Nonsuicidal Self-Injury and Gender: Patterns of Prevalence, Methods, and Locations among Adolescents
Abstract
Nonsuicidal self-injury (NSSI) among adolescents is a growing concern. However, little is known about gender and features of this behavior. Gender differences in NSSI among a sample of 7,126 adolescents were investigated, 1,774 of whom reported having engaged in NSSI. Gender differences in prevalence, method, and location of NSSI were examined. Findings revealed that females reported higher rates of NSSI, more cutting and scratching, and more injuries to arms and legs than their male counterparts. Males reported more burning and hitting-type behavior, as well as injuries to the chest, face, or genitals. This highlights an interesting pattern of NSSI, which future research should consider to accurately examine NSSI in females and males.
Often perceived to be restricted to cutting behaviors observed among teenage girls, the study of nonsuicidal self-injury (NSSI) has shown that this behavior can actually affect both genders and take many forms. NSSI is defined as “purposely inflicting injury that results in immediate tissue damage, done without suicidal intent and not socially sanctioned” (Nixon & Heath, 2008, p. 4). It can include a number of behaviors, such as cutting, scratching, burning, and hitting oneself. Likewise, individuals who hurt themselves have reported injuring a wide variety of locations on the body.
Estimated prevalence rates have contributed to the growing concern regarding this behavior, with the majority of community-based studies finding that between 14% and 24% of adolescents report having engaged in NSSI at least once in their lifetime (Heath, Schaub, Holly, & Nixon, 2008). Although NSSI is different from suicide, and by definition is nonsuicidal in intent, research has shown that specific features of NSSI are related to a higher risk for suicide attempt; the longer one engages in NSSI, the more methods one uses, and the less physical pain one experiences during NSSI are all associated with suicide attempt (Nock, Joiner, Gordon, Lloyd-Richardson, & Prinstein, 2006). Although many people who self-injure may never attempt suicide, NSSI is still a salient risk factor for suicidal behavior, including both attempted and completed suicide; although the rate of completed suicide of those who engage in NSSI is unknown, the majority of those who self-injure report at least one lifetime suicide attempt (Sher & Stanley, 2009).
Studies that explore the functions of NSSI have helped to increase the overall understanding of this behavior and continue to inform and contextualize research (Klonsky, 2007). Individuals often engage in NSSI as a means of regulating overwhelming negative emotion. Indeed, emotion dysregulation has been found to the most commonly reported function of this behavior (e.g., Gratz, 2003; Heath, Toste, Nedecheva, & Charlebois, 2008; Klonsky, 2007; Nock & Prinstein, 2004).
Because NSSI is considered to be a maladaptive coping strategy used to regulate emotion and it is known that females and males tend to differ in the ways in which they manage and regulate their emotions (Offer & Schonert Reichl, 1992; Underwood, 1997), it is possible that NSSI behaviors would manifest differently in females and males. Emerging research seems to suggest that such differences exist (e.g., Izutsu et al., 2006), but the known details of these differences are limited. Studies of community-based adolescent samples have not found consistent results with regard to gender and NSSI prevalence. Some studies suggest that females do engage in NSSI more than males (e.g., Muehlenkamp & Gutierrez, 2007), while others have found similar rates between female and male adolescents (e.g., Lloyd-Richardson, Perrine, Dierker, & Kelley, 2007).
There are a number of different factors that can affect both the overall observed prevalence rates, as well as possible gender differences in NSSI prevalence. The way in which NSSI is defined can either inflate prevalence rates or exclude behaviors that would otherwise fall under the definition of the behavior. For example, when researchers explicitly include the method “picking at a wound” in their definition of NSSI, they tend to find prevalence rates ranging from 37% to 47% (e.g., Lloyd-Richardson et al., 2007; Yates, Tracy, & Luthar, 2008). In fact, Yates et al., (2008) found that this specific variable produced a disproportionately high level of endorsement among participants. They found that after removing that one variable, gender differences in prevalence became significant, when they had not previously been so.
