least once. Participants were referred to research staff for determination of eligibility by their outpatient military mental health provider upon disclosure of suicide ideation, or following discharge from an inpatient psychiatric hospitalization for a suicide attempt. Inclusion criteria included current (i.e., past week) suicidal ideation with intent to die and/or a suicide attempt within the past month; active duty military status; age 18 years or older; ability to speak English; and ability to understand and complete informed consent procedures. The only exclusion criterion was the presence of a medical or psychiatric condition that would preclude informed consent or participation in outpatient treatment (e.g., active psychosis or mania). The current study’s procedures were reviewed and approved by the Institutional Review Boards of Madigan
All analyses were conducted using the SAS 9.3. For baseline group comparisons, Mann–Whitney U was used for continuous variables and logistic regression was used for binary variables. Of the 176 participants, 152 (86.3%) met eligibility criteria for the study and were therefore included in prospective analyses. Univariate and multivariate Cox proportional hazard regression model survival analysis of time to the first suicide attempt was used to determine the association of previous NSSI and suicide attempts with subsequent suicide attempt during the 2-year follow-up period. Time to suicide attempt was measured by calculating the total number of days from enrollment to first suicide attempt. For participants without a suicide attempt, the total number of days from enrollment to the last assessment was calculated. The Cox regression model was selected because it utilizes all available data from all participants regardless of dropout or length of follow-up. Estimates of the proportions of participants in each making at least one suicide attempt during the 2-year follow-up period were calculated using the Kaplan–Meier method, which similarly accounts for dropouts and limited follow-up.
3. Results Forty-six (30.3%) participants reported a history of NSSI, and 116 (76.3%) reported a history of suicide attempt. At baseline, there were no gender differences in history of NSSI (30.8% of men, 26.3% of women; OR = .80 [.27, 2.37], P = .689) or history of suicide attempts (78.2% of men, 63.2% of women; OR = .48 [.17, 1.32], P = .156), and no differences in total number of lifetime NSSI episodes (men:
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
M = 1.40, SD = 5.66; women: M = 5.84, SD = 22.61; Mann–Whitney U = .12, P = .902) or total number of lifetime suicide attempts (men: M = 1.34, SD = 1.00; women: M = 1.16, SD = 1.34; U = −1.22, P = .223). In terms of follow-up suicide attempts by NSSI history at intake, there were 12 participants with a history of NSSI (estimated proportion: 39.7%) who made a subsequent suicide attempt as compared to 14 participants without a history of NSSI (estimated proportion: 19.3%). Results of the multivariate Cox regression models predicting time to suicide attempt, controlling for race and symptom severity (i.e., depression, posttraumatic stress, hopelessness, and suicide ideation), indicated that participants with a history of NSSI were more than twice as likely to make a suicide attempt during the 2-year follow-up period than participants with no history of NSSI (Wald χ 2 = 4.02, P = .045; hazard ratio [HR] = 2.25, 95% CI: 1.02, 4.96). Survival curves by history of NSSI are displayed in Fig. 1. In terms of follow-up suicide attempts by suicide attempt history at intake, there were 20 participants with a history of suicide attempt (estimated proportion: 25.7%) who made a subsequent suicide attempt as compared to 6 participants without a history of suicide attempt (estimated proportion: 31.4%). Results of the multivariate Cox regression models predicting time to suicide attempt indicated that a history of suicide attempt was not associated with increased risk for suicide attempt during the 2-year follow-up period (Wald χ 2 = .07, p = .787; HR = .88, 95% CI: .34, 2.26). Survival curves by suicide attempt history are displayed in Fig. 2. Results did not change when history of NSSI and history of suicide attempt were entered into the Cox regression model simultaneously (see Table 2). In terms of co-occurrence of NSSI and suicide attempts, 41 of the 136 (30.1%) participants with a history of suicide
Fig. 1. Survival curves for time to first suicide attempt according to history of nonsuicidal self-injury (NSSI).
3
Fig. 2. Survival curves for time to first suicide attempt according to history of suicide attempt.
attempt had also engaged in NSSI. Participants were next categorized into 4 groups according to their histories of NSSI and suicide attempts: 12 (6.8%) participants reported a history of NSSI but no suicide attempts (i.e., “NSSI only”); 96 (54.0%) reported a history of suicide attempt but no NSSI (i.e., “SA only”); 41 (23.3%) reported a history of both NSSI and suicide attempts (i.e., “NSSI + SA”); and 28 (15.9%) reported neither NSSI nor suicide attempts (i.e., “No History”). Sociodemographic and clinical characteristics of each group are summarized in Table 1. The No History group had a significantly greater proportion of participants self-identifying as Caucasian and Native American than the NSSI + SA group. The No History group also had a significantly smaller proportion of participants self-identifying as Hispanic as compared to all other groups. The 4 groups did not differ from each other in terms of military-specific characteristics or baseline symptom severity. Participants in the NSSI group were significantly more likely to be diagnosed with social anxiety than all other groups. The NSSI and NSSI + SA groups also had a higher proportion of participants diagnosed with borderline personality disorder. There were no difference between the NSSI only and NSSI + SA groups in terms of the total number of lifetime NSSI episodes (U = .27, p = .984) and no difference between the SA only and NSSI + SA groups in terms of the total number of previous suicide attempts (U = 4.59, p = .614). During the 2-year follow-up period, there were 4 suicide attempts in the no history group (estimated proportion: 32.6%), 10 suicide attempts in the SA only group (estimated proportion: 15.7%), 2 suicide attempts in the NSSI only group (estimated proportion: 22.2%), and 10 suicide attempts in the NSSI + SA group (estimated proportion: 41.7%). Relative to the NSSI + SA group, participants in the SA group were approximately 60% less likely to have made a suicide attempt during follow-up (Wald χ 2 = 3.845, P =
4
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
Table 1 Demographic and clinical characteristics by group.
