Next Article in Journal
How Do the Different Humor Styles of Streamers Affect Consumer Repurchase Intentions?
Previous Article in Journal
Listen or Read? The Impact of Proficiency and Visual Complexity on Learners’ Reliance on Captions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Behavioral Sciences
Volume 15
Issue 4
10.3390/bs15040543
behavsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Relationships Between Cumulative Family Risk, Emotion Regulation Difficulties, and Non-Suicidal Self-Injury in Adolescents: A Person-Centered Analysis

by
Xiaoxian Liu
,
Hengyuan Fan
,
Ruijuan Xiong
,
Lei An
,
Yiming Wang
,
Ruojuan Du
and
Xiaosheng Ding
*
Faculty of Education, Henan Normal University, Xinxiang 453007, China
*
Author to whom correspondence should be addressed.
Behav. Sci. 2025, 15(4), 543; https://doi.org/10.3390/bs15040543
Submission received: 15 February 2025 / Revised: 12 April 2025 / Accepted: 15 April 2025 / Published: 17 April 2025
Browse Figures
Review Reports Versions Notes

Abstract

:
The present study used a person-centered approach to examine the latent patterns of cumulative family risk and emotion regulation difficulties in adolescents and their relationships with non-suicidal self-injury. A sample of 1046 primary and secondary school students was analyzed using latent class analysis and latent profile analysis to identify subgroups of cumulative family risk and emotion regulation difficulties, respectively. The results were as follows: (1) Two latent classes of cumulative family risk were identified: a high-risk group (30.78%) and a low-risk group (69.22%). Adolescents in the high-risk group had significantly higher self-injury scores. (2) Three latent profiles of emotion regulation difficulties were identified: a low-difficulty group (56.02%), a medium-difficulty group (32.60%), and a high-difficulty group (11.38%). Adolescents in the high-difficulty group had the highest self-injury scores. (3) The logistic regression showed that adolescents in the high-risk group were more likely to belong to the high-difficulty group, followed by the medium- and low-difficulty groups. In summary, adolescents with high levels of cumulative family risk are also more likely to exhibit high levels of emotion regulation difficulties and self-injurious behavior.

1. Introduction

Non-suicidal self-injury (NSSI) refers to the deliberate, self-inflicted damage to body tissue without suicidal intent and for reasons not socially accepted, typically involving behaviors such as cutting or burning (Nock, 2010). NSSI is a significant mental health concern that negatively impacts the physical and psychological well-being of adolescents (Westers & Culyba, 2018) and is associated with an increased risk of suicidal ideation and behaviors (Klonsky et al., 2013).
The Systemic Model of Self-Injury suggests that adolescent NSSI arises from dysfunction in family or social environments (Crouch & Wright, 2004). Risk factors within the family environment, such as disrupted family structure, limited resources, and a lack of parent–child closeness, may impair the development of children’s emotion regulation abilities (Fombouchet et al., 2023). According to the Affective Cascade Model, emotional dysregulation and the accumulation of negative emotions are critical to the initiation and development of NSSI (Selby et al., 2008).
Numerous studies have examined the relationship between family factors and adolescent NSSI, with a particular focus on the influence of cumulative family risk (Pitkänen et al., 2019; Turner & Colburn, 2022; Manna et al., 2022). Previous studies have mostly relied on a variable-centered approach (e.g., regression analysis), which may overlook adolescents’ individual differences. Furthermore, studies on the relationship between family risk and NSSI in China have primarily focused on junior and senior high school students. However, evidence from international studies shows a high prevalence of NSSI in young adolescents, ranging from 7.6% to 23.7% (Barrocas et al., 2012; Tsypes et al., 2018). Early adolescence is characterized by rapid emotional, cognitive, and social transitions, and is recognized as a critical developmental period for the emergence of maladaptive behaviors, such as NSSI (Silvers, 2022). Identifying the factors associated with NSSI during this stage is essential for assessing both current and future risks and has important implications for prevention and early intervention efforts (De Luca et al., 2023). Given these limitations, we used a person-centered approach to examine the relationships between cumulative family risk, emotion regulation difficulties, and NSSI during early adolescence.

1.1. Cumulative Family Risk and Adolescent NSSI

Family risk encompasses factors within the family microsystem that may increase the likelihood of negative developmental outcomes (Bronfenbrenner & Morris, 2006). It typically includes three components (Y. L. Wang et al., 2024): (1) structure risk, such as parental divorce or parent–child separation; (2) resource risk, involving deficits in economic, educational, or material resource; and (3) family atmosphere risk, characterized by a lack of emotional support or parent–child closeness. Yates’ (2004) Developmental Psychopathology Model of Self-Injury suggests that family risk hinders the development of essential psychological capacities (e.g., emotional regulation ability), with NSSI then serving as a compensatory mechanism to manage psychological challenges. Numerous studies have shown that family risk factors, such as left-behind experiences, low parental education, disrupted family structure, family economic pressure, family conflict, and low parent–child closeness have been shown to be predictors of adolescent NSSI (J. M. Xiong et al., 2020; Manna et al., 2022; Y. L. Wang et al., 2024).
The Cumulative Risk Effect Theory suggests that while a single risk factor may not significantly impact individual development, the likelihood of adverse outcomes increases considerably when multiple risks are experienced simultaneously (Evans et al., 2013). The accumulation of multiple family risk factors may be a more accurate predictor of adolescent NSSI than a single risk factor (Buehler & Gerard, 2013). In the current study, we used the person-centered approach of exploratory latent class analysis (LCA) to identify latent patterns of cumulative family risk among adolescents and to examine differences in NSSI across these classes.

