A longitudinal study on problematic smartphone use and Non-Suicidal Self-Injury: the mediating role of academic procrastination and anxiety

Abstract

Problematic smartphone use is a possible risk factor for non-suicidal self-injury. However, little research explored the mechanisms underlying the association between problematic smartphone use and non-suicidal self-injury. We hypothesized that academic procrastination and anxiety may mediate the relationship between problematic smartphone use and non-suicidal self-injury. A total of 965 Chinese adolescents (492 males and 473 females, mean age = 15.96 years, SD = 0.47 years) completed questionnaires regarding problematic smartphone use, academic procrastination, anxiety and non-suicidal self-injury. The findings were as follows: (1) anxiety mediated the relation between problematic smartphone use and non-suicidal self-injury. (2) Academic procrastination and anxiety sequentially mediated the relation between problematic smartphone use and non-suicidal self-injury. The result reveals that problematic smartphone use has indirect effect on non-suicidal self-injury, which is mediated by academic procrastination and anxiety. The results offer valuable insights into the connections between problematic smartphone use and non-suicidal self-injury, highlighting the important role of anxiety in this dynamic. We can prevent maladaptive coping styles causing NSSI by improving the ability to regulate emotions.

Introduction

Non-suicidal self-injury (NSSI) is defined as behavior in which an individual intentionally damages their body without suicidal intent and for purpose not sanctioned by society [1, 2]. NSSI happens in many methods, such as hitting, burning, cutting, scratching the skin, and interfering with wound healing [3]. NSSI is a widely concerned psychosocial and cultural phenomenon, meanwhile, the data show an increasing trend in modern societies [4]. One meta-analysis reported a detection rate of 27.4% in the middle school student population in mainland China [5], which proved a high prevalence of NSSI behaviors among adolescents in China. NSSI has a negative impact on family communication and interpersonal relationships [6, 7], and is associated with mental illness such as depression [8], anxiety [9], and suicide attempts [10]. The growing researches has indicated the importance of understanding the factors that may be associated with NSSI, especially in adolescents. Thus, there is a theoretical and practical necessity to explore the influencing factors of NSSI in Chinese adolescents.

The purpose of the current study is to explore the predictors of NSSI behavior. NSSI is often engaged in as a coping behaviour and is often used to regulate painful emotional states [11]. According to a recent review of influencing factors exploring the NSSI, a significant behavioural issue among the 80 risk factors was found to be problematic behaviours, such as problematic smartphone use [12]. And excessive use of internet or smartphone was significantly associated with increased risk of NSSI [13]. Despite increasing early smartphone use and growing research on problematic use (PSU) and non-suicidal self-injury (NSSI), most studies focus narrowly on their direct relationship, overlooking deeper mediating factors.

Problematic smartphone use and non-suicidal self-injury

The term “problematic smartphone use (PSU)” denotes the lack of control or regulation over smartphone use, and it is associated with maladaptive problems, excessive use, loss of control, and functional impairment [14, 15]. A large-scale study in China demonstrated that PSU was significantly associated with NSSI among adolescents, and this relationship was stronger in females [16]. Although there has been a clear relationship between PSU and NSSI, these studies have limited empirical understanding of the psychological mechanisms involved. Some researchers have suggested that the effect of PSU on NSSI may be realised through a number of mediating variables [17]. Fewer studies have directly explored the relationship between PSU and NSSI, but several studies examining the relationship between high intensity smartphone use and self-injurious behavior have revealed a strong predictive effect [13]. Furthermore, PSU was found to play a prominent role in increasing the risk of NSSI among Chinese adolescents in a recent large-scale study in China [16]. Moreover, PSU may lead to a variety of negative consequences, including depression [18, 19], sleep problems [20], and suicidal attempts [21]. Although many studies only explore the relationship between PSU, anxiety [22], depression [23], mental health problem [24], and suicidal behavior [25], few studies have directly explored the relationship between the PSU and NSSI. This suggests that PSU may be highly correlated with dysphoria, resulting in self-injurious behaviors. Self-control theory is a psychological framework that explains how individuals manage to prioritize long-term goals over immediate urges or temptations. It posits that self-control involves a dual-system model, where the “cool” system, which is reflective and rational, overrides the impulses of the “hot” system, which is impulsive and automatic [26]. According to self-control theory, adolescents have lacked self-control due to their addiction to smartphones, while one undesirable way to compensate for self-control is self-injurious behavior. Thus, we hypothesize that :

H1. PSU may be a significant risk factor for increased NSSI in adolescents.