Likewise, methodological variations between studies can affect these findings; research has shown that prevalence rates can differ depending on whether the study uses self-report checklists or open-ended questions about NSSI, and on whether anonymous surveys or individual interviews are used (Heath, Schaub, Holly, & Nixon, 2008). Studies that assess NSSI using the Functional Assessment of Self-Mutilation (FASM; Lloyd-Richardson et al., 2007), a checklist of all possible NSSI behaviors, tend to find higher rates than those that use anonymous surveys and interviews; this method of assessment broadens and obscures the definition of NSSI (Heath, Schaub, Holly, & Nixon, 2008).
Beyond prevalence rates, little is known about how females and males differ in other characteristics of NSSI, such as the methods used to injure or the locations on the body that are injured. Although NSSI methods are generally low in their risk for lethality (Walsh, 2006), research suggests a positive correlation between number of NSSI methods used and history of suicide attempts (Nock et al., 2006). Additionally, some methods, such as cutting, cause more severe tissue damage than others, such as scratching (Klonsky & Glenn, 2008). Location on the body has also been linked to the risk of the behavior; when an individual changes the location at which they injure themselves, it can signify a change in their psychological well-being (Walsh, 2006). Because of the limited amount of research exploring specific methods and locations of NSSI and the risk of lethality that they pose, more understanding is needed about these features of NSSI and the ways in which different people use different methods and locations.
Current Knowledge of Gender and NSSI
Prevalence Most research on NSSI in community samples has focused on either adolescent or young adult samples. The limited research on gender differences in NSSI has yielded conflicting results for prevalence rates; there are studies that show females engaging in NSSI more than males, and studies that show both genders engaging in NSSI at comparable rates. However, on close review, it appears that the majority of studies that find a gender difference focus specifically on adolescent samples (e.g., Laye-Gindhu & Schonert-Reichl, 2005; Muehlenkamp & Gutierrez, 2007; Muehlenkamp, Williams, Gutierrez, & Claes, 2009; Nixon, Cloutier, & Jansson, 2008; Plener, Libal, Keller, Fegert, & Muehlenkamp, 2009; Ross & Heath, 2002; Yates et al., 2008). On the other hand, studies that fail to find such a difference tend to study samples of young adults (e.g., Heath, Schaub, Holly, & Nixon, 2008). Additionally, studies that failed to find a gender difference in adolescent samples do not necessarily rule out the hypothesis that such differences exist in this age group. For example, Muehlenkamp and Gutierrez (2004) did not find significant gender differences in prevalence among a group of adolescents who had engaged in NSSI but had never attempted suicide (NSSI only). At first glance, this would suggest no gender differences in prevalence; however, they did find significant gender differences in prevalence within a group who had both engaged in NSSI and had attempted suicide (NSSI + SA). These findings are consistent with other studies; Muehlenkamp and Gutierrez later found a significant difference between the prevalence of females (8.6%) and males (4.1%) who fell into the NSSI + SA group. Plener et al. (2009) also found significant gender differences, with 6.1% of females and 3.5% of males falling into an NSSI + SA group.
This pattern would seem to suggest that gender differences in NSSI can be expected within adolescent samples, if not young adult samples. Two implications specific to adolescents can be inferred from this pattern. First, if there are differences between female and male adolescents in their rates of NSSI, there may also be differences in other features of the behavior, such as the methods used or the locations on the body that are injured. Additionally, if gender differences in prevalence rates do not remain stable from adolescence to young adulthood, it is possible that gender differences in other features of NSSI are also dynamic. Therefore, results from studies of young adult samples may not necessarily apply to adolescents. It is important, then, to examine these features specifically within an adolescent sample.
Methods Although many studies have examined gender differences in prevalence, few studies have indicated whether or not there exist such differences in the reported methods used to self-injure. The research studies that have examined this characteristic of the behavior have provided few details on gender and methods of NSSI, with a resulting gap in the literature.
Lloyd-Richardson et al. (2007) separated different NSSI behaviors into two groups based on the clinical severity of the injury: moderate/severe NSSI (i.e., cutting/carving on skin, burning skin, giving self a tattoo, scraping skin, erasing skin) and minor NSSI (i.e., hitting self on purpose, pulling hair out, inserting objects under nails or skin, picking at skin or wounds to draw blood). No differences were found between females and males between the moderate/severe and minor NSSI groups. Although females and males were not compared for specific methods of NSSI, these results would suggest that gender is not related to the severity of the type of injury.
