Protective role of perceived social support against adolescent self-injury addiction: Serial mediation of negative emotion and impulsivity

Abstract

BACKGROUND

Repeated non-suicidal self-injury (NSSI) behaviors are frequently associated with negative emotions, impulsivity and addictive-like characteristics.

AIM

To explore the protective role of perceived social support against self-injury addiction among adolescents engaging in NSSI.

METHODS

This cross-sectional study enrolled 102 Chinese adolescents with NSSI from outpatient clinics at a local hospital between January 2024 and April 2024, and their perceived social support, depression and anxiety emotions, impulsivity, and self-injury addiction were assessed using the Multidimensional Scale of Perceived Social Support (MSPSS), Self-Rating Depression Scale (SDS), Self-Rating Anxiety Scale (SAS), Barratt Impulsiveness Scale (BIS-11), and addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory (ROSI-addiction).

RESULTS

Among adolescents with NSSI, ROSI-addiction scores were negatively correlated with MSPSS factors and positively correlated with SAS scores, SDS scores, and all BIS-11 factors except Cognitive instability. Meanwhile, MSPSS factors were negatively correlated with SAS and SDS scores; SAS and SDS scores exhibited positively correlations with the BIS-11 dimensions of Attention, Perseverance, Self-control and Cognitive complexity. Moreover, perceived social support demonstrated both direct and indirect negative relations with self-injury addiction, mediated by anxiety, depression and impulsivity.

CONCLUSION

Negative emotions and impulsivity mediated the inverse correlation between perceived social support and self-injury addiction in adolescents with NSSI. Enhancing available social support, and regulating anxiety, depression and impulsivity may be fundamental for intervention of in NSSI.

Key Words: Adolescent; Impulsivity; Negative emotions; Perceived social support; Self-injury addiction

Core Tip: This study reveals a novel psychological mechanism: Perceived social support protects adolescents with non-suicidal self-injury from self-injury addiction by sequentially alleviating negative emotions (anxiety/depression) and impulsivity. These findings highlight the critical role of social support as a protective factor and propose that interventions targeting emotional regulation and impulse control may effectively break the cycle of self-injury addiction. This serial mediation model offers a theoretical basis for developing multi-level clinical strategies for adolescents with non-suicidal self-injury.

INTRODUCTION

Non-suicidal self-injury (NSSI) is defined as the deliberate, self-inflicted damage of body tissue without suicidal intent, a behavior that is typically socially condemned[1]. Previous studies have reported that NSSI increases accidental injuries and predicts future suicide attempts among children, adolescents, and adults[2-4]. Moreover, a meta-analysis indicated that during the coronavirus disease 19 pandemic, the pooled prevalence of NSSI reached 32.40% among adolescent samples across Asia, Europe, and America[5]. Supporting this, a Chinese cross-sectional study further reported that among individuals with psychiatric disorders, NSSI was most prevalent in adolescents aged 10-19 years, and the prevalence of NSSI behaviors in females was significantly higher than in males[6].

Patients with NSSI often exhibit a strong craving for self-harm, an inability or unwillingness to cease the behavior, and the use of self-harm as a coping mechanism[7,8]. These addictive features are significantly associated with the clinical severity of NSSI[9]. Research based on the addiction model also indicates that repetitive self-harm shares behavioral similarities with addictive disorders[10]. Moreover, genetic and neurobiological evidence supports the notion that NSSI and addiction share underlying mechanisms, including the involvement of genes such as SLC6A4 and dysregulation in the dopamine and opioid systems[11,12]. Therefore, it is imperative to develop effective intervention strategies specifically aimed at addressing self-injury addiction among adolescents with NSSI behaviors.

Adequate social support has been shown to improve adolescents' mental health[13]. For instance, support from peers and teachers can foster positive outcomes even among adolescents in stressful family environments[14]. Notably, perceived social support, namely an individual's subjective appraisal of the adequacy of emotional, practical, informational, or financial support and satisfaction with it, are more influential than received support[15]. Conversely inadequate perceived social support, such as self-reported negative parenting or adverse childhood experiences[8], represents a significant interpersonal risk factor for NSSI and is linked to negative emotions like depression and anxiety, which are closely associated with NSSI[16-18]. While emotion regulation is a common intrapersonal function of NSSI, there is evidence that repeated self-harm may lead to addictive patterns[19], few studies have specifically examined how perceived social support influences self-injury addiction. It is plausible that such support mitigates negative emotions, thereby reducing addiction risk—though not all adolescents with poor social support and negative emotions develop addictive self-harm.

