Risk and protective factors for suicide are under-researched in developing Asian countries. This study investigated the potential protective role of self-compassion and its subscales (compassionate and uncompassionate self-responding) in moderating the relationship between thwarted interpersonal needs and suicide ideation in young adults.
Method
Indonesian university students (n = 510; age = 18–25 years) completed an online questionnaire that measured thwarted belongingness, perceived burdensomeness, recent suicide ideation, depressive symptoms, self-compassion, and demographic factors. Hierarchical linear regression models tested possible moderation effects.
Results
Significant correlations were found between suicidal ideation, thwarted interpersonal needs (i.e., perceived burdensomeness and thwarted belongingness), depressive symptoms, and self-compassion (including both the positive and negative subscales of compassionate and uncompassionate self-responding), all in the expected directions. A three-way interaction was observed between compassionate self-responding, perceived burdensomeness, and thwarted belongingness in determining suicidal ideation severity. The findings suggest that individuals with high levels of compassionate self-responding tend to experience lower levels of suicidal ideation, even in the presence of high perceived burdensomeness and thwarted belongingness.
Conclusions
Compassionate self-responding (i.e., responding to life stressors with self-kindness, a sense of common humanity, and mindfulness) may be an important protective factor in buffering the adverse effects of thwarted interpersonal needs on young people’s suicidal ideation. Therefore, interventions that enhance individuals’ ability to treat themselves with compassion may help reduce suicidal ideation.
Preregistration
This study is not preregistered.
Opmerkingen
Ferdi W. Djajadisastra and Jennifer S. Ma contributed equally to this work.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The World Health Organization (WHO, 2014) recognizes that suicide is a global public health challenge and aims to reduce suicide rates around the world by one-third by the year 2030 (WHO, 2021a). However, suicide and suicide-related behaviors are under-researched in developing countries in the Asian region, where over half of the world’s suicide incidents occur (Beautrais, 2006; WHO, 2008). It has been estimated that 20.00% of the South, South-East, and Pacific Asian population simply do not have available suicide data, limiting the extent to which countries are able to systematically understand and prevent suicides at their local and national levels (Vijayakumar et al., 2005). Stigma, socio-cultural and religious attitudes, legal prohibitions, and limited access to primary and mental health care contribute significantly to the underreporting of suicides across these culturally diverse regions (Chen et al., 2012; Milner & De Leo, 2010).
Indonesia is an emerging middle-income, South-East Asian country with an age-standardized suicide rate of 2.60 per 100,000 people (WHO, 2021b). However, it is also a country recently identified as having the highest underreporting rates for suicide (859.10%) in a nationally representative sample (Onie et al., 2024). Moreover, studies suggest that suicide ideation (SI), a major risk factor for later suicidal behavior and attempts (Nock et al., 2008; Ribeiro et al., 2016), was experienced by 40.00% of the adult Indonesian population during the COVID-19 pandemic, with the highest incidence among young adults aged 18 to 24 years (Liem et al., 2022).
It is clear that suicidality and its prevention represent a more significant challenge to Indonesia’s population health than has been previously indicated. Yet, Indonesia’s health system has yet to display the same levels of readiness to support responses to suicidal behavior (i.e., availability of emergency and mental health services, mental health literacy in the community) seen in Western contexts (Setiyawati et al., 2022). Only 1.50% of Indonesia’s budget is dedicated to mental health services (Pratiwi & Perdana, 2019), compared to 7.00% in Western countries like Australia (Australian Institute of Health & Welfare, 2023). There is an urgent need for more suicide prevention advocacy and research in Indonesia. Utilizing leading theoretical models of suicide in such efforts may help to facilitate the awareness and identification of effective and acceptable targets for the prevention of suicide and suicide-related behaviors in the Indonesian context.
The interpersonal-psychological theory of suicidal behavior (IPTS; Joiner, 2005; Van Orden et al., 2010) consolidates a wide range of established risk factors for suicide and suicide-related behaviors to provide testable predictions of who will develop thoughts of suicide and who will go on to attempt suicide. The IPTS proposes that two, overarching cognitive processes occur when individuals experience disruptions in their interpersonal relationships, giving rise to passive thoughts of suicide or SI (i.e., “I wish I was dead”): thwarted belongingness (TB) and perceived burdensomeness (PB). TB involves feelings of loneliness (e.g., “I am alone”) and of lacking reciprocal, caring relationships. PB involves viewing oneself as a liability to others (e.g., “I am a burden”) and experiencing feelings of self-hatred (Van Orden et al., 2010). According to the IPTS, the experience of either TB or PB is sufficient for an individual to develop a passive desire for suicide (Ribeiro & Joiner, 2009). However, the risk of developing an active desire for suicide (i.e., I want to kill myself) is thought to be highest when both TB and PB co-occur and are viewed as unchangeable. This more active form of SI is seen as an immediate precursor to suicidal intent and subsequent lethal and near-lethal suicide attempt/s in the IPTS.