Male Age, M (SD), y Years of service, M (SD), y Race Caucasian African–American Asian Pacific Island Native American Other Hispanic Military rank E1–E4 E5–E6 E7–E9 Warrant officer No. of deployments 0 1 2+ DSM-IV diagnosis Major depressive Social anxiety Posttraumatic stress Alcohol dependence Substance dependence Bipolar I Bipolar II Borderline Symptom severity scales, M (SD) BDI-II PCL-M BHS BSSI-C Total NSSI episodes, median (IQR) Total suicide attempts, median (IQR)
No history (n = 28)
SA (n = 95)
NSSI (n = 12)
NSSI + SA (n = 41)
82.1 27.57 (6.71) 5.32 (4.62)
88.4 27.18 (6.12) 5.23 (4.19)
83.3 27.92 (6.02) 6.33 (4.33)
87.8 28.20 (6.50) 6.83 (5.00)
82.1 a 14.3 3.6 0.0 14.3 a 3.6 3.6 a,b,c
72.6 13.7 1.1 4.2 4.2 4.2 21.1 a
75.0 0.0 0.0 0.0 0.0 8.3 33.3 b
58.5 a 14.6 2.4 0.0 0.0 a 19.5 31.7 c
75.0 17.9 7.1 0.0
74.7 21.1 2.1 1.1
58.3 41.7 0.0 0.0
70.7 24.4 4.8 0.0
17.9 42.9 39.3
23.4 33.0 43.6
16.7 41.7 41.7
9.8 48.8 41.5
75.0 0.0 a 42.9 3.8 7.7 0.0 0.0 7.4 a,b
77.2 7.1 b 35.3 12.2 14.5 1.2 0.0 a 3.4 c,d
70.0 40.0 a,b,c 50.0 10.0 20.0 0.0 10.0 a 25.0 a,c
86.5 11.1 c 51.4 18.9 16.2 0.0 2.7 20.0 b,d
34.82 (11.56) 58.89 (15.91) 12.82 (4.26) 12.32 (9.45) N/A N/A
32.21 (14.09) 54.24 (17.07) 12.41 (6.56) 8.72 (8.34) N/A 1 (1)
31.64 (12.88) 58.64 (19.46) 13.55 (6.91) 6.73 (6.20) 1.5 (2) N/A
30.80 (14.67) 55.05 (18.12) 11.98 (6.37) 10.34 (8.71) 2 (2) 2 (2)
By row, values with the same superscript significantly differ from each other at p b .05; NSSI = nonsuicidal self-injury; SA = suicide attempt; BDI-II = Beck Depression Inventory, Second Edition; PCL-M = Posttraumatic Stress Disorder Checklist, Military Version; BHS = Beck Hopelessness Scale; BSSI-C = Beck Scale for Suicide Ideation, Current.
.050, HR = .40, 95% CI: .16, 1.00). A significant difference between the SA and NSSI + SA groups occurred at the 2-year follow-up (z =3.04, P = .002). There were no other Table 2 Results of Cox regression analyses predicting time to first suicide attempt. Predictor
Wald χ 2
p
HR
(95% C.I.)
Caucasian Hispanic Suicide attempt NSSI BDI-II PCL-M BHS BSSI-C
1.71 1.88 0.07 4.14 3.45 1.57 0.11 3.92
.191 .170 .787 .042 .063 .210 .739 .048
1.96 0.53 0.88 2.30 0.96 1.02 0.98 1.07
(0.72, 5.39) (0.21, 1.31) (0.34, 2.26) (1.03, 5.12) (0.91, 1.00) (0.99, 1.05) (0.89, 1.09) (1.00, 1.14)
NSSI = nonsuicidal self-injury; BDI-II = Beck Depression Inventory, Second Edition; PCL-M = Posttraumatic Stress Disorder Checklist, Military Version; BHS = Beck Hopelessness Scale; BSSI-C = Beck Scale for Suicide Ideation, Current.
significant between-groups differences. Results remained unchanged when adjusting for baseline symptom severity level and demographic variables. The proportions of participants in each group who made a suicide attempt over time are displayed in Fig. 3. Among those who had a history of suicide attempt at baseline, the total number of NSSI episodes was associated with significantly increased risk for suicide attempt during follow-up (Wald χ 2 = 4.07, p = .044; HR = 1.09, 95% CI: 1.00, 1.19) but total number of suicide attempts was not (Wald χ 2 = 1.65, p = .199; HR = .66, 95% CI: .35, 1.24). Among those who had no history of suicide attempt at baseline, the total number of NSSI episodes was associated with significantly increased risk for suicide attempt during follow-up (Wald χ 2 = 3.96, p = .047; HR = 1.07, 95% CI: 1.00, 1.15). Taken together, results suggest that each instance of NSSI at baseline was associated with a 7%–9% increase in risk for suicide attempt during the subsequent two years.
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7 1.00
Proportion with suicide attempt
0.90 0.80 0.70 0.60
No history (n=26)
P = .002
0.50
SA (n=80) NSSI (n=10)
0.40
NSSI+SA (n=36)
0.30 0.20 0.10 0.00 6 mos
12 mos
18 mos
24 mos
Fig. 3. Proportion of participants who made at least one suicide attempt at each follow-up assessment; NSSI = nonsuicidal self-injury, SA = suicide attempt.