1.2. Emotion Regulation Difficulties and Adolescent NSSI

Emotion regulation difficulties involve challenges in identifying, understanding, accepting, and managing emotions (Gratz & Roemer, 2004). The Integrated Model of Self-Injury highlights emotion regulation as a key internal factor that predicts NSSI (Nock, 2009). The Affective Cascade Model posits that emotional dysregulation and the accumulation of negative emotions play significant roles in the initiation and maintenance of NSSI (Selby et al., 2008). Compared to non-injurers, individuals who engage in NSSI have been shown to experience greater challenges in regulating negative emotions (Wolff et al., 2019; Mettler et al., 2021; Duncan-Plummer et al., 2023).
According to the Multidimensional Model of Emotion Regulation Difficulties, there are six dimensions of emotion regulation: emotional response acceptance, goal-directed behavior, impulse control, emotional awareness, emotional clarity, and use of regulation strategies (Gratz & Roemer, 2004). Individual differences in these dimensions may form distinct latent patterns of emotion regulation difficulties (Chesney et al., 2019), each potentially linked to different associations with NSSI (Peterson et al., 2019). In the current study, we used the latent profile analysis (LPA), which is a variation in the latent class analysis used when the observed variables are continuous (Wen et al., 2021), to identify latent patterns of emotion regulation difficulties among adolescents and to examine differences in NSSI across these profiles.

1.3. Cumulative Family Risk and Emotion Regulation Difficulties

Evidence indicates that family risk significantly predicts children’s emotional insecurity and their use of emotion regulation strategies (Harold & Sellers, 2018), and emotion regulation difficulties may be more serious in children exposed to cumulative family risk factors (Gao & Han, 2016). Therefore, further exploration is needed to understand how different patterns of cumulative family risk affect adolescents’ emotion regulation difficulties. We used logistic regression to further examine the predictive role of latent classes of cumulative family risk on latent profiles of emotion regulation difficulty among adolescents.

1.4. The Present Study

First, we conducted LCA to identify latent patterns of cumulative family risk among adolescents and examined differences in NSSI across the identified classes. Second, LPA was conducted to identify latent patterns of emotion regulation difficulties in adolescents, followed by an analysis of the differences in NSSI across the identified profiles. Finally, logistic regression was used to assess latent classes of cumulative family risk as predictors of latent profiles of emotion regulation difficulty in adolescents.
The following hypotheses were proposed: (1) adolescents’ cumulative family risk can be categorized into distinct latent classes; (2) adolescents’ emotion regulation difficulties can be categorized into distinct latent profiles; (3) adolescents with higher levels of cumulative family risk will report higher levels of NSSI; (4) adolescents with higher levels of emotion regulation difficulties will report higher levels of NSSI; (5) adolescents in the higher-risk family group will be more likely to belong to latent profiles characterized by greater emotion regulation difficulties.
Person-centered analysis (e.g., LCA and LPA) identifies subgroups within a population that share similar characteristics. In contrast to variable-centered analysis, which examines average effects across all individuals, this method emphasizes individual heterogeneity (Bolck et al., 2004). This approach not only refines the precision of variable relationships but also offers insights for designing targeted interventions. Consequently, researchers can customize interventions to address the distinct needs of various subgroups effectively (Wen et al., 2021).

2. Materials and Methods

2.1. Participants

The sample included students from the 5th and 6th grade of an elementary school and from the 7th and 8th grade of a middle school in a city in Henan Province, China. A total of 1173 questionnaires were distributed using cluster sampling. Finally, 1046 valid responses were retained, yielding a response rate of 89.2%, comprising 330 students in the 5th grade (28.8%), 257 in the 6th grade (22.4%), 239 in the 7th grade (20.9%), and 220 students in the 8th grade (19.2%). Of these, 528 were boys (50.5%) and 518 were girls (49.5%).

2.2. Procedure

Before the data collection, the researchers communicated thoroughly with the school and obtained informed consent from school officials, parents, and students. The assessment was administered in class groups, with trained psychology graduate students serving as examiners. The assessment lasted 25 min, after which the examiners collected the questionnaires.
Given the sensitive nature of NSSI, several measures were implemented to minimize potential distress among participants. Prior to the commencement of the research, individuals were informed that the questionnaire would include items related to self-injury and emotional experiences. Participation was fully voluntary, and participants retained the right to withdraw at any point without consequence. If any participant experienced discomfort during the survey, the assessment was immediately discontinued, and psychological support was made available, including individual counseling and group-based interventions.

2.3. Measures

2.3.1. Adolescent Self-Injury Questionnaire

The Revised Adolescent Self-Injury Questionnaire (Feng, 2008), comprising 18 multiple-choice items. Self-injury was assessed by calculating the product of its frequency and severity, with higher scores reflecting higher levels of self-injury. The frequency is assessed using a 4-point scale (0 = never, 3 = often), while the severity is assessed using a 5-point scale (0 = no harm, 4 = very severe harm). A score of 1 or greater over the past year was used as the threshold for the presence of self-injury. In the present study, the construct validity of the scale was supported by confirmatory factor analysis (CFA), indicating an acceptable fit to the data (χ2(135) = 388.86, CFI = 0.93, TLI = 0.91, RMSEA = 0.06, SRMR = 0.05). The internal consistency reliability was satisfactory, with a Cronbach’s α of 0.81.