Academic procrastination as a mediator

Academic procrastination has been described as a trait-like characteristic of habitually postponing learning-related activities [27]. Although few studies have directly examined the relationship between academic procrastination and NSSI, empirical studies have found that academic procrastination can lead to emotional problems such as depression [28], physical problems such as palpitations and nausea, as well as sleep problems [29], which are highly predictive of NSSI [30, 31].

Previous studies have demonstrated that PSU directly predicts students’ academic procrastination [15, 32]. The more frequency of smartphone use among college students, the more it is likely to lead to poor academic performance. For instance, Sapci et al. [33] demonstrated a negative association between smartphone use and academic performance among college students using Apple’s Screen Time data. Based on the above literature review, we can conclude that PSU has a predictive effect on academic procrastination. Thus we hypothesize :

H2. Academic procrastination mediates the relationship between PSU and NSSI.

Anxiety as a mediator

Anxiety was defined as the disruptive feelings of uncertainty, dread, and fearfulness [34]. Researchers demonstrated that people who had anxiety symptoms reported an increased risk of NSSI [35]. People with more anxiety symptoms were more likely to adopt a maladaptive coping behaviour to reduce their suffering, such as NSSI [35, 36].

A large number of studies suggested PSU was a predictive factor of anxiety [37–39]. For instance, Hartanto and Yang [40] investigated whether separation from a smartphone affects state anxiety, and found that separation from a smartphone increases anxiety, which in turn mediates the negative effect of smartphone separation. In addition, the researchers found that individuals with more serious PSU problems had more anxiety over time [41]. Cheever et al. [41] also concluded that smartphone dependence, coupled with an unhealthy attachment to continued use, can lead to increased anxiety without a smartphone. Therefore, the present study hypothesizes:

H3. Anxiety has a mediating role between PSU and NSSI.

Academic procrastination and anxiety

Academic procrastination is associated with poorer academic performance and an increased risk of mental health problems such as anxiety, depression and stress [42]. According to self-control theory, previous research has primarily focused on academic procrastination as a way to compensate for negative emotions and as a manifestation of self-regulation failure [43, 44]. However, recent studies have indicated that academic procrastination can also predict negative affect [12]. For example, people with higher levels of academic procrastination reported significantly higher anxiety [12]. Anxiety may play a mediating role among PSU, academic procrastination and NSSI. Given that, we hypothesize:

H4. Academic procrastination and anxiety have a sequential mediation effect on the relation between PSU and NSSI.

Current study

In summary, many empirical studies have focused on exploring predictors of NSSI in adolescents [30, 31]. Thereby, we adopted a longitudinal study and conducted two waves of data collection for adolescents in this study to test the influencing mechanism of PSU on NSSI. This study aimed to examine the meditative roles of academic procrastination and anxiety between PSU and NSSI, to deepen our understanding about how PSU is associated with NSSI and provide valuable insights for NSSI interventions.

Methods

Participants and procedure

Adolescents from two public senior high schools in Shandong Provinces in China took part in this study. Before investigation, this study was approved by the local ethics committee of the participating schools and all adolescents, teachers, and parents involved in this study provided informed consent. We collected two-wave data in December 2020 (Time 1) and June 2021 (Time 2). At Time 1, 1029 students completed the Smartphone Application-Based Addiction Scale. At Time 2, 965 of them completed the anxiety sub-scale of the Depression Anxiety Stress Scales-21, the Tuckman’s Academic Procrastination Scale, and the non-suicidal self-injury list. The final sample included 965 students. Results of t-test revealed that students who dropped out of the study after the first assessment (64 students) were not significantly different from the analyzed sample on the problematic smartphone use used at Time 1, t = 1.83, p > .05. Reflecting the demographics, 51.0% of the students were males and 49.0% were females, 32.7% were only children and 67.3% were non-only children, 48.8% came from villages or towns and 51.1% came from urban areas (2 students did not report the local demographics), and their ages ranged from 14 to 17 years at Time 2 (M = 15.96 years, SD = 0.47 years, 18 students did not report their ages).