4. Discussion Results of the current study highlight the significance of NSSI as a risk factor for suicide attempts among active duty military personnel in outpatient treatment for suicide ideation and/or a recent suicide attempt. Approximately 30% of participants who had previously made a suicide attempt at baseline had also engaged in NSSI, a rate that is consistent with previous estimates among military personnel and veterans [6] but higher than the 14% co-occurrence rate previously reported in a sample of adolescents in outpatient psychiatric treatment [3]. In the current sample there were no gender differences in terms of lifetime rates of NSSI or total number of NSSI episodes. This finding converges with studies conducted in adolescent samples [4,12,20], general population samples [7,21], and military samples [5], although some studies conducted among college students have found higher rates of NSSI among women [6,22]. We also found a significantly higher rate of diagnosis of social anxiety among participants in the NSSI only group, which diverges from previous studies suggesting that social anxiety is higher among those with a history of both NSSI and suicide attempts as compared to those with a history of NSSI or suicide attempt only [23,24]. The reasons for this difference are not clear, but could be related to differences in sample characteristics. In contrast to previous studies, our sample was a predominantly male clinical sample. Because men with social anxiety are significantly more likely to seek out treatment than women [25], the present results may have been influenced by differences in setting and gender distribution. Additional research is needed to further consider this possibility. Consistent with expectations and previous research in adolescent samples [3,13], a history of NSSI was associated with significantly increased risk for suicide attempts during the following 2 years. In contrast, a history of suicide attempt was not a significant predictor of future suicide
5
attempts. Furthermore, the number of NSSI episodes was significantly associated with increased risk for later suicide attempt, with each NSSI conferring 7%–9% increased likelihood for a suicide attempt during follow-up. This converges with previous research [26] and suggests that participants who had repeatedly engaged in NSSI were higher risk than those who had only engaged in NSSI once. Critically, these findings did not change even when taking into account baseline symptom severity and severity of suicidal ideation. Results further indicated that Soldiers who had a history of NSSI and suicide attempt at intake were the highest risk subgroup. These Soldiers were over twice as likely to make a suicide attempt during the subsequent 2 years as Soldiers who had previously made a suicide attempt but had not engaged in NSSI, even though the severity of their psychiatric symptoms and suicide ideation was no different from other, lower risk groups. This pattern is consistent with the fluid vulnerability theory of suicide [27], which posits that some individuals are more likely to make suicide attempts than others regardless of symptom severity because they have elevated baseline risk for suicide. Baseline risk for suicide is comprised of historical and idiographic variables that influence the individual’s propensity to experience suicidal crises and make suicide attempts. For example, individuals who lack emotion regulation and distress tolerance skills are vulnerable to experiencing suicidal crises in response to triggering events and stressors. Supporting this perspective is research indicating that reduction of or escape from negative emotions is primary motive for suicide attempts in military personnel [28]. NSSI similarly functions as a coping strategy for emotional distress [29] and emerges before a first suicide attempt in 90% of military personnel reporting a history of both NSSI and suicide attempt [30]. NSSI may therefore serve as a “stepping stone” to later suicide attempts, a perspective shared by other contemporary models of suicide (e.g., the interpersonal-psychological theory [31]). From the perspective of fluid vulnerability theory, NSSI may therefore serve as a behavioral marker of elevated baseline risk for suicide. Consistent with this perspective, risk for later suicide attempts increased as the number of NSSI episodes increased. Given the emotion regulation function of NSSI, it is possible that repeated engagement in NSSI serves as an indicator of greater deficits in this area. Additional research is needed to test these assumptions and further clarify the relationship of NSSI to suicide attempts in military personnel. Conclusions should be made with consideration due to several limitations. First, the current study was conducted with a predominantly male, active duty sample of US Army personnel. Results therefore might not generalize to service members in other branches of service, veterans who are no longer in active military service, military personnel in the National Guard or Reserve components, and/or non-US military personnel. Furthermore, the considerable gender skew in the current sample (i.e., N 85% male) restricted our ability to consider gender differences. Given that previous
6
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
studies suggest that the relationship between NSSI and suicide attempts may be stronger among women than men [8], conclusions based on the present data may not extend to female service members, especially those with a history of military sexual assault. Additional studies that replicate these findings are needed. Despite these limitations, the current study provides critical information about the relationship between NSSI and suicide attempts in military personnel, and it suggests that additional research in this area could yield valuable developments relevant to suicide prevention in this population. 4.1. Conclusions Taken together, results of the current study suggest that NSSI is a strong predictor of suicide attempts among high-risk military personnel in outpatient treatment, especially among those who have also made a suicide attempt in the past. From a clinical perspective, these findings suggest that clinicians working with military personnel should assess for NSSI in their suicide risk assessments, especially among those who have recently made a suicide attempt. Although a history of suicide attempts has traditionally been a central component to risk formulation with patients, the present results suggest that military personnel who have repeatedly engaged in NSSI and made a suicide attempt likely warrant the greatest amount of clinical attention by clinicians. From a research perspective, these findings suggest that NSSI warrants much greater attention among military personnel than has been received up to this point, and should be assessed in future studies in this population. Given the strong association of NSSI with subsequent suicide attempts, NSSI should be routinely assessed in future studies focused on suicide. Furthermore, treatment studies are needed to test the efficacy of interventions to prevent attempts among military personnel who engage in repetitive NSSI.
[4]
[5]
[6]
[7]
[8]
[9] [10]
[11] [12]
[13]
[14]
[15] [16] [17] [18]
Acknowledgment This project was supported in part through research funding by the Department of Defense award #W81XWH-09-1-0569 (M. David Rudd, Principal Investigator). The views expressed in this article are those of the authors and do not necessarily represent the official position or policy of the U.S. Government, the Department of Defense, the U.S. Army, or the U.S. Air Force.
[19]
[20]
[21]
[22]
References [23] [1] Ramchand R, Acosta J, Burns RM, Jaycox LH, Pernin CG. The war within: preventing suicide in the US military. Santa Monica, CA: RAND Corporation; 2011. [2] Crosby AE, Ortega L, Melanson C. Self-directed violence surveillance: uniform definitions and recommended data elements, version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. [3] Asarnow JR, Porta G, Spirito A, Emslie G, Clarke G, Wagner KD, et al. Suicide attempts and nonsuicidal self-injury in treatment resistant
[24]
[25] [26]