Differences exist in the frequency of reporting certain methods of NSSI. For example, cutting and scratching is the most frequently reported form of NSSI for all adolescents in general. It is the most commonly reported form of NSSI for females, but the second most common form for males (behind hitting-type behaviors; Laye-Gindhu & Schonert-Reichl, 2005). Although cutting-type behavior was not the method reported the most often for boys, it was not statistically reported whether or not females and males differ in their use of cutting and scratching; a sizable proportion of the male sample in this study still reported engaging in cutting and scratching for NSSI. Therefore, although females and males may differ somewhat on the order of methods most frequently reported, it is not clear whether adolescent females and males differ on cutting, or on other specific methods of NSSI. Few studies examining such differences exist, and those that do often conflict with one another. For example, one study reports that males are more likely to self-hit (Izutsu et al., 2006), while another finds no difference in this behavior (Zoroglu et al., 2003). Other studies have reported that gender had a salient effect on individual methods of NSSI, but do not specify which methods or in which direction (Yates et al., 2008). Clearly, more information is needed about gender differences for methods of NSSI among adolescents.
Body Locations No studies examining NSSI among community-based samples of adolescents have directly analyzed the effect of gender on the location of the body where the self-injury occurs. This does not mean that there are no gender differences for location; such differences have been observed in other age groups. For example, one study that examined NSSI in young adults found that females in that age group were more likely than their male peers to injure the wrists and thighs; male young adults were more likely than females to injure the hands (Whitlock, Eckenrode, & Silverman, 2006). However, considering that gender differences in prevalence rates do not appear consistent between studies of adolescents and those of young adults, it is not known whether these findings for locations of NSSI can be generalized to adolescents.
Research Objectives
The goal of this study was to investigate differences between female and male adolescents who engage in NSSI with regard to the characteristics of the behavior. Specifically, this study compared high school-aged females and males on reported prevalence rates of NSSI, endorsed methods of self-injury, and the injured locations on the body. This study focused specifically on adolescents, as recent literature seems to indicate that gender-based patterns of NSSI may be specific to certain age groups. To investigate these differences, we had three specific objectives for the study.
Our first objective was to determine and compare the percentage of females and males who reported engaging in NSSI within a sample of high school students. It was anticipated that a significantly higher percentage of females would report engaging in NSSI than males. This hypothesis was based on the body of literature on high school samples which has predominantly found that females of this age report higher prevalence rates than their male counterparts (Laye-Gindhu & Schonert-Reichl, 2005; Muehlenkamp & Gutierrez, 2007; Nixon et al., 2008; Plener et al., 2009; Ross & Heath, 2002; Yates et al., 2008). In addition to the above hypothesis, gender differences in NSSI frequency, number of methods, and lifetime number of NSSI acts were also measured, although no hypotheses were posed regarding these variables as these aspects remain aspect was largely exploratory.
The second objective was to compare the methods of NSSI endorsed by females and males in a high school sample. It was hypothesized that overall, cutting, scratching, and hitting oneself would be the methods endorsed by the most participants (Laye-Gindhu & Schonert-Reichl, 2005; Nixon et al., 2008; Ross & Heath, 2002). Drawing on the literature that indicates gender differences in overall method use in both adolescents (Yates et al., 2008) and adults (Klonsky & Glenn, 2008), it was hypothesized that males and females would differ significantly on choice of methods.
The third objective was to compare the locations where female and male adolescents report having injured themselves. It was hypothesized that females would be more likely than males to report injuring their arms and legs. This hypothesis was based on previous findings observed in the young adult literature, and the generalization of these findings to an adolescent sample (Whitlock et al., 2006). The current study is the first to investigate gender differences in location of NSSI in an adolescent sample and, as such, no specific hypotheses were posited regarding other reported locations on the body.
Method
Participants
Data were collected from a sample of 7,126 high school students in the greater Kansas City metropolitan area, recruited to complete a survey on health-related behaviors. The sample consisted of 3,623 females (50.8%) and 3,503 males (49.2%), who ranged in age from 11 to 19 years (M = 14.92; SD = 1.61). Participants were recruited from classes in grades 6 through 12 across 11 school districts in the greater Kansas City metro area; this included 13 high schools and 18 middle schools. Participants were randomly selected from these schools to avoid sample selection bias. The sample was comprised of 15% of the student population from each grade at each school. The parents of all randomly selected participants were informed about the nature of the survey through a letter sent home with the child. Parents could choose to refuse consent; a total of six students did not participate for this reason. The majority of participants reported their ethnicity as White (67.0%), followed by Black (14.5%), Hispanic (5.7%), Multi-ethnic (4.7%), Asian/Pacific Islander (2.1%), Native Alaskan/Native American (1.6%), and other (4.5%).
