Keywords

Introduction

Self-compassion, or a way of self-relating that has been described as kindness and compassion turned inward (Neff, 2003), has been associated with better overall emotional well-being in adults in many studies and confirmed in meta-analyses (e.g., MacBeth & Gumley, 2012; Zessin et al., 2015). Furthermore, studies demonstrate a lower physiologic stress response among those who have higher trait self-compassion (Breines et al., 2014; Breines et al., 2015; Svendsen et al., 2016) or who have been primed with self-compassion inductions (Arch et al., 2014); these are discussed in-depth in another chapter in this volume. Overall, self-compassion has been associated with increased happiness, optimism, connectedness to others, and curiosity and inversely associated with mental health disorders, including anxiety and depression in adult clinical and community samples (Neff, 2009).

Although research among adolescent populations is newer and therefore not as extensive as that of adults, studies indicate that self-compassion has similar beneficial effects on the mental health and well-being of adolescents (Marsh et al., 2018). In this chapter, we will first describe why self-compassion is critically important for adolescents and follow with an overview of the adolescent research to date. This chapter does not provide a comprehensive or systematic review but rather includes a diverse sample of the research studies that have been published and suggests ways in which the findings can support healthy adolescent functioning.

The Need for Self-Compassion in Adolescence

Adolescence is a crucial and often tumultuous developmental period, during which the main task of teens is to explore and develop their identity through negotiating social roles, adapting to changing peer and family relationships, and exploring career and vocational goals (Erikson, 1963; Marcia, 1980). While this exploration of identity is taking place, the adolescent brain is going through significant restructuring; myelination of neurons allows the brain’s processing speed to increase dramatically, and new neuronal pathways are established while unused pathways are pruned away (Giedd, 2008), allowing for more efficient information processing. Differences in the rate of maturation of brain systems, such as the prefrontal cortex (or attentional control system) and the amygdala (within the limbic system), result in adolescents being more sensitive to emotional highs and lows without always having the advantage of the ability to think clearly about the consequences of their actions (Keating, 2004). This aspect of the adolescent brain has been commonly referred to as “being in a car with a sensitive gas pedal and bad brakes” (Steinberg, 2005). These brain changes are reflected in adolescents’ ability to think more abstractly; for the first time, adolescents are able to think about their thinking or metacognate. Metacognition, although advantageous in many ways, can also promote greater self-consciousness which can then lead to greater self-criticism. At the same time, changes in sex hormones, in particular, estrogen and testosterone levels in females (Andersen et al., 2022; Copeland et al., 2019), result in increases in depression and anxiety.

In addition to the normative ongoing restructuring of the brain and other physiological changes which take place during this period (Giedd, 2008), this developmental stage may be complicated by environmental challenges that may have long-term deleterious effects on adult health. These challenges often include social stressors such as transitions to different schools and peer and academic pressures (Forbes & Dahl, 2010; Steinberg & Morris, 2001) and may also include in-person and cyberbullying (Patchin & Hinduja, 2013), exposure to violence (Lambert et al., 2005), structural racism (Lambert et al., 2009), physical safety concerns, misogyny (Dosil et al., 2020), inadequate academic resources (Crede et al., 2015), food insecurity (Shanafelt et al., 2016), homelessness (Wang et al., 2019), and lack of future academic and economic opportunities (Johnson et al., 2014). Thus, the complex interaction of various physiological changes taking place at this developmental period, combined with the onslaught of environmental challenges, sets the stage for adolescents to be highly self-critical, anxious, and depressed, particularly for females.

Although adolescence has always been recognized as a time of increased vulnerability to mental health problems, over the last decade, mental health disorders among adolescents have skyrocketed. For example, between 2010 and 2015, high symptoms of mental illness increased by 33% among a US sample of adolescents aged 13–18, and suicides increased by 31%; these changes were mostly driven by females and occurred across all race and ethnicities sampled (Twenge et al., 2018). Mental health disorders in adolescence, particularly when unrecognized or untreated, can establish a maladaptive developmental trajectory and are predictive of substance abuse, poor academic achievement, and mental and physical health disorders in adulthood (Patel et al., 2007). According to the 2018 American Psychological Association survey (American Psychological Association, 2018), US adolescents reported higher stress than any other age group and were least likely to report that their mental health was either excellent or very good. Notably, 75% of teen respondents cited gun violence in general, mass shootings, and school shootings as high on their list of stressors.

Certain subpopulations of adolescents are at higher risk for mental health disorders when compared to their peers and are important groups to support through intervention. For example, studies have shown gender differences exist in how teens respond to stress, with females being at greater risk of developing and maintaining internalizing disorders such as depression and anxiety. Beginning in early adolescence, females report more depressive symptoms than males (Bennik et al., 2014; Nolen-Hoeksema, 2012; Sontag & Graber, 2010), and this global trend continues through adolescence until age 16–19 when females are twice as likely to be depressed as males (Salk et al., 2017). Also, teens who have preexisting health disorders that were diagnosed in childhood are faced with the added normative stressors of adolescence, coupled with the drive for independence and desire to self-direct their medical treatment. For instance, teens with chronic medical illnesses, (e.g., Type 1 diabetes mellitus, eating disorders, cancer) face the added challenges of transitioning to self-care and management of their medications and treatment regimen (Finlay-Jones et al., 2020). Similarly, adolescents with developmental delays may find their health challenges exacerbated by the social and academic stressors of adolescence (North et al., 2013).