Impulsivity, a common feature of adolescent development, is defined as a predisposition toward rapid, unplanned reactions to stimuli, regardless of negative consequences to oneself or others, and involves complex cognitive and behavioral processes. In the context of NSSI, impulsivity serves as both a proximal driver and a distal maintenance factor for self-injury[20-23]. Moreover, impulsivity is closely linked to negative emotions. Research indicated that anxiety and depression exacerbated impulsive attitudes and behaviors, including suicidal ideation and attempts[24,25]. Recent studies have suggested that perceived social support mediated the relationship between anxiety and impulsivity[26]. Hence, perceived social support may help prevent repetitive self-harm by mitigating negative emotions and impulsivity. However, the precise pathway through which perceived social support influences self-injury addiction in adolescents—and the specific roles of negative emotions and impulsivity within this pathway—have not been elucidated.

Therefore, this study aimed to explore the protective role of perceived social support against self-harm addiction in adolescents with NSSI. We hypothesized that: (1) Self-injury addiction would be negatively correlated with perceived social support and positively correlated with negative emotions and impulsivity; and (2) Negative emotions and impulsivity would mediate the negative association between perceived social support and self-injury addiction.

MATERIALS AND METHODS

Participants

Between January 2024 and April 2024, a cross-sectional study was carried out using consecutive sampling to enroll 102 patients from a hospital in Zhejiang Province, China. The sample size was determined according to feasibility and patient availability, and all eligible individuals during the study period were invited to participate. The inclusion criteria were as follows: (1) Age between 12 and 18 years; (2) Fulfillment the diagnostic criteria for NSSI according to the Diagnostic and Statistical Manual of Mental Disorders, 5^th edition, as confirmed by three psychiatrists; and (3) Capability to complete scale-based assessments. Exclusion criteria included the presence of psychiatric or severe organic disorders that could impair comprehension of tasks, such as intellectual disability, schizophrenia, or autism, or being in state custody. Participants' sociodemographic characteristics are presented in Table 1. The study was approved by the local ethics committee and conducted in accordance with the Helsinki Declaration. Written informed consent was obtained from all participants' legal guardians prior to data collection.

Table 1 Sociodemographic characteristics and scores on the addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory in adolescents with non-suicidal self-injury behaviors (n = 102).

Sociodemographic characteristics	n (%)	ROSI-addiction	t/F	P value
Gender			1.20	0.234
Male	9 (8.82)	10.89 ± 4.94
Female	93 (91.17)	13.83 ± 7.19
Age (year)			1.02	0.309
12-14	32 (31.37)	14.63 ± 7.05
15-18	70 (63.52)	13.09 ± 7.05
Education level			0.40	0.693
Primary school student	49 (48.04)	13.86 ± 7.14
Junior high school student	53 (51.96)	13.30 ± 7.03
Primary caregivers			0.51	0.679
Parents	75 (73.53)	13.24 ± 7.09
Grandfather/mother	10 (9.80)	16.00 ± 7.35
Father only	8 (7.84)	14.37 ± 8.26
Mother only	9 (8.82)	12.89 ± 5.67
Others	-	-
Relationship of parents			0.07	0.976
Harmonious	70 (68.63)	13.36 ± 7.16
Disharmony	16 (15.69)	14.06 ± 6.74
Divorce	14 (13.73)	13.93 ± 7.25
Bereavement	2 (1.96)	14.50 ± 10.61
School bullying			3.38	0.021
Bully others	4 (3.92)	21.75 ± 2.22
Being bullied	26 (25.49)	12.50 ± 7.96
Bully others and being bullied	8 (7.84)	17.88 ± 5.79
Neither	64 (62.75)	12.95 ± 6.57
Diagnosis			1.52	0.189
Depressive disorder	73 (69.52)	13.56 ± 6.15
Bipolar disorder	16 (15.24)	13.25 ± 7.47
Anxiety disorder	5 (4.76)	15.40 ± 10.53
Obsessive-compulsive and related disorder	2 (1.90)	10.50 ± 0.71
Trauma and stressor-related disorders	7 (6.67)	14.29 ± 10.18
Disruptive, impulse-control and conduct disorder	2 (1.90)	12.50 ± 9.19

ROSI-addiction: Addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory.