depression in adolescents: findings from the TORDIA Study. J Am Acad Child Adolesc Psychiatry 2011;50:772-81. Nock MK, Joiner TE, Gordon KH, Lloyd-Richardson E, Prinstein MJ. Nonsuicidal self-injury among adolescents: diagnostic correlates and relation to suicide attempts. Psychiatry Res 2006;144:65-72. Klonsky ED, Oltmann TF, Turkheimer E. Deliberate self-harm in a nonclinical population: prevalence and psychological correlates. Am J Psychiatry 2003;160:1501-8. Bryan CJ, Bryan AO. Nonsuicidal self-injury among a sample of United States military personnel and veterans enrolled in college classes. J Clin Psychol 2014;70:874-5. Klonsky ED. Nonsuicidal self-injury in United States adults: prevalence, sociodemographics, topography, and functions. Psychol Med 2011;41:1981-6. Klonsky ED, May AM, Glenn CR. The relationship between nonsuicidal self-injury and attempted suicide: converging evidence from four samples. Abnorm Psychol 2013;122:31-237. Whitlock P, Eckenrode J, Silverman D. Self-injurious behaviors in a college population. Pediatrics 2006;117:1939-48. DiClemente RJ, Ponton LE, Hartley D. Prevalence and correlates of cutting behavior: risk for HIV transmission. J Am Acad Child Adolesc Psychiatry 1991;30:735-9. Klonsky ED, Muehlenkamp JJ. Self-injury: a research review for the practitioner. J Clin Psychol 2007;63:1045-56. Wilkinson P, Kelvin R, Roberts C, Dubicka B, Goodyear I. Clinical and psychosocial predictors of suicide attempts and nonsuicidal selfinjury in the adolescent depression antidepressants and psychotherapy trial (ADAPT). Am J Psychiatry 2011;168:495-501. Guan K, Fox KR, Prinstein MJ. Nonsuicidal self-injury as a timeinvariant predictor of adolescent suicide ideation and attempts in a diverse community sample. J Consult Clin Psychol 2012;80:842-9. Linehan MM, Comtois KA, Brown MZ, Heard HL, Wagner A. Suicide Attempt Self Injury Interview (SASII): development, reliability, and validity of a scale to assess suicide attempts and intentional self-injury. Psychol Assess 2006;18:303-12. Beck AT. Beck Scale for Suicide Ideation. San Antonio, TX: Psychological Corporation; 1991. Beck AT, Steer RA, Brown GK. The Beck Depression Inventory. 2nd ed. San Antonio, TX: Psychological Corporation; 1996. Beck AT, Steer RA. Manual for the Beck Hopelessness Scale. San Antonio, TX: Psychological Corporation; 1988. Weathers F, Litz B, Herman D, Huska J, Keane T. The PTSD Checklist (PCL): reliability, validity, and diagnostic utility. Paper presented at: 9th Annual Meeting of the International Society for Traumatic Stress Studies; 1993. [San Antonio, TX]. Spitzer RL, Gibbon M, Williams JB. Structured Clinical Interview for Axis I DSM-IV Disorders (SCID). Washington, DC: American Psychiatric Association; 1995. 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;37:1183-92. Briere J, Gil E. Self-mutilation in clinical and general population samples: prevalence, correlates, and functions. Am J Orthopsychiatry 1998;68:609-20. Wilcox HC, Arria AM, Caldeira KM, Vincent KB, Pinchevsky GM, O’Grady KE. Longitudinal predictors of past-year non-suicidal self-injury and motives among college students. Psychol Med 2012;42:717-26. Hamza CA, Willoughby T. Nonsuicidal self-injury and suicidal behavior: a latent class analysis among young adults. PLoS One 2013;8:e59955. Chartrand H, Sareen J, Toews M, Bolton JM. Suicide attempts versus nonsuicidal self-injury among individuals with anxiety disorders in a nationally representative sample. Depress Anxiety 2012;29:172-9. Weinstock LS. Gender differences in the presentation and management of social anxiety disorder. J Clin Psychiatry 1999;60(Suppl 9):9-13. Andover MS, Gibb BE. Non-suicidal self-injury, attempted suicide, an suicidal intent among psychiatric inpatients. Psychiatry Res 2010;178:101-5.
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7 [27] Rudd MD. The assessment and management of suicidality. Sarasota, FL: Professional Resource Press; 2006. [28] Bryan CJ, Rudd MD, Wertenberger E. Reasons for suicide attempts in a clinical sample of active duty soldiers. J Affect Disord 2013;144:148-52. [29] Nock MK, Prinstein MJ. A functional approach to the assessment of self-mutilative behavior. J Consult Clin Psychol 2004;72:885-90.
7
[30] Bryan CJ, Bryan AO, May AM, Klonsky ED. Trajectories of suicide ideation, nonsuicidal self-injury, and suicide attempts in a nonclinical sample of military personnel and veterans. Suicide Life Threat Behav, in press. [31] Joiner TE. Why people die by suicide. Cambridge, MA: Harvard University Press; 2005.
ScienceDirect Comprehensive Psychiatry 59 (2015) 1 – 7 www.elsevier.com/locate/comppsych
Nonsuicidal self-injury as a prospective predictor of suicide attempts in a clinical sample of military personnel Craig J. Bryan a, b,⁎, M. David Rudd a, c , Evelyn Wertenberger d , Stacey Young-McCaughon e , Alan Peterson e a
National Center for Veterans Studies b The University of Utah c The University of Memphis d Fort Carson e The University of Texas Health Science Center at San Antonio
Abstract Background: Nonsuicidal self-injury (NSSI) is a risk factor for suicide attempts, but it has received little attention in military populations, for whom suicide rates have doubled over the past decade. In the current study, the relationship of NSSI with future suicide attempts was prospectively examined in a sample of active duty Soldiers receiving outpatient psychiatric treatment for suicide ideation and/or a recent suicide attempt. Methods: Data were collected as part of a two-year prospective study of 152 active duty Soldiers (87% male, 71% Caucasian, mean age = 27.53) in outpatient mental health care who reported current suicide ideation and/or a suicide attempt during the month preceding intake. Suicide attempts and NSSI were assessed using the Suicide Attempt Self Injury Interview. Results: Forty percent of Soldiers with a history of nonsuicidal self-injury and 25% of Soldiers with a history of suicide attempt made a suicide attempt during the 2-year follow-up. Soldiers with a history of nonsuicidal self-injury were more than twice as likely to make a subsequent suicide attempt (hazard ratio [HR] = 2.25, P = .045). Soldiers with a history of suicide attempt were no more likely to make a subsequent suicide attempt than Soldiers without a previous suicide attempt (HR = .88, P = .787). Thirty percent of Soldiers with a history of suicide attempt had also engaged in nonsuicidal self-injury. Forty-two percent of Soldiers with histories of both nonsuicidal self-injury and suicide attempt made a subsequent suicide attempt and were more likely to make a suicide attempt during follow-up than Soldiers with a history of suicide attempt only. Number of NSSI episodes, but not number of suicide attempts, was significantly associated with increased risk for future suicide attempt. Results were unchanged when adjusting for baseline symptom severity. Limitations: Predominantly male, active duty Army sample. Conclusions: Among Soldiers in outpatient mental health care, a history of NSSI is a stronger predictor of future suicide attempts than a history of suicide attempts. Soldiers with a history of both NSSI and suicide attempt are at especially increased risk. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The suicide rate among members of the US military has more than doubled during the past decade [1]. Nonsuicidal self-injury (NSSI), which entails self-directed, deliberate behavior that results in injury or the potential for injury
⁎ Corresponding author at: National Center for Veterans Studies, 260 S. Central Campus Dr., Room 205, Salt Lake City, UT 84105. Tel.: +1 801 587 7978. E-mail address: [email protected] (C.J. Bryan). http://dx.doi.org/10.1016/j.comppsych.2014.07.009 0010-440X/© 2014 Elsevier Inc. All rights reserved.
to oneself without evidence of suicidal intent [2], is a well-established risk factor for suicide ideation and attempts [3–5], but to date it has received little empirical attention among military personnel. Prevalence rates of NSSI among military personnel are estimated to range from 4% to 14% [5,6], making them comparable to rates seen among US adults [5,7,8] and adolescents/young adults [8,9]. In psychiatric samples, NSSI is much more common, with 40%–70% of adolescent psychiatric cases reporting it in their history [3,4,8,10–12]. To date there are no published estimates of the relative prevalence of NSSI in clinical samples of military personnel.