Furthermore, a past history of physical, sexual, or emotional abuse is a risk factor for adolescent and adult mental health disorders, including anxiety, depression, and suicidality (Lindert et al., 2014; Lippard & Nemeroff, 2020; Sweeting et al., 2020; Turner et al., 2017). Childhood abuse and other adverse childhood experiences are associated with a higher risk of physical disease in adulthood and adolescence (Felitti et al., 1998; Herrenkohl et al., 2013; Lippard & Nemeroff, 2020). When adverse childhood experiences occur over a prolonged period, the toxic stress which results can cause changes in the developing brain that has lifelong impacts (Kuo et al., 2012; Shonkoff & Garner, 2012). For example, childhood sexual abuse has been linked to both internalizing and externalizing maladaptive behaviors, including depression, and worldwide, prevalence for childhood sexual abuse is 18% for females and 8% for males (Kuehner, 2017). Finding ways to combat the effects of early life diversity is instrumental to disrupting its long-term sequelae.

The stressors faced by marginalized populations, such as Black, Indigenous, and People of Color (BIPOC), exacerbate the challenges of adolescence, and many BIPOC adolescents experience greater psychological, academic, and behavioral challenges than their White counterparts both during and after adolescence (Assari et al., 2017; Hughes et al., 2016; Huynh & Fuligni, 2010; Paradies et al., 2015; Umaña-Taylor & Updegraff, 2007). For example, a 2015 systematic review and meta-analysis found a significant association between experiencing racism and negative mental health outcomes, including depression, stress, and anxiety, particularly for Asian-Americans and Latinx-Americans (Paradies et al., 2015). Additionally, youth who have recently emigrated may experience higher risk of depressive symptoms related to their immigration status and the perception of being regarded as “foreigners” (Davis et al., 2016; Lopez et al., 2016; Sirin et al., 2019). As one example, Latinx youth who recently emigrated to the United States experienced ethnic discrimination in the ninth grade and were more likely to develop depressive symptoms 6 months later (Davis et al., 2016).

Further, sexual/gender minority (SGM) adolescents experience worse mental health outcomes (i.e., greater depression, higher suicide ideation rates) than their sexual majority peers, which is likely a result of higher levels of bias-based bullying, victimization, and social stigma they experience (Gnan et al., 2019; Hatchel et al., 2019; Vigna et al., 2017). In fact, stigmatized SGM youth have two to three times the rates of depression, anxiety, self-injury, and suicide rates relative to their sexual gender majority peers (King et al., 2008; Zaza et al., 2016).

In addition, other subpopulations of adolescents which struggle with nonnormative stressors that exacerbate the normative stressors of adolescence include youth experiencing homelessness or undocumented status (Torres et al., 2018; Zapata et al., 2016), adolescents in the child welfare system (Tanaka et al., 2011), and adolescents who were maltreated or victimized as children (Játiva & Cerezo, 2014; Vettese et al., 2011). Further, youth with multiple marginalized identities, such as those listed above who also struggle with experiences of racial discrimination and are LGBTQIA+, may face an intersectional or compounding effect of stress on their mental health (Vigna et al., 2017).

Importantly, the current social climate, characterized by political instability, racial protests, the global COVID-19 pandemic, and the resulting disruption of school and social structures, has exacerbated many of these risk factors and has increased the stress level of adolescents. One recent study reported that compared to pre-pandemic measures, adolescents in a large pediatric primary care network in the Northeast United States had significant increases in depressive symptoms and suicidal thoughts, particularly among female, non-Hispanic Black, and non-Hispanic White adolescents during the pandemic (Mayne et al., 2021). Female adolescents in this study had a 34% increase in suicidal thoughts compared to before the COVID-19 pandemic, 1 year prior (Mayne et al., 2021). Another study found that the global prevalence of child and adolescent depression and anxiety has doubled in the first year of the COVID-19 pandemic, to current pooled estimate prevalence levels of 25% and 20.5%, respectively (Racine et al., 2021). In fact, Ellis et al. (2020) found that during the COVID-19 pandemic, 43% of adolescents were “very concerned” about the pandemic, as it related to their academic success and relationships with friends, and COVID-related stress was associated with greater loneliness and depression, particularly for those adolescents who spent more time on social media. Further, as connecting with peers and establishing one’s place in the social network are critical tasks in adolescent development, and in-person schools and peer interactions have been limited during the pandemic, social media use among adolescents increased during this time, with 48% spending more than 5 hours a day on social media since schools closed (Ellis et al., 2020). Along with the advantages of social contact, online social media use during the pandemic also may have increased unhealthy social media interactions, such as cyberbullying, increasing adolescent mental health challenges (Mayne et al., 2021).

Clearly, there is no magic pill that will alleviate these complex challenges that adolescents face today. However, building inner resources to contend with these stressors is one important strategy for helping to modulate adverse impacts. For example, establishing a sense of identity with a community, finding meaning behind struggles, and acknowledging one’s agency (Meyer, 2015) are all internal resources that can be strengthened and have been found to be advantageous in dealing with minority stressors (Meyer, 2015). They are also all aspects of self-compassion; a sense of common humanity emerges from identifying with a community, finding meaning behind struggles is part of growth toward self-kindness (and taught directly in an exercise in the Mindful Self-Compassion program), and acknowledging one’s agency and purpose is integral to fierce self-compassion.