From the total sample, 1,859 participants (26.1%) indicated that they had ever physically hurt themselves on purpose. On a follow-up question regarding suicidality, 115 of those who reported ever having hurt themselves also reported that they had only ever done so with suicidal intent. That is, these participants had never actually engaged in NSSI; they had only hurt themselves wanting to die. As such, these participants were excluded from the NSSI group. The remaining participants who had ever hurt themselves on purpose comprised the NSSI group for this study (n = 1,744). This group was comprised of 1,163 females (66.7%) and 581 males (33.3%). The age range of the NSSI group was 11–19 years (M = 14.78; SD = 1.49). Reported ethnicity within the NSSI group was White (67.4%), Black (10.8%), Hispanic (6.3%), Multi-ethnic (5.7%), Asian/Pacific Islander (2.2%), Native Alaskan/Native American (2.3%), and other (5.3%).
Measures
Participants in this study completed the Kauffman Teen Survey (KTS), originally developed by the Ewing Marion Kauffman Foundation. This survey was conducted regularly in Kansas City from 1984 until 2007; since 2007, the survey has been conducted by McLouth Research and Consulting and has since been renamed the McLouth Teen Survey. As it was called the KTS at the time of this study, the current article will refer to it as such. The KTS is a computer-adaptive online survey made up of 125 questions investigating a broad range of teen health-related behaviors. The majority of questions were closed-ended, formatted either as multiple choice or checklist items. Some questions allowed participants to enter information in an open-ended manner.
To determine prevalence of NSSI, each participant was asked whether or not they had ever physically hurt themselves on purpose, using the question: “Students have to deal with a lot of stress. When you have had problems to deal with, have you ever physically hurt yourself on purpose?” Participants could choose from one of four options of NSSI frequency: “never did this,”“did this only once,”“did this a few times,” or “frequently did this.” Any participants who indicated that they had ever hurt themselves on purpose were presented with follow-up questions related to self-injury. The first follow-up question asked: “You indicated that you have physically hurt yourself on purpose before. When you did this, did you choose to hurt yourself because you wanted to die?” Participants were able to choose from “no, never,”“yes, a few times,” or “yes, always.” This question was used to ensure that participants in the NSSI group truly had injured themselves with nonsuicidal intent.
Methods of NSSI were assessed by asking participants to: “Check any of the ways that you have hurt yourself on purpose without wanting to die (this is sometimes called “self-injury”).” This statement was followed by a list of common methods of NSSI: “Cut your wrists, arms, or other areas of your body”; “Burned yourself”; “Scratched yourself, to the extent that scarring or bleeding occurred”; “Banged your head against something, to the extent that caused a bruise to appear”; “Punched yourself, to the extent that you caused a bruise to appear”; and “Other.” The total number of endorsed methods was calculated for each participant.
Participants were also asked: “What parts of your body have you hurt?” Participants could then select one or more of the following options: “Arms,”“Legs,”“Stomach,”“Chest,”“Genitals,”“Face,” and “Other.” As with methods of NSSI, the total number of endorsed locations was calculated.
Procedure
Before data collection began, parents of all children in grades 6 through 12 at all participating schools were informed about the nature of the survey through a letter sent home with the child. Parents were able to refuse consent; a total of six students withdrew for this reason. This survey has been conducted regularly within these school boards, so most students were familiar with the procedure. Students were informed that no individual results would be examined.
Students were then asked to complete a questionnaire exploring issues related to teen health. The students completed the survey on classroom computers during class time. Each student completed the survey independently on their own computer; no discussion was permitted for the duration of the survey and dividers between desks were used. Students were given between 40 and 115 minutes to complete the survey.