2
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
Studies indicate that NSSI often co-occurs with, and is an especially robust risk factor for, suicide attempts. Up to 70% of adolescents in inpatient psychiatric treatment who have engaged in NSSI, for instance, have also made a suicide attempt [3,4]. Among military personnel, approximately 25% of those who have engaged in NSSI have also made a suicide attempt [6]. As a risk factor for suicide attempts, prospective studies among adolescents suggest that a history of NSSI is a stronger predictor of future suicide ideation [13] and suicide attempts [3] than previous suicide attempts. Similar research has yet to be conducted with military personnel, however. The primary aims of the current study were to prospectively examine the relationship of NSSI with suicide attempts in a clinical sample of military personnel. Two specific hypotheses were tested. First, we hypothesized that military personnel with a history of NSSI would be significantly more likely to make a suicide attempt during the 2-year follow-up period than military personnel with no history of NSSI. Second, we hypothesized that a history of NSSI would demonstrate a stronger association with future suicide attempts than a history of suicide attempts.
Army Medical Center, the University of Utah, and the University of Texas Health Science Center at San Antonio. 2.2. Assessments Suicide attempts and NSSI that occurred prior to intake and during the follow-up period were determined with the Suicide Attempt Self Injury Interview (SASII) [14], a validated clinician-administered interview that differentiates between suicide attempts and NSSI by assessing the intent, desired and expected outcome, medical severity, and other characteristics of intentional self-injury. The intensity of current (i.e., past week) suicide ideation was measured using the 19-item self-report Beck Scale for Suicide Ideation (BSSI-C) [15]. Depression symptom severity was assessed with the 21-item self-report Beck Depression Inventory [16], severity of hopelessness was assessed with the 20-item Beck Hopelessness Scale [17], and posttraumatic stress symptom severity was assessed with the 17-item PTSD Checklist-Military Version [18]. The presence of current psychiatric diagnosis was determined using the Structured Clinical Interviews for Axis I and Axis II DSM-IV disorders [19]. 2.3. Statistical analysis
2. Material and methods 2.1. Participants and procedures Participants were 176 active duty Soldiers participating in a randomized controlled trial testing a brief cognitive-behavior therapy for the reduction of suicide attempts at Fort Carson, Colorado. Participants were predominantly male (n = 153, 86.9%) and ranged in age from 19 to 44 years (M = 27.53, SD = 6.26). Self-identified race was 71.0% Caucasian, 21.6% Hispanic/Latino, 13.1% African–American, 4.5% Native American, 2.3% Pacific Islander, 1.7% Asian, and 8.0% other. Military grade distribution was 47.2% junior enlisted (E1–E4), 22.7% noncommissioned officer (E5–E6), 3.4% senior noncommissioned officer (E7–E9), and 0.6% warrant officer. Participants had served in the military for a mean (SD) of 5.70 (4.48) years, and the majority (81.2%) had deployed at least once. Participants were referred to research staff for determination of eligibility by their outpatient military mental health provider upon disclosure of suicide ideation, or following discharge from an inpatient psychiatric hospitalization for a suicide attempt. Inclusion criteria included current (i.e., past week) suicidal ideation with intent to die and/or a suicide attempt within the past month; active duty military status; age 18 years or older; ability to speak English; and ability to understand and complete informed consent procedures. The only exclusion criterion was the presence of a medical or psychiatric condition that would preclude informed consent or participation in outpatient treatment (e.g., active psychosis or mania). The current study’s procedures were reviewed and approved by the Institutional Review Boards of Madigan
All analyses were conducted using the SAS 9.3. For baseline group comparisons, Mann–Whitney U was used for continuous variables and logistic regression was used for binary variables. Of the 176 participants, 152 (86.3%) met eligibility criteria for the study and were therefore included in prospective analyses. Univariate and multivariate Cox proportional hazard regression model survival analysis of time to the first suicide attempt was used to determine the association of previous NSSI and suicide attempts with subsequent suicide attempt during the 2-year follow-up period. Time to suicide attempt was measured by calculating the total number of days from enrollment to first suicide attempt. For participants without a suicide attempt, the total number of days from enrollment to the last assessment was calculated. The Cox regression model was selected because it utilizes all available data from all participants regardless of dropout or length of follow-up. Estimates of the proportions of participants in each making at least one suicide attempt during the 2-year follow-up period were calculated using the Kaplan–Meier method, which similarly accounts for dropouts and limited follow-up.
3. Results Forty-six (30.3%) participants reported a history of NSSI, and 116 (76.3%) reported a history of suicide attempt. At baseline, there were no gender differences in history of NSSI (30.8% of men, 26.3% of women; OR = .80 [.27, 2.37], P = .689) or history of suicide attempts (78.2% of men, 63.2% of women; OR = .48 [.17, 1.32], P = .156), and no differences in total number of lifetime NSSI episodes (men:
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
M = 1.40, SD = 5.66; women: M = 5.84, SD = 22.61; Mann–Whitney U = .12, P = .902) or total number of lifetime suicide attempts (men: M = 1.34, SD = 1.00; women: M = 1.16, SD = 1.34; U = −1.22, P = .223). In terms of follow-up suicide attempts by NSSI history at intake, there were 12 participants with a history of NSSI (estimated proportion: 39.7%) who made a subsequent suicide attempt as compared to 14 participants without a history of NSSI (estimated proportion: 19.3%). Results of the multivariate Cox regression models predicting time to suicide attempt, controlling for race and symptom severity (i.e., depression, posttraumatic stress, hopelessness, and suicide ideation), indicated that participants with a history of NSSI were more than twice as likely to make a suicide attempt during the 2-year follow-up period than participants with no history of NSSI (Wald χ 2 = 4.02, P = .045; hazard ratio [HR] = 2.25, 95% CI: 1.02, 4.96). Survival curves by history of NSSI are displayed in Fig. 1. In terms of follow-up suicide attempts by suicide attempt history at intake, there were 20 participants with a history of suicide attempt (estimated proportion: 25.7%) who made a subsequent suicide attempt as compared to 6 participants without a history of suicide attempt (estimated proportion: 31.4%). Results of the multivariate Cox regression models predicting time to suicide attempt indicated that a history of suicide attempt was not associated with increased risk for suicide attempt during the 2-year follow-up period (Wald χ 2 = .07, p = .787; HR = .88, 95% CI: .34, 2.26). Survival curves by suicide attempt history are displayed in Fig. 2. Results did not change when history of NSSI and history of suicide attempt were entered into the Cox regression model simultaneously (see Table 2). In terms of co-occurrence of NSSI and suicide attempts, 41 of the 136 (30.1%) participants with a history of suicide
Fig. 1. Survival curves for time to first suicide attempt according to history of nonsuicidal self-injury (NSSI).