As adolescence is a sensitive period for intervention and often a precarious period for emotional stability, providing teens with coping tools such as self-compassion can potentially shift their mental health and behavioral trajectory to a more positive and salubrious pathway. The research providing evidence for links between self-compassion and various dimensions of well-being among adolescents is discussed next.

Associations Between Self-Compassion and Well-Being Outcomes

The first empirical study on self-compassion and adolescence was published in 2010; in this study, Neff and McGehee explored the relationship between self-compassion and psychological resilience in US adolescents and young adults (Neff & McGehee, 2010). Since then, numerous studies have been published, and overall, research findings mirror that of adults. For example, similar to the meta-analysis on adults that shows a large inverse relationship (r = −0.54) between self-compassion and psychopathology, defined as stress, anxiety, and depression (MacBeth & Gumley, 2012), Marsh et al.’s (2018) meta-analysis of adolescent samples demonstrated a similar magnitude in the relationship between self-compassion and psychopathology (r = − 0.55). Supporting this, a systematic review of studies on self-compassion and depressive symptoms in adolescents found that across both longitudinal and cross-sectional studies, self-compassion was inversely related to depressive symptoms (Pullmer et al., 2019a).

Similar associations between self-compassion and various domains of emotional health have been reported. Specifically, Neff and McGehee (2010) found that self-compassion is positively correlated with social connectedness in adolescents; this is noteworthy as establishing healthy peer relationships and stable social networks is critical factor in emotional well-being during adolescence. Additionally, this study also reported positive associations between self-compassion and secure attachment and negative associations between self-compassion and preoccupied and fearful attachment, suggesting that self-compassion is established in an individual through safe and trusting early life relationships. As there has been a paucity of opportunities to develop and cultivate close peer relationships during the time of the pandemic, the ramifications of isolation during this period may influence the emotional well-being of adolescents that may not be apparent for some time.

In understanding the implications of self-compassion for adolescent well-being, it is important to clearly delineate the differences between self-compassion and self-esteem. Due to the many physiological, cognitive, and environmental changes taking place at this developmental stage, many adolescents, particularly females, struggle with low self-esteem; the chapter on self-esteem within this handbook discusses this at length. Although both self-compassion and self-esteem are ways of self-relating, self-esteem, defined as a global evaluation of one’s self-worth (Baumeister et al., 2003), differs from self-compassion in that it is acquired and maintained by comparing oneself with others. For example, social media provides a readily accessible way for adolescents to compare themselves to others through number of “likes”, “shares,” and filtered or curated images posted on social media platforms. Unfortunately, this comparison often has a negative impact on one’s self-esteem; in fact, 45% of adolescents aged 15–21 indicated that engaging in social media makes them feel judged, and 38% report that it makes them feel bad about themselves (American Psychological Association, 2018). Yet even if the comparison does not have a negative outcome or even results in a boost in one’s self-esteem, the consequence of comparing oneself with others leads to an emotional separation from others at a time in life when what is most needed is connection with others and a sense of belonging. For example, whether one considers oneself above average in a particular trait or below average (i.e., high self-esteem or low self-esteem), an outcome is that one sees oneself as apart from others, rather than feeling accepted and included, or as an integral part of a community. Further, establishing high self-esteem is dependent upon performance and achievement, and inevitably one will not always perform at their best or even fail, and thus one’s self-esteem is unstable. In contrast, self-compassion offers a way of self-relating that is stable over time. Through self-compassion, one can always be present to support oneself (Neff & Vonk, 2009).

The relationship between self-compassion and self-esteem was investigated further in a study which used cross-lagged autoregressive structural equation modeling to determine the temporal order of constructs, i.e., which construct predicted the other (Donald et al., 2018). Over 4 years, self-compassion and self-esteem levels of 2809 adolescents were assessed. Results indicated that self-esteem consistently predicted self-compassion levels, whereas self-compassion levels did not predict levels of self-esteem, indicating that among adolescents, it may be necessary to feel valued, worthy, and deserving to give oneself compassion. More research with adolescent populations is needed to replicate these findings.

The different roles of self-compassion and self-esteem as linked to adolescents’ psychological well-being are evidenced in a study by Marshall et al. (2015). In this longitudinal study of 2448 high school students in Australia, self-compassion had a buffering effect, protecting against the negative effects of low self-esteem. For grade 9 students who were high in self-compassion, having low self-esteem did not have an effect on their mental health; in grade 10, these students had no significant change in their mental health. However, grade 9 students who were both low in self-compassion and self-esteem had worse mental health when they were assessed in grade 10. Thus, having high self-compassion appeared to protect against the negative repercussions of having low self-esteem (Flett et al., 2003; Hewitt & Flett, 1991; Marshall et al., 2015).

Self-Compassion as a Buffer Against Stress and Trauma

Managing stressful circumstances in a healthy way is an essential skill that adolescents must learn to transition to becoming adults who lead fulfilling and satisfying lives. Being able to cope with stressful circumstances productively and successfully is therefore critical to healthy adolescent development. Among adolescents, self-compassion has been shown to be associated with increased adaptive coping following stressful events. For example, one study found that undergraduates (mean age = 18 years) in Japan who were higher in self-compassion reported 1 month later that they felt more in control of a recent stressful event and felt that it was less threatening to them compared to those who were lower in self-compassion (Chishima et al., 2018). Furthermore, self-compassion was negatively related to avoidance coping; those adolescents who were higher in self-compassion were less likely to engage in maladaptive coping strategies such as denying, disengaging, or distracting themselves from a problem (Chishima et al., 2018).