2.3.2. Cumulative Family Risk Questionnaire

The cumulative family risk was assessed in terms of six factors. Each factor was coded as present or absent (0 = no risk, 1 = at risk), and the cumulative family risk index was calculated by summing the risk scores (Buehler & Gerard, 2013).
(1) Left-behind experience: Codes as 1 if the participant reported that one or both parents had been away for work for more than six months; otherwise, it was coded as 0 (Y. L. Wang et al., 2024).
(2) Parental education level: Coded as 1 if the participant reported that both parents’ education levels were below high school; otherwise, it was coded as 0 (Y. L. Wang et al., 2024).
(3) Family structure: Coded as 1 if the participant reported they did not live with both their biological father and mother; otherwise, it was coded as 0 (J. M. Xiong et al., 2020).
(4) Family economic pressure: Assessed using the Family Economic Pressure Scale (J. P. Wang et al., 2010). Participants scoring in the top 25% were coded as 1, while others were coded as 0. In this study, the scale demonstrated a Cronbach’s α of 0.79.
(5) Parent–child closeness: Assessed using the Parent–Child Closeness Scale (Zhang et al., 2011). Participants scoring in the bottom 25% were coded as 1, while others were coded as 0. The father–child and mother–child subscales had Cronbach’s α of 0.87 and 0.85.
(6) Family conflict: Assessed using the Conflict Subscale of the Family Environment Scale (J. H. Tao et al., 2015). Participants scoring in the top 25% were coded as 1, while others were coded as 0. In this study, the scale demonstrated a Cronbach’s α of 0.65.
In line with previous research practices, risk factors were selected according to the following principles (S. Xiong et al., 2024): (1) relevance: risk factors that have been shown to be associated with adolescents’ emotion regulation difficulties and self-injurious behaviors were included; (2) representativeness: the selected risk factors reflect characteristics of the Chinese sociocultural context; (3) expansibility: the selected risk factors are closely aligned with the developmental characteristics of early adolescents; (4) uniqueness: there are no inclusion or overlap relationships between the selected risk factors.

2.3.3. Difficulties in Emotion Regulation Scale

The Difficulties in Emotion Regulation Scale (DERS) developed by Gratz and Roemer (2004) was used in this study. The Chinese version was revised and validated by L. Wang et al. (2007) among Chinese adolescents and has demonstrated good reliability and validity. It consists of 36 items on six dimensions: emotional response acceptance, goal-directed behavior, impulse control, emotional awareness, emotional clarity, and the use of regulation strategies. A 5-point Likert scale was used (1 = almost never, 5 = almost always), with higher scores indicating greater difficulties in emotion regulation. In the present study, the construct validity of the scale was supported by confirmatory factor analysis (CFA), indicating an acceptable fit to the data (χ2(584) = 1576.33, CFI = 0.91, TLI = 0.93, RMSEA = 0.07, SRMR = 0.06). The internal consistency reliability was satisfactory, with a Cronbach’s α of 0.89.

2.4. Data Analysis

Tests of common method bias, descriptive statistics, and correlation analyses were performed using SPSS 27.0. LCA and LPA were conducted using Mplus 8.3. We also utilized Mplus 8.3 to further examine the relationships between the identified latent classes, latent profiles, and adolescent NSSI through logistic regression and BCH method (Bolck, Croon, and Hagenaars method).
LCA and LPA were conducted using Mplus 8.3, and models with 2–5 classes were compared. Model evaluation criteria included the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), sample size-adjusted BIC (aBIC), entropy, the Lo–Mendell–Rubin adjusted likelihood ratio test (LMR-LRT), and the bootstrap likelihood ratio test (BLRT). Lower AIC, BIC, and aBIC values indicate better model fit. Entropy values closer to 1 reflect higher classification accuracy, with values above 0.8 suggesting over 90% accuracy (Nylund et al., 2007). Significant p-values for the LMR-LRT or BLRT indicate that a model with k classes fits significantly better than a model with k-1 classes (Scotto Rosato & Baer, 2012).

3. Results

3.1. Test of Common Method Bias

The Harman single-factor test identified 22 factors with eigenvalues greater than 1. The first factor accounted for 17.32% of the variance, below the critical threshold of 40% (Podsakoff et al., 2003). There was no serious common method bias in this study.

3.2. Descriptive Statistics and Correlations

Table 1 shows that cumulative family risk and NSSI were significantly positively correlated, and both were significantly positively correlated with emotion regulation difficulties (total score and all dimension scores). Among the dimensions of emotion regulation difficulties, all dimensions, except for emotional awareness, demonstrated significant positive correlations with each other.

3.3. Latent Classes of Cumulative Family Fisk

A latent class model was established, and models with two to five classes were compared. The fit indices are presented in Table 2. Entropy values for the 2-class, 4-class, and 5-class models all exceeded 0.80, indicating satisfactory accuracy. Starting from the 2-class model, the BIC and aBIC progressively increased with the addition of more classes, indicating a decline in model fit. Furthermore, from the 2-class model onward, the p-values for LMR-LRT and BLRT were no longer significant. The 2-class model was chosen as the final model for this study.
Table 3 and Figure 1 provide the information used to characterize the classes in the 2-class model. The conditional probabilities (probability of being at risk) for each family risk factor were significantly higher in the C1 group compared to the C2 group, with particularly notable differences in the dimensions of family conflict (χ2 = 134.80, p < 0.001), father–child closeness (χ2 = 476.33, p < 0.001), and mother–child closeness (χ2 = 923.84, p < 0.001). The C1 group was labeled the “high-risk group” (30.78%) and the C2 group was labeled the “low-risk group” (69.22%).
As shown in Table 4, the BCH method was conducted. The self-injury level in the high-risk group (M = 6.33) was significantly higher than that in the low-risk group (M = 1.33).

3.4. Latent Profiles of Emotion Regulation Difficulties

A latent profile model was established, and models with two to five profiles were compared. The model fit information is shown in Table 5. As the number of classes increased, the AIC, BIC, and aBIC indices gradually decreased, indicating improved model fit. The entropy values for both the 2-class and 3-class models reached 0.80, indicating satisfactory accuracy. The LMR-LRT and BLRT values for the 3-class model were significant, indicating the 3-class model provided a better fit than the 2-class model. The 3-class model was selected as the optimal model for this study.
As shown in Table 6 and Figure 2, within the 3-class model, the mean values of Group C3 were the highest in all dimensions except emotional awareness, followed by Group C2 and Group C1. Particularly notable differences among the three groups in the use of regulation strategies were observed (F = 1981.96, p < 0.001). Group C1 was labeled as the “low-difficulty group”, Group C2 as the “medium-difficulty group”, and Group C3 as the “high-difficulty group”.
As shown in Table 7, the BCH method was conducted. The self-injury level in the high-difficulty group (M = 13.22) was significantly higher than in the other two groups. Furthermore, the self-injury level was significantly higher in the medium-difficulty group (M = 3.06) than in the low-difficulty group (M = 0.65).