Results
Prevalence
Among the full sample of 7,126 adolescents (3,623 females; 3,503 males), 32.1% of females and 16.6% of males reported having engaged in NSSI at least once in their lifetime. A chi-square analysis showed that females reported a significantly higher rate of NSSI than did their male counterparts, χ2(1) = 231.93, p = .000, Φ = .094. Phi (Φ) was used as a measure of effect size for all 2 × 2 chi-squares, and Cramer’s V was used for larger chi-squares. Additionally, Cohen’s d was used to calculate effect size for all t tests.
After analyzing NSSI prevalence, further investigations were conducted into NSSI frequency within the NSSI group (n = 1,744). Because significant gender differences had already been found between those who had ever engaged in NSSI and those who had not, analyses were only conducted for the frequency levels of “did this only once,”“did this a few times,” and “frequently did this.” Overall, a significant difference between males and females in frequency was found, χ2(2) = 6.26, p = .044, V = .060; these results are summarized in Table 1. Individual chi-squares for each level of frequency were then carried out, revealing that only the difference for “did this frequently” was significant, χ2(1) = 6.11, p < .05, Φ = .059. Additionally, there was no difference between females (M = 21.56; SD = 53.094) and males (M = 19.47; SD = 52.348), t(1,015) = .545, p = n.s, d = .040, on the number of times engaging in NSSI in one’s lifetime.
| Response | Female | Male | χ 2 | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Did this only once | 332 | 28.5 | 172 | 29.6 | .21 |
| Did this a few times | 563 | 48.4 | 305 | 52.5 | 2.59 |
| Frequently did this | 268 | 23.0 | 104 | 17.9 | 6.11* |
| Total | 1,163 | 100 | 581 | 100 | |
- *p < .01
Methods of NSSI
Analyses focusing on the characteristics of NSSI included only the participants NSSI group (n = 1,744). For the question tapping methods of NSSI, 269 females and 302 males were missing data; therefore, the sample for these analyses was made up of the remaining 1,173 participants. As outlined in Table 2, results indicated that females were significantly more likely than males to report cutting, χ2(1) = 55.34, p = .000, Φ = .217, and scratching, χ2(1) = 18.49, p = .000, Φ = .126. Males were more likely than females to endorse burning, χ2(1) = 7.91, p = .006, Φ = −.082; banging head, χ2(1) = 23.93, p = .000 Φ = −.143; and punching, χ2(1) = 19.60, p = .000, Φ = −.129. A t test was conducted to investigate gender differences in total number of methods reported; females (M = 2.16; SD = 1.29) and males (M = 2.18; SD = 1.41) did not differ significantly, t(1,171) = .238, p = n.s., d = .015.
| Response | Female | Male | χ 2 | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Cutting | 702 | 78.6 | 156 | 55.9 | 55.34* |
| Scratching | 465 | 52.0 | 104 | 37.3 | 18.49* |
| Burning | 190 | 21.3 | 82 | 29.4 | 7.91* |
| Banging head | 207 | 23.2 | 106 | 38.0 | 23.93* |
| Punching | 210 | 23.5 | 103 | 36.9 | 19.60* |
| Other | 155 | 17.3 | 57 | 20.4 | 1.373 |
| Total NSSI group | 894 | 100 | 279 | 100 | |
- *p < .05
Location of NSSI
As with the analyses for method of NSSI, 269 females and 302 males within the NSSI group were missing data for location. Therefore, the following analyses of NSSI location included a sample of 1,173 participants. Female adolescents were significantly more likely than their male counterparts to injure the arms, χ2(1) = 11.76, p = .001, Φ = .100, and legs, χ2(1) = 25.98, p = .000, Φ = .149. On the other hand, male adolescents were more likely to injure their chest, χ2(1) = 37.03, p = .000, Φ = −.178; genitals, χ2(1) = 14.05, p = .000, Φ = −.109; and face, χ2(1) = 36.05, p = .000, Φ = −.175. There was no significant gender difference for stomach, χ2(1) = .013, p = n.s., Φ = .003, nor for the variable “other,”χ2(1) = 3.162, p = n.s., Φ = −.052. A t test was also conducted to examine gender and total number of locations. There were no differences for number of locations endorsed between females (M = 1.84; SD = 1.03) and males (M = 1.91; SD = 1.36), t(381.9) = .834, p = n.s., d = .061.The results regarding gender differences in location of NSSI are outlined in Table 3.