3
Fig. 2. Survival curves for time to first suicide attempt according to history of suicide attempt.
attempt had also engaged in NSSI. Participants were next categorized into 4 groups according to their histories of NSSI and suicide attempts: 12 (6.8%) participants reported a history of NSSI but no suicide attempts (i.e., “NSSI only”); 96 (54.0%) reported a history of suicide attempt but no NSSI (i.e., “SA only”); 41 (23.3%) reported a history of both NSSI and suicide attempts (i.e., “NSSI + SA”); and 28 (15.9%) reported neither NSSI nor suicide attempts (i.e., “No History”). Sociodemographic and clinical characteristics of each group are summarized in Table 1. The No History group had a significantly greater proportion of participants self-identifying as Caucasian and Native American than the NSSI + SA group. The No History group also had a significantly smaller proportion of participants self-identifying as Hispanic as compared to all other groups. The 4 groups did not differ from each other in terms of military-specific characteristics or baseline symptom severity. Participants in the NSSI group were significantly more likely to be diagnosed with social anxiety than all other groups. The NSSI and NSSI + SA groups also had a higher proportion of participants diagnosed with borderline personality disorder. There were no difference between the NSSI only and NSSI + SA groups in terms of the total number of lifetime NSSI episodes (U = .27, p = .984) and no difference between the SA only and NSSI + SA groups in terms of the total number of previous suicide attempts (U = 4.59, p = .614). During the 2-year follow-up period, there were 4 suicide attempts in the no history group (estimated proportion: 32.6%), 10 suicide attempts in the SA only group (estimated proportion: 15.7%), 2 suicide attempts in the NSSI only group (estimated proportion: 22.2%), and 10 suicide attempts in the NSSI + SA group (estimated proportion: 41.7%). Relative to the NSSI + SA group, participants in the SA group were approximately 60% less likely to have made a suicide attempt during follow-up (Wald χ 2 = 3.845, P =
4
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
Table 1 Demographic and clinical characteristics by group.
Male Age, M (SD), y Years of service, M (SD), y Race Caucasian African–American Asian Pacific Island Native American Other Hispanic Military rank E1–E4 E5–E6 E7–E9 Warrant officer No. of deployments 0 1 2+ DSM-IV diagnosis Major depressive Social anxiety Posttraumatic stress Alcohol dependence Substance dependence Bipolar I Bipolar II Borderline Symptom severity scales, M (SD) BDI-II PCL-M BHS BSSI-C Total NSSI episodes, median (IQR) Total suicide attempts, median (IQR)
No history (n = 28)
SA (n = 95)
NSSI (n = 12)
NSSI + SA (n = 41)
82.1 27.57 (6.71) 5.32 (4.62)
88.4 27.18 (6.12) 5.23 (4.19)
83.3 27.92 (6.02) 6.33 (4.33)
87.8 28.20 (6.50) 6.83 (5.00)
82.1 a 14.3 3.6 0.0 14.3 a 3.6 3.6 a,b,c
72.6 13.7 1.1 4.2 4.2 4.2 21.1 a
75.0 0.0 0.0 0.0 0.0 8.3 33.3 b
58.5 a 14.6 2.4 0.0 0.0 a 19.5 31.7 c
75.0 17.9 7.1 0.0
74.7 21.1 2.1 1.1
58.3 41.7 0.0 0.0
70.7 24.4 4.8 0.0
17.9 42.9 39.3
23.4 33.0 43.6
16.7 41.7 41.7
9.8 48.8 41.5
75.0 0.0 a 42.9 3.8 7.7 0.0 0.0 7.4 a,b
77.2 7.1 b 35.3 12.2 14.5 1.2 0.0 a 3.4 c,d
70.0 40.0 a,b,c 50.0 10.0 20.0 0.0 10.0 a 25.0 a,c
86.5 11.1 c 51.4 18.9 16.2 0.0 2.7 20.0 b,d
34.82 (11.56) 58.89 (15.91) 12.82 (4.26) 12.32 (9.45) N/A N/A
32.21 (14.09) 54.24 (17.07) 12.41 (6.56) 8.72 (8.34) N/A 1 (1)
31.64 (12.88) 58.64 (19.46) 13.55 (6.91) 6.73 (6.20) 1.5 (2) N/A
30.80 (14.67) 55.05 (18.12) 11.98 (6.37) 10.34 (8.71) 2 (2) 2 (2)
By row, values with the same superscript significantly differ from each other at p b .05; NSSI = nonsuicidal self-injury; SA = suicide attempt; BDI-II = Beck Depression Inventory, Second Edition; PCL-M = Posttraumatic Stress Disorder Checklist, Military Version; BHS = Beck Hopelessness Scale; BSSI-C = Beck Scale for Suicide Ideation, Current.
.050, HR = .40, 95% CI: .16, 1.00). A significant difference between the SA and NSSI + SA groups occurred at the 2-year follow-up (z =3.04, P = .002). There were no other Table 2 Results of Cox regression analyses predicting time to first suicide attempt. Predictor
Wald χ 2
p
HR
(95% C.I.)