3.5. Logistic Regression with Identified Latent Classes as Predictors of Identified Latent Profiles

As shown in Table 8, the latent class of cumulative family risk served as the independent variable (the low-risk group as the reference group, coded as 0), and the latent profile of emotion regulation difficulties was the dependent variable. When the low-difficulty group was the reference group, participants in the high-risk group were more likely to be classified into the medium-difficulty group (OR = 1.72, p < 0.05) or the high-difficulty group (OR = 5.63, p < 0.01). When the medium-difficulty group was the reference group, participants in the high-risk group were more likely to be classified into the high-difficulty group (OR = 3.27, p < 0.01).

4. Discussion

This study employed a person-centered approach, using LCA and LPA to investigate latent patterns of cumulative family risk and emotion regulation difficulties among adolescents, as well as their association with NSSI.

4.1. Latent Classes of Cumulative Family Risk and Their Association with NSSI

The LCA revealed that adolescents’ cumulative family risk could be categorized into two groups: high-risk group (30.78%) and low-risk group (69.22%). The probabilities of various risks were significantly higher in the high-risk group compared to the low-risk group, with particularly significant differences in the dimensions of family conflict, father–child closeness, and mother–child closeness. This suggests that the family atmosphere is a core component of cumulative family risk.
Adolescents in the high-risk group had significantly higher NSSI than those in the low-risk group, indicating that higher levels of cumulative family risk are associated with increased NSSI (Pitkänen et al., 2019; Turner & Colburn, 2022; Manna et al., 2022). The Systemic Model of Self-Injury suggests that a negative family atmosphere, characterized by frequent conflict and poor parent–child relationships, may be a key influence on adolescent self-injury (Crouch & Wright, 2004). Family conflict undermines emotional regulation and significantly increases emotional insecurity and depressive symptoms, raising the risk of NSSI among adolescents (Harold & Sellers, 2018). The parent–child relationship provides adolescents with emotional support and guidance (Kiel & Kalomiris, 2015). Without a close parent–child relationship, the adolescent might use NSSI as a substitute emotion regulation strategy (Y. Tao et al., 2020). In summary, cumulative family risk, particularly risk due to a negative family atmosphere, is associated with a significant increase in the likelihood of adolescent NSSI.

4.2. Latent Profiles of Emotion Regulation Difficulties and Their Association with NSSI

The LPA revealed that adolescents’ emotion regulation difficulties could be categorized into three groups: low-difficulty group (56.02%), medium-difficulty group (32.60%), and high-difficulty group (11.38%). The differences among the three groups were significant in all dimensions except emotional awareness, with particularly notable differences in the use of regulation strategies.
The Emotion Dysregulation Model indicates that a lack of effective emotion regulation strategies is a key characteristic of individuals who engage in NSSI (Gratz & Roemer, 2008). Maladaptive strategies, such as expressive suppression, not only fail to regulate emotions effectively but may also intensify emotional distress, leading to the accumulation of negative emotions and resulting in NSSI (Selby et al., 2008). Interestingly, emotional awareness difficulties do not appear to be an important part of emotion regulation difficulties. Emotional awareness primarily entails recognizing the presence of emotions (Gratz & Roemer, 2004). Oliveira et al. (2024) found that the association between emotional awareness and emotional problems was more dependent on how individuals managed their emotions. Recent studies have highlighted the roles of emotional granularity and emotional acceptance as more central components of successful emotion regulation processes. Emotional granularity refers to the ability to precisely differentiate between emotional states, which facilitates more targeted and adaptive regulation strategies (Ip et al., 2024). Emotional acceptance involves a nonjudgmental attitude toward emotional experiences, allowing individuals to reduce secondary distress and respond more flexibly to emotional challenges (Kisley et al., 2024). Future research may benefit from explicitly incorporating these dimensions to better understand the emotion regulation profiles of adolescents.
Adolescents in the high-difficulty group had the highest self-injury scores, followed by those in the medium-difficulty group and the low-difficulty group, indicating that adolescents with greater emotion regulation difficulties also exhibit higher levels of NSSI (Duncan-Plummer et al., 2023). The emotion regulation ability is a critical internal factor influencing the occurrence of NSSI (Nock, 2009). Compared to non-self-injurers, self-injurers often demonstrate lower emotional recognition, less acceptance of emotional responses, more impulsive behaviors, less use of adaptive regulation strategies, and greater difficulty maintaining goal-directed behavior (Wolff et al., 2019; Mettler et al., 2021; Duncan-Plummer et al., 2023). In conclusion, emotion regulation difficulties, particularly the absence of effective strategies, appear to contribute to adolescent NSSI.

4.3. Prediction of Emotion Regulation Difficulties Profiles by Cumulative Family Risk Classes

Adolescents in the high-risk group were more likely to be classified into the high-difficulty group, followed by the medium- and low-difficulty groups, compared to those in the low-risk group. These findings align with previous research, indicating that the family environment plays a significant role in shaping adolescents’ emotion regulation abilities (Paley & Hajal, 2022) and that family risk factors are key contributors to emotion regulation difficulties (Gao & Han, 2016; Harold & Sellers, 2018). Without a supportive family environment, NSSI may serve as a maladaptive alternative for regulating emotions (Yates, 2004; Houben et al., 2017). In conclusion, when family risk leads to emotion regulation difficulties, NSSI may also increase.