Meta-analytic research has supported the joint, interactive role of higher levels of TB and PB in contributing to increased SI severity (small effect) across undergraduate, community, and clinical populations (Chu et al., 2017). However, the majority of research on the IPTS has focused on testing theorized associations in Caucasian participants from developed countries (Ma et al., 2016). This observation mirrors broader criticisms around the Eurocentric nature of suicidology research and the significant need for more culturally diverse, and relevant, investigations and applications made within suicide theory and practice (Leong & Leach, 2007).
Of the few IPTS studies conducted in Asian countries, findings have indicated that South Korean students experience higher rates of TB and PB compared to American students (Seo, 2020; Suh et al., 2017), and PB is more strongly associated with SI in young male adults in Singapore than TB (Teo et al., 2018). Societal tendencies toward collectivism (i.e., when the self is construed in one’s relations to others and evaluated upon the ability to fit into one’s intergroup) and filial piety (i.e., respect, care, and obedience to elders) have been described as culturally relevant factors for the development of suicide risk in Asian contexts (Park et al., 2017).
Relevant to the present study’s Indonesian context, qualitative research on Javanese adults with a prior suicide attempt (n = 21) has shown that suicide is believed to be influenced by suffering arising from the experience of psychosocial stressors (e.g., chronic illness, violence, financial crisis, loss); loss of hope; alienation and loneliness; mental frailty in response to crisis; and low religiosity (e.g., weakness of faith, disobedience to religious teachings, spiritual detachment; Nurdiyanto & Subandi, 2023). Other beliefs about suicide reported by Indonesian stakeholders with lived or professional experiences around suicide (n = 51) have included viewing becoming a burden to other people (i.e., via not being able to reciprocate provided social support) as more shameful than taking one’s own life, as well as beliefs that suicide is influenced by ancient spirits, infectious, pre-ordained by God’s will (“pulung gantung”), and non-preventable (Setiyawati et al., 2024).
Such findings align with emerging literature on potential culture-specific factors for suicide, which suggests that culture uniquely influences the types of stressors that lead to suicide, the meanings associated with these stressors, and the ways suicidal thoughts are expressed (Chu et al., 2010; Leong & Leach, 2007). When intergroup relationships are viewed as negative and intractable in nature (i.e., “I am reminded every day that I will never live up to my parent’s expectations and am a burden to both them and society”) in such cultural contexts, targeting individuals’ levels of self-compassion may help to confer protection against negative TB and PB self-related cognitions and interpersonal suicide risk.
Empirically, self-compassion has been defined as an individual’s ability to approach their suffering with warmth and non-judgmental awareness, while recognizing that suffering is a universal human experience (Neff, 2003a, b). Inspired by Buddhist practices, philosophical literature, and neuroscientific research, this operationalization of self-compassion involves self-kindness versus self-judgment (i.e., being understanding, patient, and caring toward oneself rather than disapproving, judgmental, and intolerant about personal flaws); a sense of common humanity versus isolation (i.e., viewing one’s personal failings and inadequacies as part of the shared, human experience); and mindfulness versus over-identification (i.e., taking a balanced view of negative situations in a non-judgmental, present-focused manner rather than being consumed by feelings of inadequacy and everything that is viewed as wrong; Neff, 2003a, b).
Recent systematic and meta-analytic research has found that higher levels of self-compassion are moderately associated with a lower severity of suicidal thoughts and behaviors (Cleare et al., 2019; Suh & Jeong, 2021). A possible pathway through which self-compassion confers protection is via moderating the effect of psychological risk factors for suicidality on suicide-related outcomes, such as the negative, self-related cognitions associated with mental health problems and thwarted interpersonal needs. For instance, a study on American college students found that lower levels of self-compassion significantly moderated the effects of TB on SI and PB on SI (Fang, 2020). However, the paucity of research investigating the role of self-compassion on thwarted interpersonal needs in Asian populations, in addition to the long-standing debate around whether self-compassion, as theoretically operationalized by the Self-Compassion Scale (SCS; Neff, 2003a, b, 2016) is a unidimensional (Neff, 2020) or bi-factor construct (Muris & Otgaar, 2020), adds further complexity to any proposed protective pathways.