Self-compassion may also play a buffering role in protecting adolescents from negative outcomes resulting from traumatic early life experiences and chronic stress. For example, among adolescents who had experienced abuse and neglect as children and were part of the Canadian child welfare court system, those who reported greater self-compassion were less likely to experience psychological distress, problem alcohol use, or report a suicide attempt than those with greater self-compassion, even when accounting for early life maltreatment (Tanaka et al., 2011). Among another population of adolescents experiencing chronic stress, Prentice et al. (2021) investigated adolescents and young adults who struggle with chronic physical health problems requiring ongoing medical care. In this study, self-compassion was positively associated with emotional well-being and negatively associated with mental distress. Difficulties in emotion regulation mediated the relationship between self-compassion and distress; this suggests that interventions addressing emotion regulation difficulties and promoting self-compassion may be particularly helpful for teens facing the challenges of chronic medical illness (Prentice et al., 2021).

Being more self-compassionate appears to be beneficial in managing not only long-term chronic stressful events, such as childhood abuse/neglect and health problems, but also acute stressful events. The weeklong Mount Carmel forest fire occurred in northern Israel in December 2010, forcing the evacuation of an educational residential youth village in which high school students lived, 50% of whom had come from homes where they experienced chronic stressors such as poverty, violence, and substance abuse and 20% of whom were orphans. Notably, within 1 month of the fire, 88% of adolescents reported that they feared for their lives when the fire was taking place. Among these youth who were already considered at-risk, both longitudinal and multilevel mediational analyses demonstrated that having greater self-compassion predicted less post-traumatic stress, panic symptoms, depressive symptoms, and suicidality 3 and 6 months later (Zeller et al., 2015).

The potential buffering effect of self-compassion described here is also supported by evidence from physiological responses to stress in an experimental protocol. The Trier Social Stress test is a well-established research protocol that has been used extensively to elicit a physiological stress response that can then be measured in a lab setting. Adolescents were exposed to the Trier Social Stress test in which they were asked to give a speech and perform math computations in front of two neutral-faced lab-coated adults. Adolescents who self-reported greater self-compassion had less of a change in their systolic blood pressure during the lab stressor than those with lower self-compassion (Bluth et al., 2016b), indicating that self-compassionate adolescents may get less activated in stressful situations than adolescents who are less self-compassionate. However, more research using various physiological measures of evaluating stress is needed to substantiate this finding.

In addition to buffering the effects of external stressors (Flett et al., 2003; Hewitt & Flett, 1991; Marshall et al., 2015), self-compassion also appears to protect against the negative impacts of internal characteristics that increase vulnerability to adverse outcomes, such as maladaptive perfectionism. Maladaptive perfectionism is defined by having high personal standards that involve high self-criticism, being overly worried about making mistakes, and fear of being negatively evaluated by others. Many adolescents struggle with maladaptive perfectionism in various domains of their lives, including physical appearance, academics, and their interactions and relationships with peers. This type of perfectionism has been linked to depression (Flett et al., 2003; Hewitt & Flett, 1991) and is considered transdiagnostic in that it affects many aspects of health, including anxiety and eating disorders (Egan et al., 2011). For example, maladaptive perfectionism related to academic achievement among adolescents has been linked to anxiety and depression (Einstein et al., 2000) and negative affect after taking a test (Flett et al., 2009). Ferrari et al. (2018) demonstrated that by treating self-critical thoughts as passing events and as an opportunity to treat oneself with kindness, one is able to interrupt and subvert the pathway to depression. In this study, 541 adolescents in a high school setting completed questionnaires comprising measures of perfectionism, depressive symptoms, and self-compassion. As expected, the results showed a positive correlation between maladaptive perfectionism and depression and a significant moderating or buffering effect of self-compassion on the relationship between maladaptive perfectionism and depression, thus weakening this relationship (Ferrari et al., 2018). In other words, when self-compassion was high in adolescents, the link between maladaptive perfectionism and depression was not as strong, and adolescents who were more self-compassionate were less likely to become depressed.

Many adolescents and young adults struggle with maladaptive perfectionism concerning body image, the subjective assessment or judgment on one’s physical appearance. Ntoumanis et al. (2020) studied the association of upward appearance comparison (comparing oneself to another who is perceived as more attractive) on appearance evaluation (the sense of satisfaction with one’s own body image) in 396 Greek adolescents. This study investigated appearance-specific self-compassion, i.e., self-compassion that is specific to negative appearance-related thoughts, rather than global self-compassion. For these adolescents, appearance self-compassion moderated, or buffered, the effects of upward comparison on appearance evaluation; those adolescents with high self-compassion were less likely to negatively evaluate their appearance when comparing their image to someone who they considered more attractive (Ntoumanis et al., 2020).

Similarly, Rodgers et al. (2017) found that self-compassion moderated the relationship between perceived overweight status, appearance comparison, and appearance esteem in their study of 232 adolescents, aged 13 to 18 years. Appearance comparison, which is the process of comparing one’s physical appearance to others and engaging in mostly unfavorable comparisons, has been identified as a critical mechanism in the maintenance of body image concerns and mediates the relationship between weight status and body dissatisfaction. Perceived overweight status in adolescents is associated with body dissatisfaction and poor self-esteem. The two self-compassion components of common humanity and mindfulness moderated the association between perceived overweight status and appearance comparison among boys, but not among girls. In other words, for those boys who had higher levels of mindfulness and common humanity, the link between perceiving themselves as overweight had less of an effect on negatively evaluating their appearance (Rodgers et al., 2017).