| Response | Female | Male | χ 2 | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Arms | 764 | 85.5 | 214 | 76.7 | 11.76* |
| Legs | 424 | 47.4 | 84 | 30.1 | 25.98* |
| Chest | 56 | 6.3 | 51 | 18.3 | 37.03* |
| Genitals | 7 | 0.8 | 11 | 3.9 | 14.05* |
| Face | 96 | 10.7 | 70 | 25.1 | 36.05* |
| Stomach | 179 | 20.0 | 55 | 19.7 | .01 |
| Other | 116 | 13.0 | 48 | 17.2 | 3.16 |
| Total NSSI group | 894 | 100 | 279 | 100 | |
- * p < .05
Discussion
The purpose of this study was to contribute to the knowledge of gender differences in prevalence, methods, and locations of NSSI among adolescents, as empirical knowledge about gender differences in these domains is scarce. The findings of this study are consistent with past literature on NSSI prevalence within this age group, and add to the body of literature on NSSI methods and locations.
Prevalence
With regard to the first objective, our results supported the hypothesis that female adolescents reported significantly more NSSI than their male counterparts. This gender difference is consistent with previous research in adolescent samples. For frequency, significant gender differences were also found; females reported more frequent use of NSSI than males. It is important to note that within the NSSI group, nearly one quarter of self-injuring females indicated that they frequently engaged in this behavior. This suggests that not only are adolescent females more likely than their male counterparts to engage in NSSI, but that they are also likely to be engaging in the behavior regularly.
These results are consistent with the original hypothesis that females would report more NSSI than males. The findings support previous literature on NSSI in adolescent samples, which has found that females in this age group report higher prevalence rates than their male counterparts (e.g., Laye-Gindhu & Schonert-Reichl, 2005; Muehlenkamp & Gutierrez, 2007; Nixon et al., 2008). In addition to supporting the existing literature on prevalence, the current study contributes to knowledge regarding the frequency of NSSI. Although much is known about prevalence rates, there is much less information in the literature on gender differences in frequency of the behavior.
Although it would appear that more female adolescents engage in NSSI than do their male peers, there is an alternative explanation for this pattern. Research has indicated that male adolescents are less likely than their female peers to self-report NSSI (Heath, Schaub, Holly, & Nixon, 2008). Research in this field has improved the measures used to investigate NSSI; however, it is possible that a large enough group of self-injuring male adolescents did not report their self-injury on the survey.
Curiously, despite females indicating that they engage in NSSI more frequently than males, there was no gender difference for lifetime frequency; females and males had similar means for the open-ended question: “How many times have you hurt yourself?” This could be due to a number of different reasons. For example, participants were asked to distinguish between whether or not they had engaged in the behavior a few times or frequently. This is a subjective decision, and it is possible that different participants defined these terms differently. Alternatively, when asked to report their lifetime frequency of NSSI, participants were asked to recall and report a specific number of NSSI episodes, rather than choosing between different categories. Additionally, some participants’ responses may not have been entirely accurate. Some answers for this number were very high, with responses ranging from 0 to 1,000. However, participants were not excluded based on the perception of lifetime frequency being very high. Finally, the pattern of responses may be affected by how the questions were constructed; no time frame was given to more accurately assess frequency. Future research should consider asking participants how often they had injured themselves in a certain time period, such as the last month. Taking these considerations together, it is possible that this variable was not defined thoroughly enough for the participants. Future research is needed to determine whether there is in fact a difference in the frequency of NSSI between females and males, or whether this finding is an artifact of how individuals report their own level of frequency.
Methods of NSSI
It was hypothesized that cutting, scratching, and hitting oneself would be the most frequently endorsed method of NSSI, consistent with previous literature (Laye-Gindhu & Schonert-Reichl, 2005; Nixon et al., 2008; Yates et al., 2008). The results of the current study support the hypothesis; although the current study did not have a specific “hitting oneself” variable, the four most common methods were indeed cutting, scratching, punching self, and banging one’s head.
Yates et al. (2008) found that there were overall effects of gender on methods of NSSI, but did not report any specific differences. Other researchers have investigated methods of NSSI, but only within young adult samples (Andover, Primack, Gibb, & Pepper, 2010; Whitlock et al., 2006). This study is the first to extend an investigation into gender and specific methods within an adolescent sample. In the current study, gender differences were observed in many of the variables relating to method of NSSI. Females were significantly more likely than males to cut and scratch themselves; on the other hand, males were more likely than females to burn themselves, bang their heads, and punch themselves. The present results are the first to give a detailed observation of the gender differences in methods of NSSI among adolescents.