Caucasian Hispanic Suicide attempt NSSI BDI-II PCL-M BHS BSSI-C
1.71 1.88 0.07 4.14 3.45 1.57 0.11 3.92
.191 .170 .787 .042 .063 .210 .739 .048
1.96 0.53 0.88 2.30 0.96 1.02 0.98 1.07
(0.72, 5.39) (0.21, 1.31) (0.34, 2.26) (1.03, 5.12) (0.91, 1.00) (0.99, 1.05) (0.89, 1.09) (1.00, 1.14)
NSSI = nonsuicidal self-injury; BDI-II = Beck Depression Inventory, Second Edition; PCL-M = Posttraumatic Stress Disorder Checklist, Military Version; BHS = Beck Hopelessness Scale; BSSI-C = Beck Scale for Suicide Ideation, Current.
significant between-groups differences. Results remained unchanged when adjusting for baseline symptom severity level and demographic variables. The proportions of participants in each group who made a suicide attempt over time are displayed in Fig. 3. Among those who had a history of suicide attempt at baseline, the total number of NSSI episodes was associated with significantly increased risk for suicide attempt during follow-up (Wald χ 2 = 4.07, p = .044; HR = 1.09, 95% CI: 1.00, 1.19) but total number of suicide attempts was not (Wald χ 2 = 1.65, p = .199; HR = .66, 95% CI: .35, 1.24). Among those who had no history of suicide attempt at baseline, the total number of NSSI episodes was associated with significantly increased risk for suicide attempt during follow-up (Wald χ 2 = 3.96, p = .047; HR = 1.07, 95% CI: 1.00, 1.15). Taken together, results suggest that each instance of NSSI at baseline was associated with a 7%–9% increase in risk for suicide attempt during the subsequent two years.
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7 1.00
Proportion with suicide attempt
0.90 0.80 0.70 0.60
No history (n=26)
P = .002
0.50
SA (n=80) NSSI (n=10)
0.40
NSSI+SA (n=36)
0.30 0.20 0.10 0.00 6 mos
12 mos
18 mos
24 mos
Fig. 3. Proportion of participants who made at least one suicide attempt at each follow-up assessment; NSSI = nonsuicidal self-injury, SA = suicide attempt.
4. Discussion Results of the current study highlight the significance of NSSI as a risk factor for suicide attempts among active duty military personnel in outpatient treatment for suicide ideation and/or a recent suicide attempt. Approximately 30% of participants who had previously made a suicide attempt at baseline had also engaged in NSSI, a rate that is consistent with previous estimates among military personnel and veterans [6] but higher than the 14% co-occurrence rate previously reported in a sample of adolescents in outpatient psychiatric treatment [3]. In the current sample there were no gender differences in terms of lifetime rates of NSSI or total number of NSSI episodes. This finding converges with studies conducted in adolescent samples [4,12,20], general population samples [7,21], and military samples [5], although some studies conducted among college students have found higher rates of NSSI among women [6,22]. We also found a significantly higher rate of diagnosis of social anxiety among participants in the NSSI only group, which diverges from previous studies suggesting that social anxiety is higher among those with a history of both NSSI and suicide attempts as compared to those with a history of NSSI or suicide attempt only [23,24]. The reasons for this difference are not clear, but could be related to differences in sample characteristics. In contrast to previous studies, our sample was a predominantly male clinical sample. Because men with social anxiety are significantly more likely to seek out treatment than women [25], the present results may have been influenced by differences in setting and gender distribution. Additional research is needed to further consider this possibility. Consistent with expectations and previous research in adolescent samples [3,13], a history of NSSI was associated with significantly increased risk for suicide attempts during the following 2 years. In contrast, a history of suicide attempt was not a significant predictor of future suicide
5
attempts. Furthermore, the number of NSSI episodes was significantly associated with increased risk for later suicide attempt, with each NSSI conferring 7%–9% increased likelihood for a suicide attempt during follow-up. This converges with previous research [26] and suggests that participants who had repeatedly engaged in NSSI were higher risk than those who had only engaged in NSSI once. Critically, these findings did not change even when taking into account baseline symptom severity and severity of suicidal ideation. Results further indicated that Soldiers who had a history of NSSI and suicide attempt at intake were the highest risk subgroup. These Soldiers were over twice as likely to make a suicide attempt during the subsequent 2 years as Soldiers who had previously made a suicide attempt but had not engaged in NSSI, even though the severity of their psychiatric symptoms and suicide ideation was no different from other, lower risk groups. This pattern is consistent with the fluid vulnerability theory of suicide [27], which posits that some individuals are more likely to make suicide attempts than others regardless of symptom severity because they have elevated baseline risk for suicide. Baseline risk for suicide is comprised of historical and idiographic variables that influence the individual’s propensity to experience suicidal crises and make suicide attempts. For example, individuals who lack emotion regulation and distress tolerance skills are vulnerable to experiencing suicidal crises in response to triggering events and stressors. Supporting this perspective is research indicating that reduction of or escape from negative emotions is primary motive for suicide attempts in military personnel [28]. NSSI similarly functions as a coping strategy for emotional distress [29] and emerges before a first suicide attempt in 90% of military personnel reporting a history of both NSSI and suicide attempt [30]. NSSI may therefore serve as a “stepping stone” to later suicide attempts, a perspective shared by other contemporary models of suicide (e.g., the interpersonal-psychological theory [31]). From the perspective of fluid vulnerability theory, NSSI may therefore serve as a behavioral marker of elevated baseline risk for suicide. Consistent with this perspective, risk for later suicide attempts increased as the number of NSSI episodes increased. Given the emotion regulation function of NSSI, it is possible that repeated engagement in NSSI serves as an indicator of greater deficits in this area. Additional research is needed to test these assumptions and further clarify the relationship of NSSI to suicide attempts in military personnel. Conclusions should be made with consideration due to several limitations. First, the current study was conducted with a predominantly male, active duty sample of US Army personnel. Results therefore might not generalize to service members in other branches of service, veterans who are no longer in active military service, military personnel in the National Guard or Reserve components, and/or non-US military personnel. Furthermore, the considerable gender skew in the current sample (i.e., N 85% male) restricted our ability to consider gender differences. Given that previous
6
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7
studies suggest that the relationship between NSSI and suicide attempts may be stronger among women than men [8], conclusions based on the present data may not extend to female service members, especially those with a history of military sexual assault. Additional studies that replicate these findings are needed. Despite these limitations, the current study provides critical information about the relationship between NSSI and suicide attempts in military personnel, and it suggests that additional research in this area could yield valuable developments relevant to suicide prevention in this population. 4.1. Conclusions Taken together, results of the current study suggest that NSSI is a strong predictor of suicide attempts among high-risk military personnel in outpatient treatment, especially among those who have also made a suicide attempt in the past. From a clinical perspective, these findings suggest that clinicians working with military personnel should assess for NSSI in their suicide risk assessments, especially among those who have recently made a suicide attempt. Although a history of suicide attempts has traditionally been a central component to risk formulation with patients, the present results suggest that military personnel who have repeatedly engaged in NSSI and made a suicide attempt likely warrant the greatest amount of clinical attention by clinicians. From a research perspective, these findings suggest that NSSI warrants much greater attention among military personnel than has been received up to this point, and should be assessed in future studies in this population. Given the strong association of NSSI with subsequent suicide attempts, NSSI should be routinely assessed in future studies focused on suicide. Furthermore, treatment studies are needed to test the efficacy of interventions to prevent attempts among military personnel who engage in repetitive NSSI.