Self-compassion has also been shown to be a moderator in the association between non-suicidal self-injury (NSSI) and depressive symptoms in adolescence (Xavier et al., 2016). NSSI, described as self-injury that is not associated with suicidal intent, is prevalent in adolescents, particularly females. In fact, one study reported that one in four females and one in ten males reported engaging in NSSI over the previous 12 months (Monto et al., 2018). Six hundred forty-three adolescents aged 12–18 years completed questionnaires containing measures of emotional health, risk-taking and self-harm, self-compassion, and daily peer hassles (i.e., conflicts with friends regarding beliefs, opinions, and choices). The results showed that those who reported higher depression scores and daily peer hassle scores were more likely to self-injure. Importantly, self-compassion had a significant modifying or buffering effect on the relationship between depression and NSSI; higher levels of self-compassion were associated with lower levels of NSSI, independent of the adolescents’ depression levels (Xavier et al., 2016).

As suicide is the second leading cause of death among adolescents aged 15–19 in the United States and the fourth leading cause of death in China, finding ways to obviate suicide attempts is of critical concern. In one study, self-compassion moderated the association between suicidal ideation and suicidal attempts in a cohort of 520 Chinese adolescents (Sun et al., 2020). Those adolescents with higher self-compassion (measured by high scores when combining the three positive components of self-compassion and low scores when combining the negative components of self-compassion) who had thoughts of suicide were less likely to attempt suicide over a 12-month interval, compared to those with low self-compassion (Sun et al., 2020).

Sex differences in the moderating role of self-compassion have also been found. Results of a cross-sectional study of 1057 adolescents (65% female, mean age 14.5 years) supported the inverse relationship between self-compassion and perceived stress, depression, and anxiety and found that self-compassion moderated the association between perceived stress and anxiety for males only (Lathren et al., 2019) but moderated the association between perceived stress and depression equally between sexes. These findings are consistent with an earlier study showing that self-compassion was a buffer against later onset of anxiety or depression following a traumatic event (Zeller et al., 2015).

Gender and Age as Moderators of Self-Compassion

Similar to findings from a meta-analysis with adults that reported that women had lower levels of self-compassion overall compared to men (Yarnell et al., 2015), most evidence in adolescent studies have also found levels of self-compassion are greater in males (Bluth & Blanton, 2015; Bluth et al., 2016b; Castilho et al., 2017; Pullmer et al., 2019a). However, one study found no difference between male and female adolescents (Neff & McGehee, 2010). Bluth and Blanton (2015) examined differences in self-compassion at different adolescent developmental stages and found that level of self-compassion did not differ between males and females in middle school but differed significantly between males and females in high school, where girls scored lower in self-compassion than their male counterparts, as well as lower in self-compassion than middle school females. Further, the inverse relationship between self-compassion and negative affect was significantly greater among older adolescents compared to that of younger adolescents (Bluth & Blanton, 2015).

Much has been posited about why adolescent females experience greater depressive symptoms and negative affect than males (Nolen-Hoeksema & Girgus, 1994), and likely similar explanations exist for lower self-compassion. First, there may be biological explanations; puberty brings increases of estrogen and testosterone, and these changes, and particularly the amount of fluctuation in these changes, have been linked to depression (Andersen et al., 2022; Copeland et al., 2019). Social changes take place at this time as well; gender intensification theory suggests that in early adolescence, boys and girls take on more “traditional” gender roles for a time, which work in boys’ favor; boys tend to dominate and direct conversations, for example, resulting in a loss of agency for girls (Del Giudice, 2015).

Self-compassion levels have also been measured among sexual/gender minority adolescents (i.e., LGBTQIA+) and were found to have lower level of self-compassion than their sexual/gender majority peers, with a medium effect size. Vigna et al. (2017) postulated that this is an effect of the internalization of discrimination and stigma that many SGM experience daily. Indeed, minority stress theory explains that SGM adolescents suffer from substantial proximal and distal stressors, such as discrimination in school, rejection from family, and discomfort with their identity resulting in higher rates of depression, anxiety, and suicidality compared to their non-SGM peers (Goldbach & Gibbs, 2017).

In another study of 238 Canadian high school students, results indicated significantly higher self-compassion in males than in females (Pullmer et al., 2019b); this longitudinal study investigated relationships between baseline self-compassion and the mediating function of psychological distress on eating behaviors and body satisfaction at two time points, 14 to 18 weeks apart. While findings demonstrated that self-compassion was positively associated with body satisfaction, and negatively associated with eating pathology for both males and females at both time points, changes in psychological distress mediated the relationships between self-compassion and the outcomes of body satisfaction and eating pathology for females but not males. The authors suggested that it may be that there are other pathways in which self-compassion influences body satisfaction and eating pathology in males, such as concerns about muscle mass, for example (Pullmer et al., 2019b).