The most evident difference between the methods most reported by females and those reported by males is the presence of blood. Females tend to use methods that involve bleeding (cutting and scratching), while males tend to use methods that do not result in bleeding (burning, banging head, and punching). Emerging research into the importance of seeing blood during an episode of NSSI suggests that for some people, seeing blood is an important element of NSSI; however, there are no gender differences between those for whom blood is important and those for whom it is unimportant (Glenn & Klonsky, 2010). The study of the importance of blood included only those individuals who used cutting as a method of self-injury; therefore, the results cannot be generalized to all individuals who self-injure. Additionally, the sample consisted of young adults, so the results may not be representative of adolescents. Future research is needed to determine whether gender differences in methods of NSSI in adolescence are related to the importance of seeing blood. Despite differences in individual methods, there was no gender difference on number of methods endorsed. The finding that females and males endorsed an average of slightly more than two methods indicates that adolescents of both genders who self-injure are likely to have tried multiple methods of NSSI.
Location of NSSI
Results for the individual locations of NSSI indicated specific gender differences, and the hypothesis that females would be more likely than males to injure the legs was supported; however, additional differences were also found. Females were significantly more likely than males to injure their arms and legs; males were more likely than females to injure the chest, genitals, and face. Although this is consistent with the literature on gender differences in location of NSSI, such literature is very limited. In fact, only one study to date has investigated gender differences in location of NSSI in a community sample, and that study focused on young adults (Whitlock et al., 2006). The current research is the first to extend these findings to adolescents. It also adds to the specific locations that yielded gender differences; Whitlock and colleagues only found differences in the arms and legs, but the current study found additional differences in the chest, genitals, and face. Furthermore, the current findings can be integrated with the results found by Whitlock et al., as they share a common element: males appear to be more likely than females to injure areas that are more visible.
The reasons behind this pattern are unclear. Nock (2008) suggests that in some cases, individuals may engage in specific types of NSSI to communicate certain messages to others. Specifically, those who may be afraid of victimization of others may engage in NSSI that can be explained or interpreted as a “battle scar,” a physical signal that implies resilience. It is possible to interpret injuries to the chest or face as this type of injury; if a male is punching or hitting himself in the chest or face, the injuries may appear to be caused by a physical altercation with others. In this case, he may avoid victimization from his peers for his NSSI. This may be more likely behavior in males than in females as a result of social display rules regarding emotions (Polce-Lynch, Myers, Kilmartin, Forssman-Falck, & Kliewer, 1998). This interpretation, however, does not explain the fact that males are also more likely than females to injure the genitals. Additionally, this interpretation is merely hypothetical; without a greater understanding of why females and males choose the locations that they do, it is impossible to draw fully accurate conclusions about the observed gender patterns in NSSI location.
More importantly, these results highlight a concerning pattern; males appear to be more likely than females to injure more sensitive areas of the body. Injuries to the face, chest, or genitals may have more severe consequences than those to the arms and legs. For example, injuries to the face could indicate high levels of psychological and social distress (Walsh, 2006). Injuring the genitals can also have symbolic meaning that may indicate extreme negative emotions related to the sexual domain; it could also suggest that the individual may be experiencing an elevated threshold for pain (Walsh, 2006). This is particularly concerning, considering that high pain threshold and NSSI has been linked to suicide attempts (Nock et al., 2006). This could put adolescent males who self-injure at a greater medical risk to themselves than their female peers. Additionally, these areas are not considered typical areas to injure the body; research that excludes these body locations may inadvertently exclude males who are engaging in NSSI.
Limitations
One limitation in this study was that the measure used did not ask participants about the reasons behind their choices of methods or locations. As such, it is impossible to determine exactly why the observed gender differences exist. Future research should ask participants about these reasons, to better understand why females and males differ in their NSSI behavior. Another limitation is that the use of a self-report measure may have affected the validity of the results in two ways. First, although questions were written in clear and simple language, it is possible that some students were unable to comprehend the questions because of difficulties with reading. Second, despite the measures taken to ensure participant privacy and confidentiality, self-report studies of NSSI may be influenced by a social desirability bias (Nock & Cha, 2009). This may affect males in particular; if there is an established social perception that NSSI is a problem that primarily affects females, males may be less likely to feel comfortable reporting their engagement in NSSI.