Measures

Participants completed several clinical self-report measures and provided information on their sociodemographic characteristics, such as age, gender, educational level, and family background, through a questionnaire.

The Multidimensional Scale of Perceived Social Support: The Multidimensional Scale of Perceived Social Support (MSPSS) is a 12-item self-report instrument that measures perceived support from family, friends, and significant others[27]. Items (e.g., "I can talk about my problems with my friends") are rated on a 7-point Likert scale [1 (strongly disagree) to 7 (strongly agree)]. Higher total and subscale scores indicate greater levels of perceived social support. MSPSS has been shown to be reliable and valid in Chinese adolescents[28,29], and its Cronbach's α in this study was 0.906.

The Barratt Impulsiveness Scale: The Barratt Impulsiveness Scale (BIS-11) assesses the personality trait of impulsivity[30]. Its revised version comprises six first-order factors: Attention, cognitive instability, motor, perseverance, self-control, and cognitive complexity[31]. Items (e.g., "I do things without thinking") are rated on a 4-point Likert scale from 1 (never) to 4 (very frequently). The Chinese version, validated among adolescents, has shown strong internal consistency, test-retest reliability, and construct validity[32]. In the present study, its Cronbach's α was 0.802.

The Self-Rating Anxiety Scale: The Self-Rating Anxiety Scale (SAS) is a 20-item questionnaire designed to evaluate a range of anxiety symptoms, encompassing both psychological and somatic aspects. Respondents rate each item (e.g., "I feel more nervous and anxious than usual") on a 4-point scale. The Chinese version of the SAS has established reliability and validity in adolescent populations[33]. In this study, the scale demonstrated good internal consistency, with a Cronbach's α of 0.791.

The Self-Rating Depression Scale: The Self-Rating Depression Scale (SDS) comprises 20 items designed to assess depressive symptoms. Each item (e.g., "I feel down-hearted, blue and sad") is rated on a 4-point Likert-type scale ranging from 1 (none or a little of the time) to 4 (most or all of the time). Its Chinese version has been validated for young clinical patients[34]. In the current sample, the SDS achieved a Cronbach's α of 0.838.

The Revised Chinese Version of Ottawa Self-injury Inventory: The Revised Chinese version of Ottawa Self-injury Inventory (ROSI) is a brief version of the Ottawa Self-injury Inventory that measures onset, frequency, motivation to stop, types and functions, and potential addictive features of self-injury. Research in university samples supported the reliability of linking specific NSSI functions to addictive features[35]. Among them, seven items [making up the addiction subscale of the ROSI (ROSI-addiction)] were used to assess addictive traits (e.g., "Despite a desire to cut down or control this behavior, you are unable to do so"), with response options ranging from 0 (never) to 4 (always) for each addictive trait. Higher scores indicate greater addictive tendencies. The ROSI has shown satisfactory stability and validity, and it is suitable for clinical and scientific research on adolescents with NSSI[36], and its Cronbach's α in the current study was 0.822.

Statistical analysis

Student's t-test and one-way ANOVA were used to compare ROSI-addiction scores in adolescents with different sociodemographic characteristics, followed by Bonferroni post-hoc tests. When significant difference was found between, the sociodemographic variable would be included as a covariate in subsequent analyses. For hypothesis 1, the Pearson analysis examined the relationships between the MSPSS, SAS, SDS, BIS-11 and ROSI-addition scores. For hypothesis 2, MSPSS subscales were used as the independent variables, ROSI-addition as the dependent variable, and SAS/SDS and the BIS-11 factors as chain mediators, to test the mediating effects of negative emotions and impulsivity using the AMOS 24.0 (see Figure 1 for the theoretical model). To address the dimensionality of negative emotions, serial mediation models with separate anxiety and depression mediators were compared with an alternative higher-order “negative affect” model (comprising SAS and SDS). A P value < 0.05 was considered statistically significant. In addition, to assess the potential threat of common method bias, Harman's single-factor test was performed. All items from the MSPSS, SAS, SDS, BIS-11, and ROSI-addiction scales were subjected to an exploratory factor analysis.

Figure 1 The theoretical mediation model in this study. It has two sub-models based on negative emotions including anxiety and depression.