4.4. Implications

From a theoretical perspective, this study contributes to the theoretical understanding of adolescent NSSI by employing a person-centered approach, offering a distinct perspective from traditional variable-centered methods. Through LCA and LPA, we identified subgroups of adolescents exhibiting different patterns of cumulative family risk and emotion regulation difficulties. The findings revealed that adolescents with higher levels of cumulative family risk were more likely to experience greater emotion regulation difficulties, which in turn were associated with higher levels of NSSI. These results highlight the critical roles of family risk and emotion regulation in the development of NSSI, aligning with key assumptions of the Systemic Model of Self-Injury (Crouch & Wright, 2004) and the Affective Cascade Model (Selby et al., 2008).
From a practical standpoint, the findings suggest that intervention efforts should target both the environmental and emotional factors associated with adolescent NSSI. Schools could play a central role by implementing social–emotional learning programs that enhance students’ abilities in emotion recognition, expression, and regulation (Silvers, 2022). In addition, schools should facilitate parents understanding of adolescent psychological development and impart skills for maintaining harmonious interactions through initiatives such as parent schools (Harold & Sellers, 2018). Furthermore, parent-focused interventions, including workshops on nonviolent communication and responsive parenting, may help reduce cumulative family risk and foster healthier parent–child relationships (Ladis et al., 2019). However, the implementation of these supports may face significant challenges in rural or under-resourced areas. To address these issues, low-cost, scalable interventions such as mobile-based psychoeducational programs and self-guided online courses should be considered (Hennemann et al., 2018). We believe leveraging existing school infrastructures and digital technologies could help bridge the service gap and extend emotional support to adolescents in need.

4.5. Limitations and Further Research

First, this study relied on self-reported measures. Future research could employ multi-source reporting to enhance result reliability. Second, cumulative family risk includes a broad range of factors. Beyond those examined in this study, future research could investigate the influence of additional family interpersonal factors (e.g., parental mental health) on adolescent NSSI. Third, this study used a cross-sectional design, offering only the patterns of adolescents’ current emotion regulation difficulties. Future studies could utilize longitudinal designs (e.g., latent transition analysis) to explore the dynamic trends of emotion regulation ability throughout adolescence.

5. Conclusions

(1)
Adolescents can be classified into two latent classes of cumulative family risk: high-risk (30.78%) and low-risk (69.22%). Adolescents in the high-risk group had significantly higher self-injury scores than those in the low-risk group.
(2)
Adolescents exhibit three latent profiles of emotion regulation difficulties: low-difficulty (56.02%), medium-difficulty (32.60%), and high-difficulty (11.38%). Adolescents in the high-difficulty group had the highest self-injury scores, followed by those in the medium-difficulty group and the low-difficulty group.
(3)
Cumulative family risk latent classes predict emotion regulation difficulty latent profiles. Adolescents in the high-risk group were more likely to be classified into the high-difficulty group, followed by the medium- and low-difficulty groups, compared to those in the low-risk group.

Author Contributions

Conceptualization, X.L. and X.D.; methodology, X.L. and R.X.; software, H.F. and R.X.; validation, L.A.; formal analysis, H.F. and R.X.; investigation, L.A., Y.W. and R.D.; resources, X.D.; data curation, L.A., Y.W. and R.D.; writing—original draft preparation, R.X. and H.F.; writing—review and editing, H.F. and X.L.; visualization, H.F. and X.L.; supervision, X.L. and X.D.; project administration, X.D.; funding acquisition, X.L. and X.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (Grant No. 72174055) and the Henan Province Philosophy and Social Science Research Plan (Grant No. 2022BJY015).

Institutional Review Board Statement

This study was approved by the Institutional Review Board of the Henan Provincial Key Laboratory of Psychology and Behavior (approval code 20240323001) on 23 March 2024.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Acknowledgments