[4]
[5]
[6]
[7]
[8]
[9] [10]
[11] [12]
[13]
[14]
[15] [16] [17] [18]
Acknowledgment This project was supported in part through research funding by the Department of Defense award #W81XWH-09-1-0569 (M. David Rudd, Principal Investigator). The views expressed in this article are those of the authors and do not necessarily represent the official position or policy of the U.S. Government, the Department of Defense, the U.S. Army, or the U.S. Air Force.
[19]
[20]
[21]
[22]
References [23] [1] Ramchand R, Acosta J, Burns RM, Jaycox LH, Pernin CG. The war within: preventing suicide in the US military. Santa Monica, CA: RAND Corporation; 2011. [2] Crosby AE, Ortega L, Melanson C. Self-directed violence surveillance: uniform definitions and recommended data elements, version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. [3] Asarnow JR, Porta G, Spirito A, Emslie G, Clarke G, Wagner KD, et al. Suicide attempts and nonsuicidal self-injury in treatment resistant
[24]
[25] [26]
depression in adolescents: findings from the TORDIA Study. J Am Acad Child Adolesc Psychiatry 2011;50:772-81. Nock MK, Joiner TE, Gordon KH, Lloyd-Richardson E, Prinstein MJ. Nonsuicidal self-injury among adolescents: diagnostic correlates and relation to suicide attempts. Psychiatry Res 2006;144:65-72. Klonsky ED, Oltmann TF, Turkheimer E. Deliberate self-harm in a nonclinical population: prevalence and psychological correlates. Am J Psychiatry 2003;160:1501-8. Bryan CJ, Bryan AO. Nonsuicidal self-injury among a sample of United States military personnel and veterans enrolled in college classes. J Clin Psychol 2014;70:874-5. Klonsky ED. Nonsuicidal self-injury in United States adults: prevalence, sociodemographics, topography, and functions. Psychol Med 2011;41:1981-6. Klonsky ED, May AM, Glenn CR. The relationship between nonsuicidal self-injury and attempted suicide: converging evidence from four samples. Abnorm Psychol 2013;122:31-237. Whitlock P, Eckenrode J, Silverman D. Self-injurious behaviors in a college population. Pediatrics 2006;117:1939-48. DiClemente RJ, Ponton LE, Hartley D. Prevalence and correlates of cutting behavior: risk for HIV transmission. J Am Acad Child Adolesc Psychiatry 1991;30:735-9. Klonsky ED, Muehlenkamp JJ. Self-injury: a research review for the practitioner. J Clin Psychol 2007;63:1045-56. Wilkinson P, Kelvin R, Roberts C, Dubicka B, Goodyear I. Clinical and psychosocial predictors of suicide attempts and nonsuicidal selfinjury in the adolescent depression antidepressants and psychotherapy trial (ADAPT). Am J Psychiatry 2011;168:495-501. Guan K, Fox KR, Prinstein MJ. Nonsuicidal self-injury as a timeinvariant predictor of adolescent suicide ideation and attempts in a diverse community sample. J Consult Clin Psychol 2012;80:842-9. Linehan MM, Comtois KA, Brown MZ, Heard HL, Wagner A. Suicide Attempt Self Injury Interview (SASII): development, reliability, and validity of a scale to assess suicide attempts and intentional self-injury. Psychol Assess 2006;18:303-12. Beck AT. Beck Scale for Suicide Ideation. San Antonio, TX: Psychological Corporation; 1991. Beck AT, Steer RA, Brown GK. The Beck Depression Inventory. 2nd ed. San Antonio, TX: Psychological Corporation; 1996. Beck AT, Steer RA. Manual for the Beck Hopelessness Scale. San Antonio, TX: Psychological Corporation; 1988. Weathers F, Litz B, Herman D, Huska J, Keane T. The PTSD Checklist (PCL): reliability, validity, and diagnostic utility. Paper presented at: 9th Annual Meeting of the International Society for Traumatic Stress Studies; 1993. [San Antonio, TX]. Spitzer RL, Gibbon M, Williams JB. Structured Clinical Interview for Axis I DSM-IV Disorders (SCID). Washington, DC: American Psychiatric Association; 1995. 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;37:1183-92. Briere J, Gil E. Self-mutilation in clinical and general population samples: prevalence, correlates, and functions. Am J Orthopsychiatry 1998;68:609-20. Wilcox HC, Arria AM, Caldeira KM, Vincent KB, Pinchevsky GM, O’Grady KE. Longitudinal predictors of past-year non-suicidal self-injury and motives among college students. Psychol Med 2012;42:717-26. Hamza CA, Willoughby T. Nonsuicidal self-injury and suicidal behavior: a latent class analysis among young adults. PLoS One 2013;8:e59955. Chartrand H, Sareen J, Toews M, Bolton JM. Suicide attempts versus nonsuicidal self-injury among individuals with anxiety disorders in a nationally representative sample. Depress Anxiety 2012;29:172-9. Weinstock LS. Gender differences in the presentation and management of social anxiety disorder. J Clin Psychiatry 1999;60(Suppl 9):9-13. Andover MS, Gibb BE. Non-suicidal self-injury, attempted suicide, an suicidal intent among psychiatric inpatients. Psychiatry Res 2010;178:101-5.
C.J. Bryan et al. / Comprehensive Psychiatry 59 (2015) 1–7 [27] Rudd MD. The assessment and management of suicidality. Sarasota, FL: Professional Resource Press; 2006. [28] Bryan CJ, Rudd MD, Wertenberger E. Reasons for suicide attempts in a clinical sample of active duty soldiers. J Affect Disord 2013;144:148-52. [29] Nock MK, Prinstein MJ. A functional approach to the assessment of self-mutilative behavior. J Consult Clin Psychol 2004;72:885-90.
7
[30] Bryan CJ, Bryan AO, May AM, Klonsky ED. Trajectories of suicide ideation, nonsuicidal self-injury, and suicide attempts in a nonclinical sample of military personnel and veterans. Suicide Life Threat Behav, in press. [31] Joiner TE. Why people die by suicide. Cambridge, MA: Harvard University Press; 2005.