RESULTS

Harman's single-factor test showed that the first unrotated factor of all scales was below the 50% critical threshold, indicating that common method bias did not pose a serious threat in this study. Detailed data were provided in Supplementary Table 1. Furthermore, as showed in Table 1, there were no significant differences in ROSI-addiction scores between NSSI patients with different gender, age, education, family structure or relationships (Ps > 0.05). A significant group difference was found on ROSI-addition among NSSI patients with different bullying experiences (P = 0.032), although post-hoc showed no significance.

Correlations among the target variables

As shown in Table 2, ROSI-addiction was negatively correlated with each MSPSS factor and positively corrected with the SAS, SDS, and all BIS-11 factors except Cognitive instability (Ps < 0.05). All the MSPSS factors were negatively correlated with SAS and SDS (Ps < 0.05). SAS and SDS were positively correlated with BIS-11 attention, perseverance, self-control and cognitive complexity (Ps < 0.05).

Table 2 Correlations between perceived social support (by the Multidimensional Scale of Perceived Social Support), depression (the Self-rating Depression Scale), anxiety (the Self-rating Anxiety Scale), impulsivity (the Barratt Impulsiveness Scale), and self-injury addiction (the addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory) in adolescents with non-suicidal self-injury behaviors (n = 102).

		MSPSS			SDS	SAS	BIS-11
		Support from family	Support from friends	Support from significant others			Attention	Cognitive instability	Motor	Perseverance	Self-control	Cognitive complexity
MSPSS
Support from family	r value	1
	P value
Support from friends	r value	0.42c	1
	P value	< 0.001
Support from significant others	r value	0.64c	0.75c	1
	P value	< 0.001	< 0.001
SDS	r value	-0.51c	-0.21a	-0.41c	1
	P value	< 0.001	0.042	< 0.001
SAS	r value	-0.43c	-0.22a	-0.31b	0.81c	1
	P value	< 0.001	0.035	0.002	< 0.001
BIS-11
Attention	r value	-0.30c	-0.31c	0.29c	0.31b	0.33b	1
	P value	< 0.001	< 0.001	< 0.001	0.002	0.001
Cognitive instability	r value	0.10	0.02	0.11	-0.05	0.07	0.04	1
	P value	0.134	0.726	0.174	0.645	0.507	0.710
Motor	r value	-0.13	-0.10	0.30c	0.16	0.16	0.29c	0.35b	1
	P value	0.051	0.124	< 0.001	0.107	0.107	0.003	< 0.001
Perseverance	r value	-0.15a	-0.16a	0.24b	0.25a	0.25a	0.31b	0.03	0.36c	1
	P value	0.033	0.015	0.002	0.011	0.012	0.001	0.755	< 0.001
Self-control	r value	-0.28c	-0.25c	0.24b	0.32b	0.20a	0.27b	-0.12	0.32b	0.31b	1
	P value	< 0.001	< 0.001	0.002	0.001	0.046	0.006	0.247	0.001	0.002
Cognitive complexity	r value	-0.26c	-0.23b	0.20a	0.23a	0.24a	0.35c	0.14	0.57c	0.37c	0.39c	1
	P value	< 0.001	0.001	0.010	0.020	0.017	0.000	0.172	< 0.001	< 0.001	< 0.001
ROSI-addiction	r value	-0.30c	-0.32c	-0.40c	0.44c	0.50c	0.29c	0.11	0.30c	0.24b	0.24b	0.20a
	P value	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	0.174	< 0.001	0.002	0.002	0.01

^aP < 0.05.

^bP < 0.01.

^cP < 0.001.

MSPSS: Multidimensional Scale of Perceived Social Support; SDS: Self-Rating Depression Scale; SAS: Self-Rating Anxiety Scale; BIS-11: Barratt Impulsiveness Scale; ROSI-addiction: Addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory.

Chain mediating effects of depression/anxiety and impulsivity in the associations between perceived social support and self-injury addiction

The structural equation analysis assumed that MSPSS was the predictor, SAS/SDS and BIS-11 were the mediators, and ROSI-addiction was the dependent variable. AMOS 24.0 was used to assess the model fit, and the results are shown in Table 3. These data indicated a good fit, conforming the acceptability of the mediator model.

Table 3 Structural equation model fit index.