Future Directions
A review of the past research on NSSI suggests that age group may have an influence on the characteristics and gender differences with regard to the behavior. The current study was consistent with the observed trend and added new information to the existing literature. A follow-up study of a young adult sample would be a logical next step, as young adults have been described as the “highest risk group for engaging in NSSI” (Rodham & Hawton, 2009, p. 46).
The current study should be used to inform future approaches to NSSI research; specifically, the results of the current study highlight the importance of recognizing diverse methods and possible locations of NSSI. For example, a study that asks participants solely about cutting and scratching is inherently likely to find a gender difference in prevalence because females consistently use these methods more than males. Some previous studies have presented limited definitions of NSSI to their participants, which may have an effect on the validity of the findings. Finally, many of the interpretations of the results of the current study are based on research conducted on emotion regulation difficulties. Future studies should be conducted to determine the reasons adolescents might give as to why they have chosen a particular method or location when engaging in NSSI.
Implications
The findings from this research have a number of implications for both researchers working in the field of NSSI as well as for clinicians working with adolescents. Research has shown that there is a correlation between number of methods used and suicide attempts (Nock et al., 2006), and different locations can increase the risk of negative medical and social outcomes and can signify concerning levels of psychological distress (Walsh, 2006). Therefore, a greater understanding of these features of NSSI is crucial to understanding the behavior and helping clients.
For researchers, it is important that future studies of NSSI continue to include a variety of different methods and locations. Excluding such behaviors as burning or banging one’s head could exclude male adolescents who engage in NSSI; likewise, excluding injuries to the face or genitals may also exclude males. These exclusions may therefore lead to incomplete assessment of NSSI and inaccurate representation of the behavior within the research literature, particularly with regard to furthering the understanding of gender and NSSI.
Clinicians should be aware that adolescent clients who are engaging in NSSI may be using more than one method to self-injure, or injuring themselves in more than one location. This is true for both female and male adolescents and is particularly important considering the correlations between number of methods and suicide attempts, and between location and different signs of psychological and social distress (Nock et al., 2006; Walsh, 2006). It also is important for clinicians to recognize the variety of methods being used by female and male adolescents alike; if clinicians conceptualize NSSI as simply “cutting,” they may miss the behavior in clients who burn, scratch, or punch themselves. Finally, males who engage in NSSI may be injuring themselves in severe locations, such as the chest, face, or genitals. This should be kept in mind when assessing NSSI; injuries in severe locations may require more immediate referral to medical services.
Conclusions
NSSI is a concerning behavior among adolescents, but the current study shows that it is a behavior that can present very differently depending on individual characteristics, such as gender. A female adolescent may be more likely to be cutting her arms and legs more frequently than another male adolescent, who injures himself by burning himself on the chest or genitals. Although these two behaviors look different, they are both used by the individuals to cope with stress, and they both involve intentional tissue damage without the desire to kill oneself. Although it is a distinct behavior from suicide, NSSI can be a risk factor for suicide attempt (Sher & Stanley, 2009). Even individuals who have no suicidal intent and injure themselves in low-lethality ways can accidentally injure themselves more seriously than they intend to (Newman, 2009). It is therefore important for clinicians not only to treat clients when they present with NSSI, but to also be aware of the different features of NSSI in order to better understand the behavior, its traits, and its functions.
The current study revealed a pattern whereby female and male adolescent self-injurers differed across a number of variables. Females reported a higher prevalence of NSSI than males. Females were more likely than males to report cutting and scratching, and injuring the arms and legs. Males, on the other hand, were more likely than females to punch themselves and to injure the chest and face. These results will not only help researchers and clinicians to better understand the different forms that NSSI can take, but will also support the conceptualization of NSSI as an emotion regulation strategy, since the findings seem to be consistent with past research on difficulties with managing emotion. A better understanding of the features of, and reasons for, NSSI will allow psychologists and other professionals to be better prepared to help these adolescents who are struggling with their ability to cope and manage overwhelming negative emotions (Heath & Nixon, 2008; Walsh, 2006).