Some researchers have argued that the links between self-compassion, wellbeing, and psychopathological risk are better captured by considering the two distinct subdimensions of compassionate self-responding (CS; i.e., sum of the positively oriented self-kindness, sense of common humanity, and mindfulness SCS subscales) and uncompassionate self-responding (UCS; i.e., sum of the negatively oriented self-judgment, isolation, and over-identification SCS subscales) versus the total SCS score (Muris & Otgaar, 2022; Muris et al., 2016). CS has been found to be negatively associated with mental health problems, whereas UCS has been positively linked to psychopathology, suggesting that the negative components measured by the SCS may actually represent maladaptive mechanisms and not self-compassion (Brenner et al., 2018; Muris & Petrocchi, 2017; Muris et al., 2018).
Meta-analytic research examining the strength of correlations between CS, UCS, and suicide-related behaviors has also found CS to have a small negative effect (r = − 0.16) and moderate heterogeneity with suicidality outcomes, while UCS had a stronger, positive association (r = 0.23) but also displayed larger levels of heterogeneity (Per et al., 2022). Moreover, in Asian contexts, research suggests that UCS may play a more important role in the development of suicide risk when cross-sectionally studied in a sample of 1143 college students from Jiangsu Province, China (Ge, 2023), whereas CS disrupts the transition from SI to a later attempt when longitudinally examined in a sample of 520 Chinese adolescents (Sun et al., 2020). Taken together, this literature highlights the value and need for investigating the role of self-compassion, and its sub-components, on interpersonal suicide risk, as they may have differential risk and protective impacts across the suicide spectrum and cultural contexts.
Relating personally to one’s suffering with warmth, non-judgmental awareness, and a recognition of such suffering as being part of the universal human experience (i.e., self-compassion) may help individuals to buffer the effects of negative cognitions associated with suicide risk. However, it is currently unclear as to how self-compassion, and its sub-components, may be effectively targeted in interventions to meaningfully change the interactive experience of thwarted interpersonal needs and suicide risk (Bianchini, 2022). Furthermore, the application of research on risk and protective factors for suicide in developing Asian countries, such as Indonesia, remains limited (Chen et al., 2012; Jordans et al., 2014). This study aimed to investigate the protective role of self-compassion, and its positive and negative self-relating styles, on the relationship between the IPTS interpersonal risk factors and SI in Indonesian university students. In line with prior research, it was hypothesized that (1) SI would be significantly, positively correlated with the IPTS interpersonal risk factors of TB and PB, and the UCS component of self-compassion, but negatively correlated with self-compassion (total score and CS); (2) the relationship between thwarted interpersonal needs (i.e., TB and PB) and SI would be moderated by self-compassion, CS, and UCS; and (3) there is a three-way interaction between PB, TB, and self-compassion (or one of its positive and negative subscales) in determining the level of SI. The relationship between the interpersonal risk factors and SI was predicted to be stronger at low levels of self-compassion (i.e., low total SCS, low CS subscale, and high UCS subscale scores) compared to high levels of self-compassion (i.e., high total SCS, high CS subscale, and low UCS subscale scores).
Method
Participants
An a priori power analysis was conducted using G*Power 3.1 (Faul et al., 2007) to estimate the sample size needed to detect a medium effect (Cohen’s f2 = 0.15, α = 0.05, power = 0.80) in a regression model with eight predictors: depression, TB, PB, self-compassion (total score and subscales), TB × self-compassion, PB × self-compassion, TB × PB, and TB × PB × self-compassion. The power analysis indicated that a minimum of n = 109 participants was needed.
Out of the 529 participants recruited for this study, 510 were included in the final analysis (n = 13 excluded for failing to provide consent; n = 6 excluded for attending universities outside of Indonesia). Thus, the number of participants met the requirements of the power analysis. Participants’ ages ranged from 18 to 25 years (M = 19.74; SD = 0.06), and 76.67% (n = 391) identified as women. The majority (n = 437; 85.68%) were sampled from universities on Java Island. Only 5.49% (n = 28) reported having received a professional psychiatric diagnosis, 27.84% (n = 142) reported SI, and 10.98% (n = 56) reported a history of a lifetime suicide attempt.