Measures

Problematic smartphone use. Problematic smartphone use was assessed using the Smartphone Application-Based Addiction Scale developed by Csibi et al. [45]. This 6-item scale is rated using a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree. An example item is “I fiddle around more and more with my smartphone”. The problematic smartphone use score was obtained by summing the six item scores, with higher scores indicating higher levels of problematic smartphone use. It has been shown good reliability and validity among Chinese participants [46]. In this study, the Cronbach’s α for this scale was 0.846.

Academic procrastination. We assessed adolescents’ academic procrastination using Tuckman’s Academic Procrastination Scale [47]. This 16-item scale is rated using a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree. An example item is “I needlessly delay finishing jobs, even when they’re important”. The final score was calculated by summing the 16 item scares, with higher score indicating more academic procrastination behaviors. This scale has shown good reliability and validity in a recent study in China [32]. In this study, Cronbach’s α was 0.922.

Anxiety. Adolescents’ levels of anxiety was measured by the anxiety sub-scale of the Depression Anxiety Stress Scales-21 [48], which has demonstrated good reliability and validity in Chinese adolescent populations [49]. Participants rated 7 items in terms of how often they experience them in the past six months, using a 5-point scale (1 = never, 5 = always). An example item is “I have a dry mouth”. Scores of anxiety were calculated by summing the 7 item scores, with higher scores indicating stronger psychological distress. In this study, the Cronbach’s α for this scale was 0.813.

Non-suicidal self-injury. Non-suicidal self-injury was accessed by the non-suicidal self-injury list which has good reliability and validity in Chinese samples [50, 51]. It included 7 self-injury behaviors, namely self-cutting, burning, biting, punching, scratching skin, inserting sharp objects in the nail or skin, and banging the head or other parts of the body against the wall. Adolescents were asked to indicate the frequency of engagement in the following behaviors (e.g., self-cutting) to deliberately harm themselves but without suicidal intent in the past six months. Each item is rated using a 4-point Likert scale (1 = never, 4 = six times or more). The higher total scores indicate more frequent non-suicidal self-injury. In this study, the Cronbach’s α for this scale was 0.779.

Statistical analysis

Data from two waves were analyzed using SPSS 22.0. Missing values (< 0.1%) were imputed using the mean for each respective item. Outliers were identified using z-scores and replaced with the item mean; however, alternative methods were considered to ensure that this approach did not distort the data distribution.

To assess common method bias, we conducted a Harman’s single-factor test. The first factor accounted for 27.55% of the variance, suggesting that common method bias is not a major concern.

Preliminary analyses, including independent samples t-tests, ANOVAs, descriptive statistics, and Pearson correlation analyses, were conducted to examine the relationships among gender, age (T2), problematic smartphone use (T1), academic procrastination (T2), anxiety (T2), and non-suicidal self-injury (T2).

For the mediation analysis, Model 6 of the PROCESS macro v3.5 for SPSS was employed to test the proposed multiple mediation model [52], which hypothesizes that problematic smartphone use influences NSSI through sequential effects on academic procrastination and anxiety. A bootstrap procedure with 5000 resamples was used to estimate the total, direct, and indirect effects. Effects were deemed significant if the 95% confidence interval for the bootstrapped path coefficient did not include zero.

Results

Descriptive and correlation analysis

Prominently, t-tests revealed that compared to males, females were significantly higher in academic procrastination (T2), t = -4.37, p < .001, and anxiety (T2), t = -2.20, p < .05. Females and males did not differ on problematic smartphone use (T1) and non-suicidal self-injury (T2), p > .05. Only children (32.7%) and non-only children (67.3%) did not differ significantly on any of the research variables, p > .05. Results of ANOVA revealed that adolescents who came from different places differed on problematic smartphone use (T1), F (2) = 6.49, p < .01. Adolescents coming from urban areas were lower in problematic smartphone use (T1) than those coming from villages and towns were (p < .05).

The descriptive statistics and correlation matrix of age (T2), problematic smartphone use (T1), academic procrastination (T2), anxiety (T2), and non-suicidal self-injury (T2) are summarized in Table 1. Age was non-significantly associated with all research variables (p > .05). Moreover, problematic smartphone use (T1), academic procrastination (T2), anxiety (T2), and non-suicidal self-injury (T2) were all significantly and positively correlated (p < .001). According to the preliminary analysis, only gender was related to the outcome variables. Thereby, we included gender as a covariate in our model.