There is some evidence that the response to self-compassion interventions also differs by gender. Bluth and Eisenlohr-Moul (2017) reported a trend for a greater increase in self-compassion in females than males following an 8-week self-compassion program. This is similar to findings from a study of a 6-week mindfulness program in which there was preliminary evidence that females were more engaged than males; for example, only 60% of boys used mindfulness stress-reduction techniques during the post-intervention stress test, compared to 100% of the girls (Bluth et al., 2017). As developmental maturity occurs at different ages for males and females, differences may be related to males’ relative lack of maturity and emotional states compared to females of the same age (Steinberg & Morris, 2001). Also, as adolescent males often adopt traditional male hegemonic norms, they may be reluctant to engage in compassion programs that they perceive as being “weak” or “soft” (Kirby & Kirby, 2017); rather, programs that emphasize the cultivation of courage or strength may be more appealing.

Self-compassion levels can also vary according to the age of the adolescent. Stolow et al. (2016) reported that older adolescents (aged 12–16 years) showed higher levels of self-criticism and depressive symptoms, and lower self-esteem and self-compassion, when compared to younger adolescents (aged 9–10 years). Although not measured in this study, previous literature attributes this to the fact that older female adolescents have lower self-compassion, higher negative affect, and less life satisfaction than their younger female adolescent counterparts (Bluth & Blanton, 2015). Further, older adolescents have also shown greater increases in self-compassion scores following an intervention when compared to younger adolescents (Bluth & Eisenlohr-Moul, 2017); this may be because having lower self-compassion levels at baseline affords a greater opportunity to raise level of self-compassion across an intervention. Interestingly, the meta-analysis of studies on self-compassion that examined gender differences found that although men reported higher self-compassion levels than women, this association diminished with increasing age, perhaps due to increased development of the common humanity aspect of self-compassion in older individuals (Yarnell et al., 2015); that is, with age, people are more able to see that experiencing emotions, particularly emotions that are challenging, is part of the experience of being human.

Self-Compassion as a Mediator: The Role of Self-Compassion in Predicting Outcomes

Self-compassion has been shown to be a mechanism which explains the relationship between various risk or protective factors that adolescents are exposed to and their psychological outcomes. For example, Neff and McGehee (2010) reported that self-compassion partially mediated the relationship between early life influences, including maternal support and family functioning and adolescent well-being, defined by depression, anxiety, and connectedness. Adolescents’ reports of maternal support and better family functioning led to greater self-compassion, which was then associated with less depression and anxiety and greater sense of connectedness. This suggests that healthy family functioning provides support for establishing and maintaining self-compassion in the individual, which then leads to better mental health during adolescence.

In another study, self-compassion mediated the relationship between shameful memories and traumatic experiences that had been experienced earlier in life and depressive symptoms occurring in adolescence. Shameful memories arise from past experiences in which one is the target of exclusion, rejection, or criticism; the recollection of these experiences later in life can produce intense emotional responses, including feelings of shame, self-criticism, and isolation (Castilho et al., 2017). Adolescents are particularly vulnerable to shame memories because they are engaged in the development of self-identity and social connectedness, and memories of past experiences have a strong effect on the maturation of a positive self-identity (Castilho et al., 2017). Among 1100 adolescents in Portugal, Castilho et al. (2017) found that self-compassion mediated the relationship between adolescents’ shame memories and depressive symptoms, such that shame memories were linked with lower self-compassion, which in turn was associated with greater depressive symptoms. Higher levels of self-compassion were associated with higher emotional self-regulation and self-soothing, which may have contributed to the adolescents’ ability to use more effective strategies to deal with difficult emotions. One crucial strategy for the healthy functioning of adolescents is the formation of supportive social bonds; this study concluded that self-compassion may facilitate a sense of interpersonal connectedness and of common human experience, which may help adolescents build social bonds and lessen shame and self-criticism, which is then linked with lower depression (Castilho et al., 2017).

Similarly, 109 adolescents with poor school performance who had experienced being victimized (i.e., victimization as a child by a parent or siblings; having experienced assault, robbery, or kidnapping; sexual victimization; indirect victimization such as observing others being victimized; Internet victimization) were assessed (Játiva & Cerezo, 2014). Those who had experienced poly-victimization (i.e., several types of victimization) and directly experienced victimization (as compared to indirect victimization) had worse psychological maladjustment than others, and self-compassion partially mediated the relationship between victimization and psychological maladjustment. In other words, those adolescents who had experienced more victimization and direct victimization had lower levels of self-compassion, which was then associated with higher levels of psychological maladjustment and more internalizing and externalizing problems (Játiva & Cerezo, 2014).

Wu et al. (2019) found that self-compassion mediated the relationship between peer acceptance and non-suicidal self-injury (NSSI) in 816 Chinese adolescents. Increased peer acceptance led to increased self-compassion, which then was associated with decreased depressive symptoms and decreased NSSI. This mediation model was moderated by behavioral impulsivity; for those adolescents with higher levels of impulsivity, increased self-compassion was linked with decreased NSSI, while for adolescents with lower levels of impulsivity, there was no significant relationship between self-compassion and NSSI (Wu et al., 2019). The implications of this study are that for adolescents who are impulsive, such as those with attention deficit hyperactivity disorder (ADHD) who may struggle with being accepted by their peers, cultivating self-compassion may be particularly important for decreasing depressive symptoms and NSSI.