Procedure
This paper presents an analysis of a subset of data obtained from a larger study that examined the psychometric properties of Indonesian-translated suicide measures (Geshica et al., 2023). Convenience sampling was used to recruit university students from across the 26 provinces of Indonesia to participate in an online survey that assessed participants’ levels of interpersonal risk (i.e., TB and PB), SI, depressive symptoms, self-compassion, and demographic factors (age, gender, university location, diagnosis of a mental health problem, and lifetime history of suicide attempt). A dedicated website provided potential participants with information about the study and was shared through research networks using snowballing. Online surveys were administered using Google Forms between 2 and 17 July, 2021. Informed consent was provided electronically.
Measures
All of the measures used in this study were originally developed in English. In cases where Indonesian-translated versions were unavailable for use, the authors (FWD and LG) translated the measures into Indonesian. Where possible, translation quality was discussed with the original authors of the measures prior to use. Feedback from the original authors of the Interpersonal Needs Questionnaire (Van Orden et al., 2012) and the Self-Compassion Scale (Neff, 2003a) was used to inform the present research.
Interpersonal Needs
The 15-item version of the Interpersonal Needs Questionnaire (INQ-15; Van Orden et al., 2012) was used to assess recent experiences of TB and PB, as outlined by the IPTS. The INQ-15 consists of 9 items that assess TB (range score 9–63) and 6 items for PB (range score 6–42). Statements such as “I feel disconnected from other people” and “people in my life would be better off if I were gone” were rated on a Likert scale ranging from 1 (Not at all true for me) to 7 (Very true for me). Higher scores indicate a greater degree of thwarted interpersonal needs. In this study, the TB and PB subscales demonstrated good to excellent reliability (ω = 0.83 and ω = 0.94, respectively).
Depressive Symptoms
The 8-item version of the Patient Health Questionnaire (PHQ-8; Kroenke et al., 2001, 2009) was used to measure depressive symptoms in the general population based on the Diagnostic Statistical Manual of Mental Disorders (DSM-4; American Psychiatric Association, 2000) criteria. Respondents rated the frequency of problems such as having little interest in doing things, feeling down, or having trouble with sleep, energy, appetite, and concentration over the past 2 weeks on a 4-point Likert scale (Not at all to Nearly every day). The PHQ-8 had good internal consistency in this study (ω = 0.88).
Suicide Ideation
Severity of SI was measured using the suicidality subscale of the Depressive Symptom Inventory (DSI-SS; Metalsky & Joiner, 1997). The DSI-SS comprises of 4 items that assess the frequency and intensity of suicidal ideation and impulses experienced, over the past 2 weeks, on a scale from 0 to 3. Higher scores indicate greater severity of SI. The DSI-SS has been found to display good convergent validity with measures for depressive symptoms, hopelessness, TB, PB, and impulsivity, and the ability to differentiate between respondents with and without suicidal history in population-based samples (Von Glischinski et al., 2016). In this study, the DSI-SS demonstrated excellent reliability (ω = 0.94).
Self-Compassion
The short-form Self-Compassion Scale (SCS-SF; Neff, 2003b; Raes et al., 2011) was used to measure self-compassion, as well as its individual facets of CS (compassionate self-responding; i.e., tenderness, patience, and empathy toward oneself) and UCS (uncompassionate self-responding; i.e., bogged down by one’s suffering and negative experiences). Participants rated the extent to which they did or did not take a balanced view of themselves and situations; saw their failings as part of the human condition or as an isolating experience; and responded with care and tenderness or self-judgment in relation to personal flaws on a scale from 1 (Almost Never) to 5 (Almost Always). Higher scores on the full scale (range score 12–60) indicate greater levels of self-compassion, while higher scores on the individual CS and USC subscales (range score 6–30) reflect higher levels of compassionate or uncompassionate self-responding. Factor analytical research has supported the use of the SCS subscales (Muttaqin et al., 2020) and total scale (Sugianto et al., 2020) in Indonesian populations. In this study, the Indonesian-language SCS-SF demonstrated acceptable internal consistency for the CS and UCS subscales (ω = 0.80 and 0.79, respectively), and sufficient internal consistency for the total scale (ω = 0.76).
Data Analyses
Pearson’s correlations were used to examine the strength of associations between all of the study variables. The independent variables were centered before the regression analysis. Several hierarchical linear regression models were built to test whether the relationship between thwarted interpersonal needs (i.e., TB and PB) and SI would be moderated by self-compassion (or its subscales, CS and UCS). The Huber-White robust estimation method was used due to violations of regression assumptions related to the non-normal distribution of the dataset (Huber, 1967; White, 1980).