Table 1.

Descriptive statistics of all variables and pearson’s r of all variables

Variables	Total Score Range	M	SD	1	2	3	4
1 T2Age		15.96	0.46	1
2 T1Problematic Smartphone Use	6–30	14.50	5.14	− 0.019	1
3 T2Academic Procrastination	16–80	43.98	11.38	0.005	0.252***	1
4 T2Anxiety	7–35	11.91	4.41	− 0.005	0.366***	0.452***	1
5 T2Non-suicidal Self-injury	7–28	7.94	2.11	0.009	0.131***	0.387***	0.190***

Note. N = 965. ^*** p < .001

Testing for the multiple mediation model

To further investigate the association between problematic smartphone use (T1) and non-suicidal self-injury (T2), we established a multiple mediation model (Fig. 1). In this model, problematic smartphone use (T1) was specified as the independent variable, non-suicidal self-injury (T2) as the dependent variable, academic procrastination (T2) and anxiety (T2) as mediating variables, and gender as a covariate. As shown in Table 2, problematic smartphone use (T1) was significantly and positively associated with academic procrastination (T2) (β = 0.374, p < .001), and anxiety (T2) (β = 0.103, p < .01). Academic procrastination (T2) was significantly and positively related to anxiety (T2) (β = 0.413, p < .001), which in turn predicted non-suicidal self-injury (T2) (β = 0.376, p < .001). Whereas, the direct effect of problematic smartphone use (T1) on non-suicidal self-injury (T2) (β = 0.033, p > .05) and the direct effect of academic procrastination (T2) on non-suicidal self-injury (T2) (β = 0.013, p > .05) were not significant. Thus, the multiple mediation model was partially supported.

Fig. 1.

The assumptive multiple mediated model

Table 2.

Testing the pathways of the multiple mediation model

Predictors	Model 1 (T2Academic Procrastination)			Model 2 (T2Anxiety)			Model 3 (T2Non-suicidal Self-injury)
	β	t	95%CI	β	t	95%CI	β	t	95%CI
gender	0.140	4.72***	[0.082, 0.198]	0.013	0.46	[-0.044, 0.070]	− 0.034	-1.12	[-0.093, 0.025]
T1Problematic Smartphone Use	0.374	12.33***	[0.314, 0.433]	0.103	3.28**	[0.042, 0.165]	0.033	0.99	[-0.032, 0.097]
T2Academic Procrastination				0.413	13.29***	[0.352, 0.474]	0.013	0.36	[-0.056, 0.082]
T2Anxiety							0.376	11.23***	[0.310, 0.442]
R 2	0.15			0.21			0.15
F	87.04***			86.82***			43.19***

Note. N = 965. ^** p < .01, ^*** p < .001

Bootstrap test of mediating effect

Based on the results of the multiple mediation model testing, a bootstrap test was used to test the significance of total effect, direct effect, indirect effects, and total indirect effect in this multiple mediation model. The standardized effect, Boot SE, 95% confidence interval, and relative effect size of the total effect, direct effect, three indirect effects, and total indirect effect are listed in Table 3. The results revealed that total effect of problematic smartphone use (T1) on non-suicidal self-injury (T2) was 0.134, 95% CI = [0.070, 0.198], whereas the direct effect of this relation was 0.032, 95% CI = [-0.032, 0.097]. Furthermore, two mediating effects (namely T1Problematic Smartphone Use → T2Anxiety → T2Non-suicidal Self-injury, T1Problematic Smartphone Use → T2Academic Procrastination → T2Anxiety → T2Non-suicidal Self-injury) were significant. Specifically, the mediating effect of anxiety (T2) was 0.039, 95% CI = [0.015, 0.067], and the sequential mediating effect of academic procrastination (T2) and anxiety (T2) was 0.058, 95% CI = [0.041, 0.077], whereas, the mediating effect of academic procrastination (T2) was not significant (indirect effect = 0.005, 95% CI = [-0.023, 0.034]). Thus, the multiple mediation model indicated that problematic smartphone use (T1) significantly predicted non-suicidal self-injury (T2) completely and indirectly via the mediating effect of anxiety (T2) and the sequential mediating effect of academic procrastination (T2) and anxiety (T2).