In a study with 1872 US adolescents, Vigna et al. (2017) examined the relationship between bias-based bullying and victimization related to sexual/gender minority (SGM) status. Bias-based bullying is the targeted, stigmatized harassment of those who are perceived as marginalized or “other” in some way by their peers. SGM youth reported higher levels of exposure to adversity, risk-taking behaviors, bias-based bullying, peer victimization, and depression and anxiety, and lower levels of self-compassion, compared to their SGM majority peers, and self-compassion attenuated the relationship between bias-based bullying/peer victimization and both anxiety and depression. The degree to which self-compassion mediated the association of bias-based bullying and anxiety and depression varied according to the adolescents’ level of exposure to bias-based bullying; those with higher levels of bias-based bullying reported less of a mediating effect of self-compassion on this association (Vigna et al., 2017). These results suggest that for those with relatively lower levels of bias-based bullying, greater self-compassion could potentially facilitate the pathway to reducing the adverse effects of bias-based bullying and internalization of stigma and thereby improve the mental health outcomes for SGM youth.

Interventions: Cultivating Self-Compassion in Adolescents

Recognizing the beneficial correlates of self-compassion in adolescents, and the positive outcomes of the Mindful Self-Compassion (MSC) program developed by Neff and Germer (2013), an adaptation of MSC was created for adolescents, called Mindful Self-Compassion for Teens (formerly known as Making Friends with Yourself) (MSC-T). MSC-T is currently an eight-session group program involving weekly sessions of 1.5-hour duration that teaches skills of self-compassion. Implementation of MSC-T has demonstrated decreases in depression, anxiety, negative affect, and stress and increases in self-compassion pre- to post-intervention (Bluth et al., 2021; Bluth & Eisenlohr-Moul, 2017; Bluth et al., 2016a; Galla, 2016). Like MSC, MSC-T has a foundation of mindfulness and focuses on teaching skills of self-compassion; in particular, participants in both MSC and MSC-T are taught to recognize that most of us tend to be much harder on ourselves than we are on others and that harsh self-criticism often is detrimental to our mental health. For example, in the MSC-T program, adolescents learn to recognize when their inner critic arises and are taught to then actively take steps to practice being kinder and more compassionate to themselves. Perhaps most importantly, adolescents discover that they are not alone in their struggles and that other adolescents encounter many of the similar emotional pitfalls as do they. Often, this is very eye-opening to adolescents.

In addition to explicitly introducing and teaching self-compassion practices in the MSC-T program, self-compassion is also taught implicitly in mindfulness programs and retreat settings. In these programs, self-compassion is embedded within the context of teaching mindfulness. For example, in the adolescent mindfulness programs such as Learning to BREATHE (Broderick, 2021) and iBme (www.ibme.com), concepts of self-compassion are taught implicitly within the program, and self-compassion practices are included, although the focus of the program is learning mindfulness skills. Galla (2016, 2017) measured change in self-compassion across two studies in two subsequent years in a mindfulness retreat program for adolescents and reported that within-person change in self-compassion was the “driver” behind more of the outcomes than within-person change in mindfulness. Self-compassion predicted decreases in perceived stress, rumination, depressive symptoms, and negative affect and increases in positive affect and life satisfaction, for example (Galla, 2016).

Limitations and Future Directions

As research on adolescents and self-compassion is just over a decade old, there are many gaps in the literature that need to be addressed. First, most studies are cross-sectional which precludes determining a causal link between self-compassion and mental health outcomes. Although Krieger et al. (2016) provided support that self-compassion was antecedent to depression in a cross-lagged panel analysis with an adult sample, longitudinal studies are needed among adolescent populations to support direction of effects. For example, it may be that among adolescents, those with less depressive symptoms are more able to be self-compassionate, rather than self-compassion predicting depressive symptoms, or it may be that these relationships are bidirectional. Additionally, longitudinal studies with at least three time points are needed to determine mediation pathways, as cross-sectional mediation studies have distinct limitations, and should be used only as a preliminary way to conceptualize mediation (MacKinnon, 2008).

Also, there is a lack of standardization of outcome assessments across studies, including standardized measures assessing emotional well-being and mental health. As one example, in their meta-analysis, Pullmer and Chung et al. (2019a) reported that studies utilize numerous measures of depressive symptoms. Using standard measures would allow for comparison across studies and make synthesizing the results of these studies more accurate. Recent debates in the literature over the factor structure of the self-compassion scale (e.g., Muris et al., 2019; Neff et al., 2018) have added variability to the way self-compassion is defined and measured. Hopefully, the publication of the youth self-compassion scale (Neff et al., 2021) will help to standardize self-compassion assessment in youth. In addition, there is lack of standardization of definitions across different age categories (e.g., young teens versus older teens, middle school versus high school), and using agreed-upon definitions of these different stages of development would make it more possible to compare across studies. Similarly, acknowledging multiple genders and establishing criteria for multiple genders within research studies (i.e., providing various gender choices) would better align with current conceptualizations of gender.

There is also a dearth of interventional studies, and the field will benefit from implementing manualized interventions, such as MSC-T, and assessing outcomes in various populations and subpopulations of youth. MSC-T can also be tailored to meet the needs of different populations; for example, Bluth et al. (2021) adapted MSC-T for transgender adolescents with promising findings, and Boggiss et al. (2020) adapted MSC-T to a brief format for adolescents with Type 1 diabetes and disordered eating. Recognizing the differences in self-compassion with regard to gender and age is critical in developing adolescent self-compassion interventions, as it is prudent to tailor interventions to the unique developmental and emotional needs of the participants.

Further, as adolescents are generally adaptable to online learning, and implementing MSC-T online with minor modifications has been shown to be feasible (Bluth et al., 2021), it may be that online implementation of self-compassion interventions is the most effective way to reach adolescents, particularly those in remote areas or those who, due to illness or other reasons, are unable to attend a program in-person (Finlay-Jones et al., 2020). Further, it would be valuable to engage the adolescents as community advisors on how to best tailor the intervention to meet the specific needs of their subpopulation, as well as how to best reach these adolescents and recruit them into the program. For example, in a review of self-compassion as an active component in the prevention of anxiety and depression in adolescents and young adults, Egan et al. (2022) interviewed 20 young people, to ascertain their understanding of self-compassion and how to best offer self-compassion programming to adolescents and young adults. Findings were illuminating; for example, certain terms, such as “intervention” and even “self-compassion” and “self-love,” were turnoffs for some, whereas for one participant, appealing to cultivating wisdom and courage was more attractive and would get more “buy-in.” Many voiced that they had similar fears of compassion as that of adults; that is, they were afraid of losing motivation to accomplish their goals and expressed that self-compassion sounded “weak” or “lazy.” Clearly, it would be important to clarify these misconceptions when teaching self-compassion to adolescents. Additionally, many contributed that designing programs that addressed the diversity which exists among young people is essential, as the “one-size-fits-all” model does not necessarily work. Ideally, these interventional studies need to be randomized and controlled, preferably with an active control group but, at minimum, with a waitlist control or a treatment-as-usual control group. Follow-up assessments (e.g., at 3 months, 6 months, and 1 year) are needed to determine whether outcomes are maintained over time. Lastly, intervention studies need to be replicated by different researchers in different settings to support generalizability.

Further, ascertaining the intervention components that are the most effective in achieving outcomes is needed. Various guided meditations, exercises, didactic instruction, and discussion comprise the interventions. Determining the critical components that are vital in achieving positive outcomes would be advantageous in further tailoring the interventions. Also, it may be that certain components are more efficacious in a particular culture or among a particular subpopulation of adolescents, and other components may be beneficial in other settings. Chio et al. (2021) found differential effects of self-compassion components in dialectical and non-dialectical cultures among adults; there may be similar effects for youth populations as well. Thus far, no research to our knowledge has investigated for whom and under what conditions which components of the interventions are most effective.

In addition, it is important to evaluate the fidelity of the intervention and whether it is being delivered in the way it was intended. Fidelity checks should be included to ensure that all components of the intervention are being taught. Also, the amount of training and experience that the instructors have should be noted, as well as their own practice history with mindfulness and self-compassion. Being able to embody mindfulness and self-compassion, as well as thorough training in teaching contemplative practices in groups, is highly recommended and included as one of the six domains in the Mindfulness-Based Interventions: Teaching Assessment Criteria, a tool created to systematically evaluate the integrity of mindfulness-based interventions (Crane et al., 2013; Griffith et al., 2021). Additionally, experience working with adolescents in an educational or therapeutic capacity should be a requirement, as teaching adolescents has specific challenges unique to their developmental stage.

Future research should use larger and more diverse sample populations to determine whether findings can be generalized to the general population. Self-compassion can be studied in adolescent subpopulations, particularly in marginalized or intersecting groups, including BIPOC, sexual/gender minorities, adolescents experiencing homelessness or undocumented status, those with a history of childhood abuse, those struggling with chronic health conditions, and recent immigrants. The differences between different adolescent genders merit investigation, as it is possible that different interventional approaches and formats may be more effective for different genders. For example, when teaching adolescents who have experienced trauma, it is important to take measures to ensure their emotional safety, providing them with options such as keeping their eyes opened or closed or even disengaging with the exercise completely. For some subpopulations, such as those who have experienced physical or sexual abuse, those with eating disorders, or those who identify as transgender, it may be important to not include the compassionate body scan, as this could be unnecessarily triggering.

Various implementation settings merit study as well, including schools, hospitals, juvenile detention systems, foster care programs, and afterschool programs. Most notably, as most adolescents worldwide attend public or private schools, implementing MSC-T or other self-compassion programs in schools is an important direction to increase the availability and accessibility of such programs. An adaptation of MSC-T for school implementation has been created but has not yet been tested empirically in school settings. The ability to reach all students, not only those whose parents or caregivers have the resources or opportunity to enroll them in self-compassion programs, may have far-reaching implications for the overall mental health of adolescents and, more distally, their long-term academic and behavioral functioning.

Finally, self-compassion and its associations with mental health may vary at different points in the life cycle and at different developmental stages of childhood and adolescence. Ideally, research studies would elucidate the windows of sensitivity in youth development at which the beneficial effects of self-compassion interventions would be most efficacious. For example, it may be that implementing a program to cultivate self-compassion in late childhood for females may help to evade the mental health plummet that is so common among adolescent females. It may be that other genders need a different approach at a different stage of development.

In conclusion, adolescents today struggle with an array of mental health challenges that have implications for lifelong maladaptive trajectories. Self-compassion has been shown to be associated with positive mental health factors, and interventions which cultivate self-compassion have been instrumental in providing coping mechanisms to shift the way in which adolescents respond to these stressors. Implementing self-compassion interventions has tremendous potential for adolescents, and future research is needed to replicate and confirm the current findings.