As shown in Fig. 1, Model 1 included the main effects of TB, PB, and recent depressive symptoms (as a control) in relation to SI, with either the SCS-SF (Model 1A) or its subscales for CS (Model 1B) or UCS (Model 1C) added as predictors. Model 2 incorporated the two-way interactions between each of the thwarted interpersonal needs and self-compassion (Model 2A), CS (Model 2B), or UCS (Model 2C). Finally, Model 3 added the two-way interaction between TB and PB, and its three-way interaction with self-compassion (Model 3A), CS (Model 3B), or UCS (Model 3C) on SI.
Fig. 1
Regression models for investigating the moderating role of self-compassion in the association between thwarted interpersonal needs and suicide ideation
×
A moderator graph was created using simple slopes analysis to represent the significant three-way interaction between TB, PB, and the UCS subscale on SI. Following the recommendations of Preacher et al. (2006), participants were categorized into four groups based on their levels of TB and PB (i.e., low TB-low PB, high TB-low PB, low TB-high PB, and high TB-high PB). Participants were considered to have high levels of PB or TB if they were at + 1 SD, and low PB or TB at − 1 SD. Lastly, we examined whether the linear trends of each group differed significantly from one another based on pairwise comparisons of the slopes. The p-value was rounded to three decimal places and the significance threshold was set at 0.05. The analyses were conducted using STATA 16 (StataCorp, 2019).
Results
Descriptive and Correlational Analysis
Table 1 presents the descriptive statistics and correlations between the interpersonal risk factors, SI, depressive symptoms, and self-compassion, including its total score and subscales. As predicted, SI was significantly positively correlated with the interpersonal risk factors of TB and PB, depressive symptoms, and the UCS component of self-compassion. The positively valanced CS and self-compassion measures were negatively correlated with SI.
Table 1
Correlations between interpersonal risk, suicide ideation, depressive symptoms, and self-compassion (n = 510)
Variables
1
2
3
4
5
6
7
1. TB (INQ-15)a
-
-
-
-
-
-
-
2. PB (INQ-15)a
0.45*
-
-
-
-
-
-
3. SI (DSI-SS)
0.22*
0.58*
-
-
-
-
-
4. Depressive symptoms (PHQ-8)
0.40*
0.56*
0.43*
-
-
-
-
5. Self-compassion (SCS-SF)
− 0.42*
− 0.54*
− 0.43*
− 0.49*
-
-
-
6. Compassionate self-responding (CS)b
− 0.36*
− 0.38*
− 0.32*
− 0.26*
0.73*
-
-
7. Uncompassionate self-responding (UCS)b
0.30*
0.45*
0.36*
0.49*
− 0.82*
− 0.20*
-
M
31.18
17.18
1.05
16.94
38.70
23.01
20.30
SD
9.59
8.86
2.11
5.34
6.53
3.86
4.54
Range
9–63
6–42
0–12
8–32
18–56
10–30
7–30
Skewness
0.27
0.62
2.31
0.69
− 0.14
− 0.36
− 0.07
Kurtosis
2.92
2.60
8.42
2.97
2.95
2.97
2.62
PB, perceived burdensomeness; TB, thwarted belongingness; SI, suicide ideation; *p < 0.05; aRespective construct subscales from the INQ-15 were used, bRespective construct subscales from the SCS-SF were used
Moderation Analysis
The findings from the regression models and moderation analysis are summarized in Table 2. In Models 1A–C, higher levels of PB were significantly associated with greater SI, while higher levels of TB were linked to lower SI when controlling for depressive symptoms. Notably, TB only bordered on significance when UCS was included in the Model 1C. As hypothesized, higher levels of self-compassion and CS corresponded to lower SI, whereas increased levels of UCS were associated with higher SI.