The authors would like to thank the primary school for hosting them and all of their participants.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Barrocas, A. L., Hankin, B. L., Young, J. F., & Abela, J. R. (2012). Rates of nonsuicidal self-injury in youth: Age, sex, and behavioral methods in a community sample. Pediatrics, 130(1), 39–45. [Google Scholar] [CrossRef]
  2. Bolck, A., Croon, M., & Hagenaars, J. (2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis, 12(1), 3–27. [Google Scholar] [CrossRef]
  3. Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model of human development. In W. Damon, & R. M. Lerner (Eds.), Handbook of child psychology (6th ed., pp. 793–828). Wiley. [Google Scholar]
  4. Buehler, C., & Gerard, J. M. (2013). Cumulative family risk predicts increases in adjustment difficulties across early adolescence. Journal of Youth & Adolescence, 42(6), 905–920. [Google Scholar] [CrossRef]
  5. Chesney, S. A., Timmer-Murillo, S. C., & Gordon, N. S. (2019). Establishment andreplication of emotion regulation profiles: Implications for psychologicalhealth. Anxiety, Stress, & Coping, 32(3), 329–345. [Google Scholar] [CrossRef]
  6. Crouch, W., & Wright, J. (2004). Deliberate self-harm at an adolescent unit: A qualitative investigation. Clinical Child Psychology and Psychiatry, 9(2), 185–204. [Google Scholar] [CrossRef]
  7. De Luca, L., Pastore, M., Palladino, B. E., Reime, B., Warth, P., & Menesini, E. (2023). The development of Non-Suicidal Self-Injury (NSSI) during adolescence: A systematic review and Bayesian meta-analysis. Journal of Affective Disorders, 339, 648–659. [Google Scholar] [CrossRef]
  8. Duncan-Plummer, T., Hasking, P., Tonta, K., & Boyes, M. (2023). Cognitive-emotional networks in students with and without a history of non-suicidal self-injury. Journal of Affective Disorders, 329, 394–403. [Google Scholar] [CrossRef] [PubMed]
  9. Evans, G. W., Li, D., & Whipple, S. S. (2013). Cumulative risk and child development. Psychological Bulletin, 139(6), 1342. [Google Scholar] [CrossRef]
  10. Feng, Y. (2008). The relationship between adolescent self-injurious behavior and individual emotional factors and family environmental factors [Master’s Thesis, Central China Normal University]. [Google Scholar]
  11. Fombouchet, Y., Pineau, S., Perchec, C., Lucenet, J., & Lannegrand, L. (2023). The development of emotion regulation in adolescence: What do we know and where to go next? Social Development, 32(4), 1227–1242. [Google Scholar] [CrossRef]
  12. Gao, M., & Han, Z. R. (2016). Family expressiveness mediates the relation between cumulative family risks and children’s emotion regulation in a Chinese sample. Journal of Child and Family Studies, 25, 1570–1580. [Google Scholar] [CrossRef]
  13. Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26, 41–54. [Google Scholar] [CrossRef]
  14. Gratz, K. L., & Roemer, L. (2008). The relationship between emotion dysregulation and deliberate self-harm among female undergraduate students at an urban commuter university. Cognitive Behaviour Therapy, 37(1), 14–25. [Google Scholar] [CrossRef] [PubMed]
  15. Harold, G. T., & Sellers, R. (2018). Annual research review: Interparental conflict and youth psychopathology: An evidence review and practice focused update. Child Psychology and Psychiatry, 59(4), 374–402. [Google Scholar] [CrossRef] [PubMed]
  16. Hennemann, S., Farnsteiner, S., & Sander, L. (2018). Internet-and mobile-based aftercare and relapse prevention in mental disorders: A systematic review and recommendations for future research. Internet Interventions, 14, 1–17. [Google Scholar] [CrossRef] [PubMed]
  17. Houben, M., Claes, L., Vansteelandt, K., Berens, A., Sleuwaegen, E., & Kuppens, P. (2017). The emotion regulation function of nonsuicidal self-injury: A momentary assessment study in inpatients with borderline personality disorder features. Journal of Abnormal Psychology, 126(1), 89. [Google Scholar] [CrossRef]
  18. Ip, K. I., Yu, K., & Gendron, M. (2024). Emotion granularity, regulation, and their implications in health: Broadening the scope from a cultural and developmental perspective. Emotion Review, 16(4), 224–237. [Google Scholar] [CrossRef]
  19. Kiel, E. J., & Kalomiris, A. E. (2015). Current themes in understanding children’s emotion regulation as developing from within the parent–child relationship. Current Opinion in Psychology, 3, 11–16. [Google Scholar] [CrossRef]
  20. Kisley, M. A., Beblo, T., & Lac, A. (2024). Emotion acceptance questionnaire (EAQ): Factor analysis and psychometric evaluation. Journal of Personality Assessment, 107, 1–14. [Google Scholar] [CrossRef]
  21. Klonsky, E. D., May, A. M., & Glenn, C. R. (2013). The relationship between nonsuicidal self-injury and attempted suicide: Converging evidence from four samples. Journal of Abnormal Psychology, 122(1), 231–237. [Google Scholar] [CrossRef]
  22. Ladis, B. A., Macgowan, M., Thomlison, B., Fava, N. M., Huang, H., Trucco, E. M., & Martinez, M. J. (2019). Parent-focused preventive interventions for youth substance use and problem behaviors: A systematic review. Research on Social Work Practice, 29(4), 420–442. [Google Scholar] [CrossRef]
  23. Manna, G., Falgares, G., Costanzo, G., La Sala, M. C., Lamis, D. A., & Musso, P. (2022). Cumulative childhood maltreatment and non-suicidal self-injury: The mediating and moderating role of perceived social support in a sample of university students. Journal of Family Violence, 37(4), 657–669. [Google Scholar] [CrossRef]
  24. Mettler, J., Stern, M., Lewis, S. P., & Heath, N. L. (2021). Perceived vs. actual emotion reactivity and regulation in individuals with and without a history of NSSI. Frontiers in Psychology, 12, 612792. [Google Scholar] [CrossRef]
  25. Nock, M. K. (2009). Why do people hurt themselves? New insights into the nature and functions of self-injury. Current Directions in Psychological Science, 18(2), 78–83. [Google Scholar] [CrossRef]