Model	χ2/df	CFI	GFI	TLI	RMSEA
MSPSS - SAS - BIS-11 - ROSI-addiction	1.404	0.939	0.906	0.917	0.065
MSPSS - SDS - BIS-11 - ROSI-addiction	1.380	0.942	0.901	0.920	0.064

MSPSS: Multidimensional Scale of Perceived Social Support; SAS: Self-Rating Anxiety Scale; BIS-11: Barratt Impulsiveness Scale; ROSI-addiction: Addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory; SDS: Self-Rating Depression Scale; CFI: Comparative fit index; GFI: Goodness of fit index; TLI: Tucker-Lewis index; RMSEA: Root mean square error of approximation.

As shown in Table 4, when SAS and BIS-11 were used as mediators, MSPSS had a significant negative total prediction on ROSI-addiction (β = -0.35, P = 0.002). It also had a significant negative indirect prediction on ROSI-addiction through SAS alone (β = -0.11, P = 0.007) and through SAS and BIS-11 (β = -0.02, P = 0.025). When SDS and BIS-11 were used as mediators, the MSPSS also showed a significant negative total prediction of ROSI-addiction (β = -0.35, P = 0.002); it also had a significant negative indirect prediction of ROSI-addiction through SDS alone (β = -0.11, P = 0.016) and through the SDS and BIS-11 (β = -0.04, P = 0.021; Figure 2).

Figure 2 The diagrams of chain mediation models in non-suicidal self-injury adolescents (n = 102). A: With anxiety, B: With depression as one of the chain mediators. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001.

Table 4 Chain mediating effects of depression (the Self-rating Depression Scale) or anxiety (the Self-rating Anxiety Scale) and impulsivity (the Barratt Impulsiveness Scale) in the associations between perceived social support (by the Multidimensional Scale of Perceived Social Support) and self-injury addiction (the Addiction Subscale of the Revised Chinese Version of Ottawa Self-injury Inventory) in adolescents with non-suicidal self-injury behaviors (n = 102).

Variables	Effects	SE	P value	BC bootstrap 95%CI		Effect proportion (%)	Power (%)
				Lower	Upper
Model A: MSPSS - SAS - BIS-11 - ROSI-addiction							20.9
The total effect of MSPSS - ROSI-addiction	-0.35	0.11	0.002	-0.88	-0.19	100.0
The direct effect of MSPSS - ROSI-addiction	-0.19	0.10	0.057	-0.39	<0.01	54.3
The indirect effect of
Ind1: MSPSS - SAS - ROSI-addiction	-0.11	0.05	0.007	-0.24	-0.03	31.4
Ind2: MSPSS - BIS-11 - ROSI-addiction	-0.03	0.04	0.300	-0.15	0.04	8.6
Ind3: MSPSS - SAS - BIS-11 - ROSI-addiction	-0.02	0.02	0.025	-0.10	> -0.01	5.7
Model B: MSPSS - SDS - BIS-11 - ROSI-addiction							38.6
The total effect of MSPSS - ROSI-addiction	-0.35	0.11	0.002	-0.56	-0.12	100.0
The direct effect of MSPSS - ROSI-addiction	-0.18	0.12	0.138	-0.43	0.06	51.4
The indirect effect of
Ind1: MSPSS - SDS - ROSI-addiction	-0.11	0.06	0.016	-0.26	-0.02	31.4
Ind2: MSPSS - BIS-11 - ROSI-addiction	-0.02	0.05	0.512	-0.13	0.06	5.7
Ind3: MSPSS - SDS - BIS-11 - ROSI-addiction	-0.04	0.03	0.021	-0.15	> -0.01	11.4

BC: Bias-corrected; 95%CI: 95% confidence interval; MSPSS: Multidimensional Scale of Perceived Social Support; SAS: Self-Rating Anxiety Scale; BIS-11: Barratt Impulsiveness Scale; ROSI-addiction: Addiction subscale of the Revised Chinese Version of Ottawa Self-injury Inventory; SDS: Self-Rating Depression Scale.

DISCUSSION

The study demonstrated that perceived social support, negative emotions and impulsivity are key factors influencing NSSI in adolescents. It further revealed that perceived social support exerted a protective effect against NSSI addiction through a serial mediating process involving depression/anxiety and impulsivity.

