Suicide and Life-Threatening Behavior © 2015 The American Association of Suicidology DOI: 10.1111/sltb.12155

1

Frequency of Nonsuicidal Self-Injury in Adolescents: Differences in Suicide Attempts, Substance Use, and Disordered Eating AMY M. BRAUSCH, PHD,

AND

SHANNON D. BOONE, BA

The relationship between frequency of nonsuicidal selfinjury (NSSI) and suicide attempts, substance use, and disordered eating was assessed in a community sample of 4,839 adolescents, 922 of whom reported NSSI in the past year. It was expected that the engagement in risk behaviors would significantly increase as NSSI frequency increased. Participants completed the Youth Risk Behavior Survey (Centers for Disease Control and Prevention, 2009) and were subdivided into five NSSI frequency groups: none, 1 time, 2–3 times, 4–5 times, and 6 or more times. A one-way MANOVA found significant mean differences for all variables across NSSI frequency groups. The no NSSI group was significantly lower than all other groups on all variables. For suicide attempts, all frequency groups were significantly different from each other, with attempt frequency increasing with each increase in NSSI frequency. The six or more group reported significantly more substance use and disordered eating than all other groups. Overall, adolescents with more frequent NSSI represent a group at risk for concurrent unhealthy behaviors and suicide attempts. Adolescent nonsuicidal self-injury (NSSI) and suicidal behavior are recognized as a continuing health crisis. While much has been learned about self-harm behaviors in youth, there are many lingering questions (Nock, 2012). Prevalence rates of NSSI in community samples of adolescents have held steady around 15% to 30% in the United States (Brausch & Gutierrez, 2010), with similar rates in Europe (Giletta, Scholte, Engels, Ciairano, & Prinstein, 2012). Meanwhile, overall suicide deaths in 15- to 24-year-olds have slowly increased in recent years; the most recent data show 4,822 deaths in 2011, an increase of almost 5% from the previous year (McIntosh & Drapeau, 2014).

The growing body of research on NSSI includes valuable information on prevalence, function, rates of engagement by gender and function, psychosocial correlates, concurrent psychopathology, and the relationship between NSSI and suicide attempts (Nock, 2010). Previous studies have examined NSSI in various samples and have defined the behavior in multiple ways, with many studies categorizing participants on a dichotomous NSSI variable based on their report of any versus no lifetime incidents of NSSI. While this method has been helpful in the early stages of research on differences between adolescents who selfinjured and those who did not, there is a lack of evidence on the frequency of NSSI

AMY M. BRAUSCH, Psychological Sciences, Western Kentucky University, Bowling Green, KY, USA; SHANNON D. BOONE, Psychology, Western Kentucky University, Bowling Green, KY, USA.

Address correspondence to Amy M. Brausch, Psychological Sciences, Western Kentucky University, Bowling Green, KY 42101; E-mail: amy. [email protected]

2

FREQUENCY

and how increased frequency may relate to both suicide attempts and other health behaviors. The goal of the current study was to determine whether increased frequency of NSSI was related to more suicide attempts and increased substance use and disordered eating, compared with lesser frequencies of NSSI in a large community sample of adolescents. Theories Linking NSSI and Suicide Existing literature has identified several theories that explain the link between NSSI and suicidal behavior. The gateway theory views self-harm behavior on a continuum ranging from NSSI to completed suicide; individuals who engage in NSSI at one end of the spectrum may use it as a gateway to more severe self-harm behavior such as suicide attempts. Cross-sectional and longitudinal studies support this theory such that NSSI is predictive of suicidal behavior (Andover & Gibb, 2010; Prinstein et al., 2008). Several longitudinal studies (Whitlock et al., 2013) also support Joiner’s (2005) Interpersonal Theory of Suicide, which argues that NSSI represents a painful and provocative event that, through repetition and time, leads to habituation to pain and fear of death, which may increase risk for a suicide attempt. However, within this theory, NSSI is one of many painful and provocative events that can lead to increased acquired capability for suicide; other behaviors include substance abuse and disordered eating (Fink, Bodell, Smith, & Joiner, 2013; Neumark-Sztainer, Story, Dixon, & Murray, 1998; Van Orden et al., 2010). The third variable theory states that an extraneous variable exists that accounts for the correlation between NSSI and suicide, with supporters of this theory pointing out high rates of psychiatric disorders in individuals who die by suicide and who engage in NSSI (Cavanagh, Carson, Sharpe, & Lawrie, 2003; Nock, Joiner, Gordon, Lloyd-Richardson, & Prinstein, 2006). Further research regarding the predictive capabilities of NSSI frequency and other risk factors for suicide

OF

NSSI

attempts would enhance the accuracy and reliability of risk assessments. NSSI and Suicide Attempts The vast majority of studies that examine the link between NSSI and suicide attempts rely on cross-sectional designs that show concurrence of the behaviors. Previous studies have shown the overlap of these behaviors in adolescent psychiatric inpatients (Nock et al., 2006), college students (Klonsky & Glenn, 2009), and community samples of adolescents (Brausch & Gutierrez, 2010). In a comprehensive review of the link between NSSI and suicidal behavior, Hamza, Stewart, and Willoughby (2012) reported that across the 18 studies of adolescents and adults reviewed, NSSI was a robust predictor of both suicidal thoughts and behaviors and predicted suicidal behavior above and beyond depression, hopelessness, family functioning, and other psychopathology. Similarly, Klonsky, May, and Glenn (2013) identified NSSI as a more reliable predictor of suicide attempts than depression, anxiety, impulsivity, and BPD. While this study was a retrospective, crosssectional design, Hamza and colleagues’ literature review located three longitudinal studies in which NSSI was a significant predictor of suicidal behavior. However, these longitudinal studies used only clinical samples of adolescents and short periods of time between the multiple assessment points. Overall, NSSI is clearly a risk factor for suicide, and the link between the two should continue to be studied. Specific findings have emerged from studies that have investigated frequency of NSSI as it relates to other variables. Overall, adolescents who engage in more frequent NSSI also show more frequent suicide attempts (Brunner et al., 2007; Guan, Fox, & Prinstein, 2012; Prinstein et al., 2008). Other studies have found that as frequency of NSSI increases, the frequency of suicide attempts also increases, but only to a point. Whitlock and Knox’s (2007) study of university students found

BRAUSCH

AND

BOONE

7

transformation improved the distribution with a resulting skewness of 2.38 (SE = .04) and kurtosis of 4.65 (SE = .07). A one-way MANOVA was then used to test for group differences between the five NSSI groups on number of suicide attempts in the past year (transformed variable), level of substance use in the past 30 days, and level of disordered eating behaviors in the past 30 days. Significant mean differences were found for all variables across NSSI frequency groups (Table 1). Planned pairwise comparisons were conducted using Tukey’s test to investigate significant mean differences between all NSSI frequency groups. The data were analyzed using MANOVA rather than regression due to the nature of the existing data. The NSSI frequency item was available with predetermined response options for frequencies, which could not be recoded. Therefore, this variable would not have been a true continuous variable, and using the predetermined categories of frequencies was thought to be preferable. Results showed that for suicide attempts, all frequency groups were found to be

significantly different from each other, indicating an increase in suicide attempt frequency as NSSI frequency increased. For disordered eating, the no NSSI group was significantly lower than all other groups and the six or more NSSI group was significantly higher than all other groups. For substance use, all frequency groups were significantly different from each other except for the 2–3 times and 4–5 times groups, with substance use in the past 30 days increasing with each NSSI frequency group (Table 1).

DISCUSSION

Results from the current study indicated that all NSSI frequency groups were significantly different from each other. Therefore, among the adolescent sample, for each increase of NSSI frequency reported, the number of suicide attempts also increased. These results concur with and provide additional evidence for the relationship between NSSI frequency and

TABLE 1

Means, Standard Deviations, Significant Group Differences, and Effect Sizes by NSSI Frequency on Suicide Attempts, Disordered Eating, and Substance Use NSSI Frequency Groups Subscale Suicide attempts Disordered eating Substance use

No NSSI (n = 3,589)

1 time (n = 271)

2–3 times (n = 245)

4–5 times (n = 115)

0.08a (0.29)

0.29a (0.51)

0.42a (0.62)

0.50a (0.63)

0.05b (0.85)

1.03 (1.11)

1.13 (1.21)

1.20 (1.29)

3.33b (6.54)

5.99 (7.74) 8.41d (10.05)

8.73d (10.27)