  26. Nock, M. K. (2010). Self-injury. Annual Review of Clinical Psychology, 6(1), 339–363. [Google Scholar] [CrossRef]
  27. Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 535–569. [Google Scholar] [CrossRef]
  28. Oliveira, J., Pedras, S., Inman, R. A., & Ramalho, S. M. (2024). Latent profiles of emotion regulation among university students: Links to repetitive negative thinking, internet addiction, and subjective well-being. Frontiers in Psychology, 15, 1272643. [Google Scholar] [CrossRef]
  29. Paley, B., & Hajal, N. J. (2022). Conceptualizing emotion regulation and coregulation as family-level phenomena. Clinical Child and Family Psychology Review, 25(1), 19–43. [Google Scholar] [CrossRef] [PubMed]
  30. Peterson, A. L., Chen, J. I., Karver, M. S., & Labouliere, C. D. (2019). Frustrationwith feeling: Latent classes of non-suicidal self-injury and emotion regulationdifficulties. Psychiatry Research, 275, 61–70. [Google Scholar] [CrossRef] [PubMed]
  31. Pitkänen, J., Remes, H., Aaltonen, M., & Martikainen, P. (2019). Experience of maternal and paternal adversities in childhood as determinants of self-harm in adolescence and young adulthood. Journal of Epidemiol & Community Health, 73(11), 1040–1046. [Google Scholar] [CrossRef]
  32. Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879. [Google Scholar] [CrossRef]
  33. Scotto Rosato, N., & Baer, J. C. (2012). Latent class analysis: A method for capturing heterogeneity. Social Work Research, 36(1), 61–69. [Google Scholar] [CrossRef]
  34. Selby, E. A., Anestis, M. D., & Joiner, T. E. (2008). Understanding the relationship between emotional and behavioral dysregulation: Emotional cascades. Behaviour Research and Therapy, 46(5), 593–611. [Google Scholar] [CrossRef]
  35. Silvers, J. A. (2022). Adolescence as a pivotal period for emotion regulation development. Current Opinion in Psychology, 44, 258–263. [Google Scholar] [CrossRef] [PubMed]
  36. Tao, J. H., Jin, F. X., Zhang, M. R., & Cheng, Z. H. (2015). Reliability and validity of FES-CV in the adolescent population. Chinese Journal of Clinical Psychology, 23(6), 1024–1027. [Google Scholar] [CrossRef]
  37. Tao, Y., Bi, X. Y., & Deng, M. (2020). The impact of parent–child attachment on self-injury behavior: Negative emotion and emotional coping style as serial mediators. Frontiers in Psychology, 11, 1477. [Google Scholar] [CrossRef]
  38. Tsypes, A., Owens, M., Hajcak, G., & Gibb, B. E. (2018). Neural reward responsiveness in children who engage in nonsuicidal self-injury: An ERP study. Journal of Child Psychology and Psychiatry, 59(12), 1289–1297. [Google Scholar] [CrossRef]
  39. Turner, H. A., & Colburn, D. (2022). Independent and cumulative effects of recent maltreatment on suicidal ideation and thoughts of self-harm in a national sample of youth. Journal of Adolescent Health, 70(2), 329–335. [Google Scholar] [CrossRef]
  40. Wang, J. P., Li, D. P., & Zhang, W. (2010). Adolescences’s family financial difficulty and social adaptation: Coping efficacy of compensatory, mediation, and moderation effects. Journal of Beijing Normal University (Social Sciences), 4, 22–32. [Google Scholar] [CrossRef]
  41. Wang, L., Lu, Y. P., & Li, Z. Q. (2007). Test of emotion regulation scale in adolescents. Chinese Journal of Clinical Psychology, 15(3), 236–238. [Google Scholar] [CrossRef]
  42. Wang, Y. L., Zhao, J. F., & Lin, X. Y. (2024). Cumulative effects of family risk factors on adolescent non-suicidal self-injury and its gender difference. Psychological Development and Education, 40(2), 240–247. [Google Scholar] [CrossRef]
  43. Wen, Z. L., Fang, J., Shen, J. Q., Tan, Y. T., Li, D. X., & Ma, Y. M. (2021). A review of research on psychological statistics in China’s mainland from 2001 to 2020. Advances in Psychological Science, 29(8), 1331. [Google Scholar] [CrossRef]
  44. Westers, N. J., & Culyba, A. J. (2018). Nonsuicidal self-injury: A neglected public health problem among adolescents. American Journal of Public Health, 108(8), 981–983. [Google Scholar] [CrossRef]
  45. Wolff, J. C., Thompson, E., Thomas, S. A., Nesi, J., Bettis, A. H., Ransford, B., Scopelliti, K., Frazier, E. A., & Liu, R. T. (2019). Emotion dysregulation and non-suicidal self-injury: A systematic review and meta-analysis. European Psychiatry, 59, 25–36. [Google Scholar] [CrossRef] [PubMed]
  46. Xiong, J. M., Hai, M., Huang, F., Xin, L., & Xu, Y. (2020). Family cumulative risk and mental health in Chinese adolescents: The compensatory and moderating effects of psychological capital. Psychological Development and Education, 36(1), 94–102. [Google Scholar] [CrossRef]
  47. Xiong, S., Xu, Y., Zhang, B., & Chen, Y. (2024). Gender differences in the longitudinal association between cumulative ecological risk and smartphone dependence among early adolescents: A parallel mediation model. Computers in Human Behavior, 151, 108044. [Google Scholar] [CrossRef]
  48. Yates, T. M. (2004). The developmental psychopathology of self-injurious behavior: Compensatory regulation in posttraumatic adaptation. Clinical Psychology Review, 24(1), 35–74. [Google Scholar] [CrossRef]
  49. Zhang, J. T., Liu, Q. X., Deng, L. Y., Fang, X. Y., Liu, C. Y., & Lan, J. (2011). Parents-adolescents relations and adolescent’s internet addiciton: The mediaiton effect of loneliness. Psychological Development and Education, 27(6), 641–647. [Google Scholar] [CrossRef]
Behavsci 15 00543 g001
Figure 1. The conditional probabilities of each family risk factor across family risk groups.
Figure 1. The conditional probabilities of each family risk factor across family risk groups.
Behavsci 15 00543 g001
Behavsci 15 00543 g002
Figure 2. Differences in the means across emotion regulation difficulties groups.