Sociodemographic findings showed that 91.2% of NSSI adolescents were girls, 26.5% didn't live with both parents, 31.4% reported disharmonious parental relationships, 37.3% were involved in bullying or were bullied, and depression was the main diagnosis. The data distribution was consistent with the previous meta-analysis that female gender, adverse childhood experiences, bullying, and mental disorders were risk factors for NSSI[37]. A recent psychiatric survey also found depressive disorder was the primary diagnosis associated with NSSI[6]. Although a recent meta-analysis has reported comparable rates of NSSI among male and female adolescents in Asia[38]. The present results showed a significantly higher rate among females, consistent with an epidemiological study in Chinese psychiatric patients[6].

Supporting the first hypothesis, SAS, SDS, most BIS-11 factors and ROSI-addiction were positively correlated indicating anxiety, depression, impulsivity and self-injury addiction are interrelated. The lack of a significant correlation between ROSI-addiction and BIS-11 cognitive instability may be because cognitive items load across multiple impulsivity factors[30]. The negative correlation between ROSI-addiction and MSPSS factors suggested that perceived social support was a protective factor for self-injury addiction in adolescents with NSSI. These findings were consistent with previous studies that behavioral and personality disorders and depression or anxiety were risk factors for self-harm, while good family and friend relationships were protective factors, and that repeated self-injury was associated with impulsivity, particularly attentional impulsivity[39,40]. Regarding other correlations, SAS and SDS were positively correlated with BIS-11 attention, perseverance, self-control, and cognitive complexity, consistent with previous findings that anxiety and depression were predictors of impulsivity in young people[24]. Moreover, SAS and SDS were negatively correlated with all MSPSS factors in our study, which can be explained by the fact that individuals with negative affect experience social impairments, receiving less social support and perceiving their support networks as weaker than those of healthy individuals[41].

In the serial mediation model, perceived social support was associated with reduced self-injury addiction through two separated mechanisms: One involving reduced anxiety that consequently lowered impulsivity, and the other through decreased depression, which in turn lessened impulsivity. These findings not only confirm the important role of perceived social support in mitigating NSSI, but also indicate distinct underlying mechanisms for anxiety and depression[42,43]. While previous studies often treat negative affect as a unified construct, this study delineates two separate emotional pathways and identifies that although they share impulsivity as a common downstream pathway, they differ in origin and manifestation: Anxiety appear to be more related to emotional arousal in response to anticipated threats, while depression is more closely associated with anhedonia and difficulties in emotion regulation[44]. The results also support prior research on the mediating role of impulsivity between depression and self-injury and extend its applicability to the anxiety pathway[45]. Unlike studies that mainly emphasized the mediating role of social support between emotion state and addictive behaviors, this model highlights the mediating function of emotional pathways between social support and self-injury, illustrating the complex interplay of risk factors contributing to NSSI[46].

The present findings should be interpreted in light of several methodological limitations. First, the cross-sectional design precludes causal inference regarding the relationships among variables. Second, although Harman’s single-factor test indicated that common method bias was not a serious issue, dependence on self-report instruments may still introduce potential bias. Third, the modest sample size resulted in low statistical power (20.9% and 38.6%) for the serial mediation models, thereby reducing generalizability and raising the possibility of overlooked mediation effects, especially those related to particular aspects of impulsivity. Additionally, the study treated impulsivity as a unitary construct and did not differentiate among its subdomains (e.g., attentional, motor, non-planning), which may exhibit distinct relationships within the proposed model. Despite these imitations, the observation of significant indirect effects under conditions of low statistical power suggests that the identified pathways are likely meaningful. Future research should employ longitudinal or experimental designs, incorporate multimodal assessments (e.g., behavioral tasks, clinical ratings), and recruit larger and more diverse samples to enhance validity and statistical robustness. In addition, future studies should utilize refined measures of impulsivity to disentangle the effects of its specific components. Finally, the findings underscore the importance of developing and testing targeted interventions (such as mindfulness-based, cognitive-behavioral, and dialectical behavior therapies) through rigorously designed randomized controlled trials.

CONCLUSION

The study demonstrated the serial mediating effects of anxiety/depression and impulsivity between perceived social support and self-injury addiction in adolescents with NSSI. These findings highlight the importance of interventions aimed at enhancing perceived social support, alongside the regulation of negative emotions and impulsivity, to reduce self-injury addiction in this population.

ACKNOWLEDGEMENTS

The authors wish to sincerely thank all of the staff participating this project. Most importantly, we wish to thank the young volunteers.