6+ times (n = 291)

F value

0.78a (0.80) 315.5* 1.45c (1.29)

g2 .22

86.29* .07

14.58c (16.94) 154.29* .12

Notes. NSSI = Nonsuicidal self-injury; NSSI frequency is reported in the past 12 months. Suicide attempts = transformed number of reported attempts in past 12 months. Disordered eating = sum of four items from Youth Risk Behavior Surveillance assessing for unhealthy weight management practices (possible score range 0–4). Substance use = sum of 10 items that assessed frequency of substance and alcohol use in past 30 days (possible score range 0–50). a In post hoc analyses, all frequency group means were significantly different from each other. b The No NSSI group was significantly lower than all other groups. c The 6 or more times group mean was significantly higher than all other groups. d The 2–3 and 4–5 groups were significantly different from all other groups except each other. *p < .001.

4

FREQUENCY

for NSSI and substance use is different. Substance use is used for both positive and negative reinforcement, while NSSI seems to mainly be used for negative reinforcement. Additional studies have found associations between NSSI and alcohol use in high school students (Bj€arehed & Lundh, 2008). Another study noted that adolescents who met criteria for the DSM-5 proposed NSSI disorder were more likely to be smokers, have tried drugs, and consumed alcohol more frequently than adolescents with NSSI who did not meet criteria for the disorder (Zetterqvist, Lundh, Dahlstrom, & Svedin, 2013), providing evidence that increased frequency of NSSI may be related to increased substance use. Lastly, a study comparing NSSI prevalence and correlates across three countries found daily smoking and more frequent marijuana use to be related to NSSI, but not binge drinking. Further analyses also found that the relationship between NSSI and cigarette smoking was exclusive to the adolescent sample from the United States, indicating that the overlap of NSSI and substance use may differ across cultures (Giletta et al., 2012). Study Overview Previous research has demonstrated a clear link between NSSI and suicidal behaviors, as well as between NSSI and risk behaviors such as substance use and disordered eating. Research is lacking on whether or not increased frequency of NSSI impacts the likelihood that concurrent risk factors will be found among adolescents. Many cross-sectional studies examine NSSI as being present or absent, rather than examining different frequencies. The goal of the current study was to fill these gaps by examining group differences in a sample of community adolescents by comparing the number of suicide attempts, level of substance use, and level of disordered eating behavior across different frequencies of NSSI. This study assessed NSSI and suicide attempts within the past 12 months, and substance use and disordered eating within

OF

NSSI

the last 30 days. The sample was large enough to examine group differences between NSSI frequencies of none, once, 2–3 times, 4–5 times, and 6 or more. It was expected that the highest frequency NSSI group would report the most suicide attempts and highest levels of substance use and disordered eating compared with the other NSSI frequency groups. Number of suicide attempts and level of other risk behaviors were expected to decrease with each corresponding NSSI frequency group.

METHOD

Participants and Procedure Data for this study were accessed from a large regional mental health screening project in a rural area of a midwestern state in the United States. High school students in a seven-county region were assessed biennially using the Youth Risk Behavior Surveillance System (YRBS; Centers for Disease Control and Prevention [CDC], 2009). The data collection occurred in the spring of 2010 from the large sample of adolescents in 25 different high schools in the seven-county region. All 27 high schools in the region received the opportunity to participate in the survey, and 25 (92.6%) participated in 2010. Regional enrollment for all high schools for the academic year was 7,434, and 4,839 students completed the survey, for a participation rate of 65%. The biennial screening is conducted by the staff of a local coalition and is sponsored by the community hospital. The screening is to inform local prevalence and prevention purposes, and received approval from the hospital’s human subjects review board. Similarly, the author consulted with the university institutional review board with whom the author was previously affiliated, and received approval for analyzing an archival data set. The coalition utilized passive parental consent, and students were not asked to sign assent documents to protect confidentiality of students

BRAUSCH

AND

BOONE

at schools with low enrollment (e.g., total student body

Frequency of Nonsuicidal Self-Injury in Adolescents: Differences in Suicide Attempts, Substance Use, and Disordered Eating.

The relationship between frequency of nonsuicidal selfinjury (NSSI) and suicide attempts, substance use, and disordered eating was assessed in a commu...
106KB Sizes 0 Downloads 8 Views