Figure 2. Differences in the means across emotion regulation difficulties groups.
Behavsci 15 00543 g002
Table 1. The mean (M), standard deviation (SD), and correlations of the main variables.
Table 1. The mean (M), standard deviation (SD), and correlations of the main variables.
123456789
1. Emotional response acceptance1
2. Goal-directed behavior0.44 ***1
3. Impulse control0.55 ***0.54 ***1
4. Emotional awareness−0.11 ***0.030.051
5. Use of regulation strategies0.62 ***0.62 ***0.71 ***0.041
6. Emotional clarity0.34 ***0.29 ***0.37 ***0.41 ***0.43 ***1
7. NSSI0.31 ***0.25 ***0.40 ***0.07 *0.48 ***0.25 ***1
8. Emotion regulation difficulties0.70 ***0.72 ***0.80 ***0.33 ***0.87 ***0.65 ***0.44 ***1
9. Cumulative family risk index0.18 ***0.23 ***0.25 ***0.13 ***0.30 ***0.22 ***0.32 ***0.32 ***1
M12.4512.2812.2716.2816.0910.942.8980.321.6
SD5.154.715.055.246.583.787.4920.921.47
Note: * means p < 0.05, *** means p < 0.001.
Table 2. Fit indices for models derived from the LCA of cumulative family risk factors.
Table 2. Fit indices for models derived from the LCA of cumulative family risk factors.
ModelAICBICaBICEntropyLMR-LRT(p)BLRT(p)Size (% of Sample)
17021.987056.657034.41----
26359.806434.096386.450.83<0.001<0.00130.78/69.22
36357.776471.686398.630.740.0630.0978.41/23.52/68.07
46361.806515.336416.870.800.0940.37526.48/4.21/68.07/1.24
56368.766561.916438.040.840.0520.9730.96/5.64/24.47/67.11/1.82
Table 3. Chi-square tests of conditional probabilities across family risk groups.
Table 3. Chi-square tests of conditional probabilities across family risk groups.
High-Risk GroupLow-Risk Groupχ2 (df = 1)
At Risk(1)No Risk(0)At Risk(1)No Risk(0)
Left-behind experience0.090.910169.45 ***
Parental education level0.40.60.280.7216.77 ***
Family structure0.090.910171.83 ***
Family economic pressure0.510.490.240.7668.06 ***
Family conflict0.560.440.210.79134.80 ***
Father-child closeness0.780.220.080.92476.33 ***
Mother-child closeness0.860.140.020.98923.84 ***
Note: *** means p < 0.001.
Table 4. Differences in NSSI across cumulative family risk groups.
Table 4. Differences in NSSI across cumulative family risk groups.
High-Risk Group(M ± SE)Low-Risk Group(M ± SE)χ2 (df = 1)
NSSI6.33(0.65)1.33(0.19)50.91 ***
Note: *** means p < 0.001.
Table 5. Fit indices for models derived from the LPA of emotion regulation difficulties.
Table 5. Fit indices for models derived from the LPA of emotion regulation difficulties.
ModelAICBICaBICEntropyLMR-LRT(p)BLRT(p)Size (% of Sample)
138,076.1338,135.5738,097.45----
236,425.6236,519.7336,459.380.88<0.001<0.00170.08/29.92
335,940.3936,069.1735,986.590.86<0.001<0.00156.02/32.60/11.38
435,747.8635,911.3035,806.490.780.232<0.00127.44/8.70/40.34/23.52
535,627.5435,825.6535,698.610.750.202<0.00126.10/20.08/25.14/21.13/7.55
Table 6. Analyses of variance of means across emotion regulation difficulties groups.
Table 6. Analyses of variance of means across emotion regulation difficulties groups.
C1(M ± SE)C2(M ± SE)C3(M ± SE)FComparison
Emotional response acceptance2.51(0.07)3.94(0.16)5.04(0.22)462.88 ***C3 > C2 > C1
Goal-directed behavior2.67(0.07)4.01(0.12)4.87(0.17)360.84 ***C3 > C2 > C1
Impulse control2.82(0.10)4.31(0.22)6.30(0.38)711.27 ***C3 > C2 > C1
Emotional awareness3.08(0.10)3.18(0.12)3.07(0.13)0.55No significant differences
Use of regulation strategies3.68(0.12)5.97(0.23)9.21(0.31)1981.96 ***C3 > C2 > C1
Emotional clarity2.81(0.09)3.58(0.14)4.14(0.18)120.32 ***C3 > C2 > C1
Note: C1 = low-difficulty group, C2 = medium-difficulty group, C3 = high-difficulty group, *** means p < 0.001.
Table 7. Differences in NSSI across emotion regulation difficulties groups.
Table 7. Differences in NSSI across emotion regulation difficulties groups.
M ± SEχ2 (df = 2)
low-difficulty group0.65(0.13)low-difficulty group vs. medium-difficulty group29.93 ***
low-difficulty group vs. high-difficulty group76.63 ***
medium-difficulty group3.06(0.40)medium-difficulty group vs. high-difficulty group44.02 ***
high-difficulty group13.22(1.43)--
Note: *** means p < 0.001.
Table 8. The prediction of cumulative family risk latent classes on emotion regulation difficulties latent profiles.
Table 8. The prediction of cumulative family risk latent classes on emotion regulation difficulties latent profiles.
REF: Low-Difficulty GroupREF: Medium-Difficulty Group
Medium-Difficulty GroupHigh-Difficulty GroupHigh-Difficulty Group
High-risk group
(coded as 1)		ORCI(95%)ORCI(95%)ORCI(95%)
1.72 *[1.12, 2.64]5.63 **[3.52, 9.01]3.27 **[1.94, 5.50]
Note: * means p < 0.05, ** means p < 0.01.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Liu, X.; Fan, H.; Xiong, R.; An, L.; Wang, Y.; Du, R.; Ding, X. Relationships Between Cumulative Family Risk, Emotion Regulation Difficulties, and Non-Suicidal Self-Injury in Adolescents: A Person-Centered Analysis. Behav. Sci. 2025, 15, 543. https://doi.org/10.3390/bs15040543

AMA Style

Liu X, Fan H, Xiong R, An L, Wang Y, Du R, Ding X. Relationships Between Cumulative Family Risk, Emotion Regulation Difficulties, and Non-Suicidal Self-Injury in Adolescents: A Person-Centered Analysis. Behavioral Sciences. 2025; 15(4):543. https://doi.org/10.3390/bs15040543

Chicago/Turabian Style

Liu, Xiaoxian, Hengyuan Fan, Ruijuan Xiong, Lei An, Yiming Wang, Ruojuan Du, and Xiaosheng Ding. 2025. "Relationships Between Cumulative Family Risk, Emotion Regulation Difficulties, and Non-Suicidal Self-Injury in Adolescents: A Person-Centered Analysis" Behavioral Sciences 15, no. 4: 543. https://doi.org/10.3390/bs15040543

APA Style

Liu, X., Fan, H., Xiong, R., An, L., Wang, Y., Du, R., & Ding, X. (2025). Relationships Between Cumulative Family Risk, Emotion Regulation Difficulties, and Non-Suicidal Self-Injury in Adolescents: A Person-Centered Analysis. Behavioral Sciences, 15(4), 543. https://doi.org/10.3390/bs15040543

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop