Non-suicidal self-injury in Mexican young adults: Prevalence, associations with suicidal behavior and psychiatric disorders, and DSM-5 proposed diagnostic criteria

Highlights

• Lifetime and 12-month prevalence of NSSI was 18.6% and 3.2% respectively.

• One-third were recurrent self-injurers, engaging in the behavior five or more times.

• Most do not meet full DSM-5 proposed criteria due to a lack of reported impairment.

• Less than 40% of those who self-injure report impairment.

• NSSI was associated with psychiatric disorders and suicidal behavior.

Abstract

Background

Non-suicidal self-injury (NSSI) may lead to scarring, infection, accidental death and psychological distress. Little is known about NSSI in the general population of young adults in developing countries like Mexico. The current study examined the prevalence of any NSSI and each type of NSSI, the prevalence of meeting DSM-5 proposed criteria, and finally the association of NSSI with socio-demographic variables, suicidal behavior and psychiatric disorders.

Methods

This study was conducted in a community sample of 1071 young adults between 19 and 26 years of age residents of Mexico City.

Results

The lifetime prevalence of NSSI was 18.56% with females having 87% greater odds. The 12-month prevalence was 3.19%. Only 0.22% of the total sample and 6.96% of those that self-injured in the past 12 months met full criteria proposed by DSM-5, in part due to the lack of reported impairment; 39.99% of those that self-injured reported impairment. Suicidal behavior commonly co-occurred with NSSI. All lifetime anxiety, mood, disruptive behavior and substance use disorders were associated with greater risk for lifetime NSSI whereas only 12-month depression and substance use disorder was associated with greater risk of 12-month NSSI.

Limitations

The cross-sectional nature of the study precludes conclusions of causality and directionality and the study excluded institutionalized and homeless young adults.

Conclusions

NSSI is a concerning problem in young adults from Mexico City due to the important associations with all types of psychiatric disorders and suicidal behavior. Because many who self-injure do not perceive impairment, they are unlikely to seek treatment.

Introduction

Non-suicidal self-injury (NSSI), defined as deliberate and direct bodily harm without suicidal intent (Nock, 2010), includes behaviors such as cutting, burning, biting and scratching the skin for purposes not socially sanctioned. NSSI is eliciting increased attention and concern around the world. According to a meta-analysis of prevalence rates of NSSI from various developed countries and from school, university, and community-based samples the pooled lifetime prevalence rate was 17.2% (8.0–26.3) for adolescents, 13.4% (4.5–22.3) for young adults, and 5.5% (1.7–16.3) for adults (Swannell et al., 2014). Whitlock et al. (2006) reported that 5% initiate these behaviors in childhood (<10 years of age), 25% in early adolescence (10–14 years of age), 27% in middle adolescence (15–16 years of age), 34% in late adolescence (17–20 years of age), 5% in early adulthood (21–24 years of age) and 4% thereafter (>24 years of age).

NSSI is generally believed to express an underlying deficit in emotion regulation and coping skills (Chapman et al., 2006, Nock, 2009, Selby et al., 2015). Because of the transdiagnostic nature of emotion dysregulation and impaired coping skills in those with psychiatric disorders, it is therefore not surprising that the research to date indicates that NSSI is associated with a range of psychiatric disorders, from mood disorders, like major depressive disorder and bipolar disorder (Bentley et al., 2015, In-Albon et al., 2013, Kim et al., 2015), anxiety disorders such as a social phobia and posttraumatic stress disorder (Bentley et al., 2015, Dixon-Gordon et al., 2014, In-Albon et al., 2013, Polanco-Roman et al., 2014, Sacks et al., 2008), substance use disorders (MacLaren and Best, 2010), eating disorders (MacLaren and Best, 2010), behavioral disorders like ADHD, oppositional defiant disorder and conduct disorder (Cerutti et al., 2011, Hinshaw et al., 2012), and personality disorders, particularly borderline personality disorder of which NSSI is one symptom (Jenkins et al., 2015, Klonsky et al., 2003, Turner et al., 2015). NSSI within an autism spectrum disorder or psychosis is considered a different phenomenon (Nock and Favazza, 2009). Understanding the high co-occurrence rates of NSSI and psychiatric disorders is important for nosology to determine whether NSSI is an independent diagnostic entity or a transdiagnostic symptom.

While some studies have not distinguished between suicidal and non-suicidal self-injury or conceive of both as part of a continuum of suicidal behavior (Orlando et al., 2015), a large body of research has emerged in which significant differences in the etiology, functions, methods and course of NSSI and suicidal behaviors support the idea that these are two separate but often co-occurring phenomenon (Franklin and Nock, 2016, Muehlenkamp, 2005, Selby et al., 2015). Nock et al. (2006) found that as much as 70% of adolescents with NSSI have a lifetime history of a suicide attempt and NSSI has been found to be a strong predictor of subsequent suicidal behavior (Guan et al., 2012, Scott et al., 2015). Because of this high co-occurrence of NSSI and suicidal behavior, a better understanding of NSSI could also provide useful information for suicide prevention.

There is contradictory evidence as to whether there are sex differences in NSSI with some studies reporting a greater prevalence in females (e.g., Barrocas et al., 2012; Nixon et al., 2008) and others finding no sex differences (e.g., Plener et al., 2009). A recent meta-analysis on the question of sex differences in prevalence found that females tended to have higher NSSI rates than males, especially in clinical samples rather than community or university samples (Bresin and Schoenleber, 2015). This meta-analysis also found sex differences in methods of NSSI such that females were more likely to engage in certain methods such as cutting and scratching than males but were no more likely to engage in others like punching. With regards to other sociodemographic correlates, one study found that those reporting difficulties to afford basic necessities were more likely to report NSSI than those that didn´t (Nixon et al., 2008). However, another study suggests that NSSI is also prevalent among youth of high economic status (Yates et al., 2008). Arbuthnott and Lewis (2015) reviewed the literature on the parental characteristics of youth that self-injure finding mixed evidence for the association of NSSI with low parental education, parental unemployment, lower family income, non-intact families, with somewhat more consistent evidence for the increased risk among those with financial problems.

A further emerging literature regarding NSSI with conflicting evidence is the nosology of NSSI. Previously in the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV; American Psychiatric Association, 1994), NSSI was conceptualized as a symptom of borderline personality disorder (BPD). However, NSSI has been found independent of BPD, which is why it was included in the current DSM-5 as a condition for further study (American Psychiatric Association, 2013). One concern with the suggested criteria for NSSI as a disorder surfaced from the DSM-5 field trials, which struggled to obtain sufficient samples sizes for NSSI disorder to generate accurate kappa estimates in two of three sites, and in the one site that did have adequate sample size, failed to show acceptable interrater reliability using clinician-administered diagnostic interviews (Regier et al., 2013). Among varying recommendations for further research to accrue sufficient evidence for the validity of NSSI as a disorder, are to conduct studies on the way the disorder may present across different ethnicities or cultures (Selby et al., 2015).

Chu et al. (2010) have proposed a cultural theory and model of suicide. While specific to suicide and not NSSI, the three proposed principals that affect cultural differences in suicide are likely to be relevant for NSSI as well. That is, culture likely affects idioms of distress, in other words how NSSI is expressed (e.g., method), the types of stressors that lead to NSSI, and the cultural meaning associated with both the stressors and NSSI itself (e.g., cultural sanctions or tolerance of NSSI, threshold for tolerance of psychological pain).

While most of the research on NSSI has come from the United States, Europe and to a lesser extent Asia (Muehlenkamp et al., 2012a, Muehlenkamp et al., 2012b), in Latin American countries in general, and in Mexico in particular, the literature on NSSI is almost nonexistent (Thyssen and van Camp, 2014). Only recently a handful of studies have begun to address this topic. Albores-Gallo et al. (2014) conducted a study on NSSI in a community sample of youth age 11–17 in Mexico City. The prevalence rate in this sample was 5.6% using the DSM-5 proposed criteria and 17.1% using a broad question of whether the participants had ever engaged in self-injury with no suicidal intent. In another study also conducted in Mexico City with students aged 12–17, 27.9% of the sample reported self-injury (Mayer Villa et al., 2016). Finally, among university students in Mexico City, Castro-Silva et al. (in press) estimated a lifetime prevalence of 45.1% and a 12-month prevalence of 6.1%. This last study suggests that while NSSI behaviors generally begin in adolescence they continue to persist into early adulthood. However, because this study was done in university students, and the majority of young adults in Mexico are not enrolled in a university, an understanding of NSSI in young adults from the general population is important. Following Chu et al.'s proposed cultural model, we might expect in the Mexico City context, differences in the types of methods used due to culturally specific idioms of distress, differences in risk factors due to the cultural meanings assigned to these factors, and differences in prevalence due to differences in cultural sanctions regarding NSSI. If the proposed DSM-5 criteria are meant to describe a universal diagnostic entity then many of those who self-injure in Mexico should meet those same diagnostic criteria.

As a way to address this lack of literature in the field of NSSI in developing countries, this study reports the distribution of NSSI and its correlates in young adults from Mexico City. Specifically, we report the prevalence of any NSSI and each type of NSSI in the total sample and by sex, the prevalence of meeting each of the DSM-5 proposed criteria and meeting full DSM-5 proposed criteria and finally the association of NSSI with socio-demographic variables, suicidal behavior and psychiatric disorders.