Table 3.

Total effect, direct effect, indirect effects, and total indirect effect in the mediation model

		Standardized coefficient	Boot SE	Boot CI lower limit	Boot CI upper limit	Relative Effect Size
Total Effect		0.134	0.03	0.070	0.198	100%
Direct Effect		0.032	0.03	− 0.032	0.097	24%
Indirect Effects
	T1Problematic Smartphone Use → T2Academic Procrastination → T2Non-suicidal Self-injury	0.005	0.01	− 0.023	0.034	4%
	T1Problematic Smartphone Use → T2Anxiety → T2Non-suicidal Self-injury	0.039	0.01	0.015	0.067	29%
	T1Problematic Smartphone Use → T2Academic Procrastination → T2Anxiety → T2Non-suicidal Self-injury	0.058	0.01	0.041	0.077	43%
Total Indirect Effect		0.102	0.02	0.069	0.136	76%

Discussion

The sequential mediating effect of academic procrastination and anxiety

This study investigated the relationships among problematic smartphone use (PSU), academic procrastination, anxiety, and non-suicidal self-injury (NSSI) in a longitudinal sample of Chinese adolescents. Our findings indicate that T1 PSU indirectly predicts T2 NSSI primarily through increased anxiety. Although we initially hypothesized that academic procrastination would also serve as a mediator, its direct mediating effect on the PSU–NSSI relationship was not significant. Instead, when combined in a sequential model, academic procrastination and anxiety together contributed significantly to explaining the effect of PSU on NSSI. In other words, while higher PSU levels were associated with both greater academic procrastination and elevated anxiety, it was the anxiety that played the more critical role in leading to NSSI behaviors.

These results are consistent with self-control theory. High PSU may impair emotion regulation, potentially due to changes in brain regions such as the prefrontal cortex [53–55], which is pivotal in managing negative emotions. As a result, adolescents may resort to maladaptive behaviors like NSSI as a means to regulate these overwhelming feelings [17, 54]. Our study further reinforces prior research showing that individuals with higher PSU exhibit more severe anxiety symptoms [29, 56], and that these symptoms, in the absence of effective coping strategies, contribute to self-injurious behavior.

Limitations and implications

This study provides a novel perspective by shifting the focus from viewing PSU solely as a coping mechanism for depression and anxiety [57] to exploring how an existing PSU problem can lead to adverse outcomes such as NSSI. However, several limitations warrant consideration. First, the reliance on self-report measures may introduce subjectivity and bias [58]. Future research should incorporate objective measures (e.g., phone usage logs) to mitigate these issues [59]. Additionally, while our sample of Chinese adolescents offers valuable insights, the cultural specificity of the sample may limit the generalizability of the findings.

From a practical standpoint, our results underscore the importance of addressing anxiety in interventions aimed at reducing NSSI. Strategies such as cognitive-behavioral therapy, group counseling, and role-playing exercises that focus on modifying maladaptive cognitive beliefs and enhancing emotion regulation may be particularly effective. Moreover, interventions designed to reduce PSU may also help alleviate the downstream effects on anxiety and academic procrastination, ultimately reducing the risk of NSSI.

Conclusion

In summary, our study found that PSU does not directly predict NSSI; rather, its impact is mediated through anxiety, both independently and as part of a sequential pathway with academic procrastination. These findings highlight the importance of addressing emotional regulation difficulties in adolescents with high PSU. Interventions that target both PSU and anxiety may be key to reducing NSSI behaviors.

Abbreviations

PSU

Problematic smartphone use

NSSI

non-suicidal self-injury

Author contributions

Ling Bao: Conceptualization Investigation, Methodology, Resources, Writing-original draft. Wen Zhang: Software, Investigation. Jinzhe Zhao: Data curation, Methodology. Jingyu Geng: Conceptualization, Data analysis, Methodology, Supervision, Writing-review & editing.

Funding

No funding.

Data availability

The data of this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

The research was conducted in accordance with the Declaration of Helsinki and received approval from the Human Research Ethics Committee of Beijing Normal University (Approval Number: 202109280047). All participants were fully informed about the study and provided written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.