Table 2
Regression analysis of interpersonal risk, self-compassion, and its subscales in relation to suicidal ideation (n = 510)
Model
B
SE
t
p
Model 1A (F(4, 505) = 45.38, p < 0.001, adj. R2 = 0.37)
Depressive symptoms
0.05
0.02
2.56
0.011*
TB
− 0.02
0.01
− 2.53
0.012*
PB
0.11
0.01
8.90
< 0.001*
Self-compassion
− 0.05
0.01
− 4.12
< 0.001*
Model 1B (F(4, 505) = 46.88, p < 0.001, adj. R2 = 0.37)
Depressive symptoms
0.07
0.02
3.46
0.001*
TB
− 0.02
0.01
− 2.64
0.009*
PB
0.12
0.01
9.05
< 0.001*
CS
− 0.07
0.02
− 2.94
0.003*
Model 1C (F(4, 505) = 40.58, p < 0.001, adj. R2 = 0.36)
Depressive symptoms
0.06
0.02
2.73
0.006*
TB
− 0.02
0.01
− 1.96
0.051
PB
0.12
0.01
9.74
< 0.001*
UCS
0.04
0.02
2.38
0.018*
Model 2A (F(6, 503) = 35.59, p < 0.001, adj. R2 = 0.41)
Depressive symptoms
0.04
0.02
2.02
0.044*
TB
− 0.02
0.01
− 2.41
0.016*
PB
0.10
0.01
8.04
< 0.001*
Self-compassion
− 0.05
0.01
− 4.15
< 0.001*
TB × Self-compassion
− 0.00
0.00
1.05
0.294
PB × Self-compassion
− 0.01
0.00
− 4.32
< 0.001*
Model 2B (F(6, 503) = 35.04, p < 0.001, adj. R2 = 0.39)
Depressive symptoms
0.06
0.02
2.79
0.005*
TB
− 0.02
0.01
− 2.54
0.011*
PB
0.12
0.01
9.21
< 0.001*
CS
− 0.05
0.02
− 2.58
0.010*
TB × CS
− 0.00
0.00
− 0.28
0.780
PB × CS
− 0.01
0.00
− 2.46
0.014*
Model 2C (F(6, 503) = 32.17, p < 0.001, adj. R2 = 0.38)
Depressive symptoms
0.05
0.02
2.18
0.030*
TB
− 0.02
0.01
− 1.71
0.088
PB
0.11
0.01
8.69
< 0.001*
UCS
0.06
0.02
2.76
0.006*
TB × UCS
− 0.00
0.00
− 1.25
0.213
PB × UCS
0.01
0.00
3.08
0.002*
Model 3A (F(8, 501) = 27.43, p < 0.001, adj. R2 = 0.41)
Depressive symptoms
0.04
0.02
2.07
0.039*
TB
− 0.03
0.01
− 2.58
0.010*
PB
0.10
0.01
7.97
< 0.001*
Self-compassion
− 0.05
0.01
− 3.64
< 0.001*
TB × Self-compassion
0.00
0.00
0.49
0.623
PB × Self-compassion
− 0.01
0.00
− 4.11
< 0.001*
TB × PB
− 0.00
0.00
− 1.28
0.200
TB × PB × Self-compassion
− 0.00
0.00
− 1.06
0.289
Model 3B (F(8, 501) = 28.70, p < 0.001, adj. R2 = 0.40)
Depressive symptoms
0.06
0.02
2.89
0.004*
TB
− 0.03
0.01
− 3.00
0.003*
PB
0.11
0.01
9.05
< 0.001*
CS
− 0.04
0.02
− 1.72
0.086
TB × CS
− 0.00
0.00
− 0.35
0.726
PB × CS
− 0.00
0.00
− 1.94
0.052
TB × PB
− 0.00
0.00
− 1.15
0.251
TB × PB × CS
− 0.00
0.00
− 2.01
0.045*
Model 3C (F(8, 501) = 23.96, p < 0.001, adj. R2 = 0.38)
In Models 3A–C, only the three-way interaction between TB, PB, and CS was found to be significant (B = − 0.00, t = − 2.01, p < 0.05). Self-compassion (p > 0.05) and CS (p > 0.05) did not function as moderator variables in the association between TB and PB toward SI. There was an increase in explained variance from Model 1B (R2 = 0.37) to Model 3B (R2 = 0.40), indicating that the inclusion of the three-way interaction provided minor, cross-sectional explanatory power in relation to SI.
As can been seen in Fig. 2, among individuals with low levels of CS, those with high levels of both TB and PB reported the highest levels of SI. Likewise, individuals with low TB and high PB showed elevated SI compared to other groups, though slightly lower than those with high levels of both TB and PB. Individuals with low levels of both TB and PB, as well as those with high TB and low PB, exhibited the lowest levels of SI. At high levels of CS, individuals with low levels of both TB and PB, as well as those with high TB and low PB, reported the lowest levels of SI. Surprisingly, those with high levels of both TB and PB had lower SI than individuals with high PB but low TB.
Fig. 2
Relationship between thwarted belongingness, perceived burdensomeness, and suicide ideation at low and high levels of compassionate self-responding (n = 510)
×
A significant difference in linear trends between the high TB and high PB group and the low TB and low PB group (dy/dx = 0.18, p < 0.05) was identified in the slopes analysis (Table 3). Similarly, a significant difference was observed between the high TB and high PB group and the high TB and low PB group (dy/dx = 0.12, p < 0.05). No other significant pairwise differences were found.
Table 3
Pairwise comparison between slopes for the thwarted interpersonal needs groupings on suicide ideation
Contrasts
dy/dx
p
2 vs 1
0.09
0.094
3 vs 1
0.18
< 0.001*
4 vs 1
0.12
0.023*
3 vs 2
0.09
0.205
4 vs 2
0.04
0.556
4 vs 3
-0.06
0.382
dy/dx, the derivative of y with respect to x; 1, High thwarted belongingness, high perceived burdensomeness; 2, Low thwarted belongingness, high perceived burdensomeness; 3, High thwarted belongingness, low perceived burdensomeness; 4, Low thwarted belongingness, low perceived burdensomeness; *p < .05
Discussion
This study explored whether self-compassion serves as a protective factor in moderating the relationship between thwarted interpersonal needs and recent SI in young Indonesian adults. Building on previous research (Fang, 2020), this investigation also analyzed whether the two components: CS and UCS had differential impacts on this association.
In alignment with past research (Fang, 2020; Per et al., 2022; Sun et al., 2020), SI showed a positive correlation with UCS, TB, and PB, while demonstrating a negative correlation with self-compassion and CS. After controlling for participants’ depressive symptoms, self-compassion, CS, and UCS were each found to moderate the relationship between PB and SI, but not between TB and SI. These findings generally align with research on the interpersonal-psychological theory of suicide (IPTS; Joiner, 2005) showing that PB has a stronger relationship with suicidality outcomes compared to TB across both Western and Asian contexts (Chu et al., 2017; Ma et al., 2016). Becoming a burden to other people via, for example, not being able to reciprocate the provided social support from others has also been specifically identified as a culturally important driver of suicide in Indonesia (Setiyawati et al., 2024). As such, PB may represent a more relevant target for initiatives aimed at reducing risk of suicidal thoughts more broadly in both the population and in young Indonesian adults.
When accounting for the possible moderating effect of self-compassion, and its sub-components, in the three-way interaction between TB and PB on SI, only CS was identified to have a moderating role. Here, low levels of CS, combined with high levels of TB and PB, conferred the highest risk of SI, providing support for the IPTS’ prediction regarding high levels of both TB and PB contributing to more severe SI risk. Subsequent analysis revealed that low levels of PB were linked to lower SI across different levels of CS. However, low levels of TB were not necessarily protective against SI when PB was high, even if an individual had high levels of CS.
The inconsistent findings for TB, as a main and interactive effect on SI, may have been affected by the study data being collected during the COVID-19 pandemic lockdown period (Nangoy & Diela, 2021). Due to the COVID-19 lockdown policies, participants’ levels of reported interpersonal suicide risk, SI, and depressive symptoms, as well as their ability to utilize regular, external supports was likely impacted. With many university students developmentally being in the emerging adult phase of their lives, where their identity and social lives revolve more greatly around peer groups compared to family relationships (Arnett, 2000), it is noteworthy that interactions with peers during the pandemic was highly limited to online communication through social media platforms (Pandya & Lodha, 2021). This context may have qualitatively and quantitatively altered how individuals perceived and experienced loneliness and the lacking of reciprocal, caring relationships (i.e., TB), as well as their methods of interpersonally coping. For instance, studies have shown that excessive screen time is associated with an increased risk of suicidal behaviors in adolescents (Coyne et al., 2021). Therefore, while some participants reported feeling connected to others, it is possible that social media–based interactions do not offer the same protective benefits against interpersonal suicide risk compared to the wider range of pre-COVID-19 opportunities for social interaction.
While initially counterintuitive, other possible explanations for the unusual TB findings may be that the normalization of heightened feelings of social disconnection during the pandemic could have helped some to feel less alone in their perceptions of aloneness, and thus lowered SI risk; or that low levels of TB were more an indicator of other risk factors, such as individuals not seeking interpersonal support when needed because they view themselves as low-risk, thus heightening their SI risk (Jetten et al., 2020). The complex interplay between social media interaction with peer groups, being confined at home alone or with family during the COVID-19 pandemic, and interpersonal suicide risk represents an interesting avenue for future research.
Taken together, the findings suggest that targeting the increase of self-compassionate responses to one’s failures and flaws (i.e., via self-kindness, sense of common humanity, mindfulness) may be a more effective pathway for buffering the harmful effects of thwarted interpersonal needs, particularly PB, on SI compared to decreasing uncompassionate responses (i.e., self-judgment, isolation, over-identification). Promoting individuals’ awareness of, and practical skills around, self-compassionate responding in relation to life and interpersonal stressors aligns with strengths-based, suicide prevention approaches. These approaches tend to advocate for the incorporation of culturally diverse perspectives and multiple sources of resilience (as opposed to risk factors) that are not currently captured by Western-oriented models (Allen et al., 2022; Dudgeon et al., 2020). Given that stigma, socio-cultural and religious attitudes, legal prohibitions, and limited access to primary and mental health care remain significant challenges to suicide underreporting in Asia, there is a strong need for more research on the nuanced role that culture may have on suicide resilience processes in countries like Indonesia.
A major strength of this study is that it is the first to investigate potential moderators of interpersonal suicide risk on SI in an Indonesian sample. This endeavor extends the application of the IPTS to diverse cultural contexts and, in doing so, plays a small part in addressing the over-reliance on sampling from developed, Western countries (Henrich et al., 2010). Furthermore, this study examined the differential roles of self-compassion, CS, and UCS on SI risk in young adults, providing much needed evidence to the ongoing debate around how to best understand and operationalize the construct of self-compassion (Muris & Otgaar, 2020; Neff, 2020). The distinctive contribution of CS in buffering the negative effect of PB on SI was particularly highlighted. Introducing self-compassion techniques from the Mindful Self-Compassion program (Neff & Germer, 2018) and Compassion-Focused Therapy (Sommers-Spijkerman et al., 2018) in multi-pronged, strengths-based suicide prevention and intervention initiatives may help to support young people experiencing thwarted interpersonal needs and symptoms of depression.
Limitations and Future Directions
The study also had several limitations. First, the cross-sectional nature of the study prevented observations about the causal direction of identified associations. Longitudinal studies are needed to map how expected inter-relationships between self-compassion and its sub-components, interpersonal risk and SI, operate over time. Second, because the sample consisted of Indonesian university students with Internet access, the sample may not be representative of the young Indonesian population more broadly. Future research should include young Indonesians from a variety of socioeconomic backgrounds, and utilize more longitudinal research methods, to facilitate greater generalizability of the findings. Lastly, the authors translated and back-translated the measures in cases where official Indonesian-language versions of the study’s self-report measures were unavailable. Although efforts were made to seek feedback from the original authors of the measures prior to use, responses were not obtained for all. It would be beneficial for future Indonesian-based studies to further evaluate, qualitatively and quantitatively, the construct validity of measures that have been adapted or translated from Western contexts for local applicability.
To conclude, suicidality and its prevention represent a more significant challenge to Indonesia’s population health than previously indicated. Yet the application of research on risk and protective factors for suicide in developing, Asian countries is limited. Utilizing leading theoretical models of suicide in such efforts may help to facilitate the awareness and identification of effective and acceptable targets for the prevention and intervention of suicide and suicide-related behaviors. In this study, having higher levels of compassionate self-responding (i.e., self-kindness, sense of common humanity, mindfulness) was identified as a potentially important protective factor that buffered the adverse effects of thwarted interpersonal needs, particularly feelings of being a liability and self-hatred (i.e., PB), on SI risk in young Indonesian adults. Longitudinal research in diverse populations is needed to further investigate the nuanced role culture may have on interpersonal suicide risk and suicide resilience processes.
Acknowledgements
FWD is supported by the Australia Awards Scholarship from The Australian Government Department of Foreign Affairs and Trade. JSM is supported by a Suicide Prevention Australia Post-Doctoral Research Fellowship funded by The Australian Government Department of Health’s National Suicide Prevention Research Fund (ID: 40966).
Declarations
Ethical Clearance
This research received ethical approval from the Faculty of Psychology at Universitas Gadjah Mada (No. 3487/UN11/FPsi.1.3/SD/PT.01.04/2021) and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
Informed Consent
All persons gave their informed consent electronically prior to their inclusion in the study.
Conflict of Interest
The authors declare no competing interests.
Use of Artificial Intelligence
During the preparation of this work, FWD used Grammarly in order to improve the English language. After using this tool/service, the authors reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
share
DELEN
Deel dit onderdeel of sectie (kopieer de link)
Optie A:
Klik op de rechtermuisknop op de link en selecteer de optie “linkadres kopiëren”
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.