Depressive and mobile phone addiction symptoms in Chinese adolescents with major depressive disorder: the mediating effect of alexithymia

Abstract

Background

Mobile phone addiction (MPA) is associated with depression, yet the underlying mechanisms are not clear. This study aimed to explore the socio-demographic and clinical correlates of MPA symptoms, the associations between depressive and MPA symptoms, and whether the associations are mediated by alexithymia among adolescents with major depressive disorder (MDD).

Methods

This cross-sectional study was conducted from January to July 2021 in seven hospitals across the northern, central, and southern regions of Anhui Province, China. MPA symptoms, depressive symptoms, and alexithymia were assessed using the Mobile Phone Addiction Scale (MPAS), the Center for Epidemiologic Studies of Depression Symptom Scale (CES-D) and the 20-item Toronto Alexithymia Scale (TAS-20), respectively.

Results

A total of 286 adolescents with MDD were included. Univariate analyses revealed that adolescents with abnormal parental marriage, poorer academic performance, higher total scores of CES-D and TAS-20, and higher subscale scores of difficulty identifying feeling (DIF) and externally oriented thinking (EOT) were likely to have more severe MPA symptoms (all P < 0.05). Multivariate linear regression analyses showed that depressive symptoms were positively correlated with the severity of MPA symptoms in adolescents with MDD (all P < 0.05). Alexithymia and EOT partially mediated the associations between depressive and MPA symptoms.

Conclusion

MPA symptoms are common in adolescents with MDD, and the effect of depressive symptoms on the severity of MPA symptoms is mediated partially through alexithymia. Therefore, effective identification and intervention for alexithymia may be important strategies to help clinical staff reduce the risk of MPA among adolescents with MDD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-025-07271-8.

Introduction

With the development of information technology, internet use has been increasing steadily and rapidly, both in China and worldwide. From 2000 to 2023, the number of global internet users has increased by 1,392%, which now represents 67.9% of the population around the world [1]. Moreover, recent data showed that there were 1.09 billion internet users in China, and 99.9% of them used mobile phones to access the internet [2]. The mobile phone has revolutionized the way people live, helping them to communicate and learn more easily and efficiently, especially for adolescents and students [3]. However, excessive mobile phone use can also lead to various problems, including addiction and mental health issues such as depression and anxiety [4].

Mobile phone addiction (MPA) is a type of behavioral addiction, also known as “compulsive mobile phone use”, “excessive mobile phone use”, or “problematic mobile phone use” in previous studies [5]. Similar to internet addiction, the core features of MPA include compulsive behaviour, tolerance, withdrawal, and functional impairment [6, 7]. The reported prevalence of MPA symptoms among adolescents varies widely between different countries or regions. For instance, several surveys conducted in different countries (India, China, and Iran) showed that the prevalence of MPA symptoms among adolescents ranged from 11.1 to 53.3% [8–10]. Individuals with MPA symptoms are prone to losing control and constantly using their mobile phones for gratification, which interferes with their daily lives [11]. For adolescents, MPA symptoms can negatively affect their academic performance, interpersonal relationships, and quality of life, and even lead to impulsive and violent behaviour [5, 12]. However, the etiopathological pathways and processes of MPA have not yet been clarified, and the empirical data in its research area needs to be further refined [13].

Major depressive disorder (MDD) is a significant cause of disability and death in adolescents, and prevalence rates are increasing every year. A meta-analysis showed that the global point prevalence of depressive symptoms among adolescents increased from 24% (2001–2010) to 37% (2011–2020) [14]. In recent years, several studies have investigated the associations between depressive and MPA symptoms in adolescents. Several large-sample cross-sectional studies have found that depressive symptoms are positively associated with MPA symptoms in student and adolescent population [14–17]. Compared with the general population, patients with depression had a 4.2-fold increased risk of MPA, after controlling for demographic variables [18]. A prospective cohort study in Korea revealed that depressive and MPA symptoms were mutually predictive across time periods, suggesting a possible bidirectional association [19]. However, the underlying mechanisms of the associations have not been well explained.

Alexithymia may play a mediating role in the associations between depressive and MPA symptoms. The term alexithymia, first proposed by Sifneos, described a difficulty in identifying, expressing, describing or distinguishing emotions in individuals [20]. On the one hand, alexithymia has been found to be associated with several psychiatric or mental disorders, including depression, anxiety, and schizophrenia [21, 22]. A meta-analysis of 3,752 subjects demonstrated that alexithymia was moderately correlated with the severity of depression in both clinical and general population samples [23]. Another two cross-sectional surveys of students and adolescents showed that higher alexithymia scores were significantly associated with depressive symptoms [24, 25]. Initially considered a personality trait, alexithymia was later defined as a defensive response to stressful or traumatic events [26]. Two prospective cohort studies have also suggested that alexithymia may be a state-dependent phenomenon in patients with depression [27, 28]. On the other hand, alexithymia has been associated with a high risk of substance (alcohol [29] and tobacco [30]) and behavioral (gambling [31] and internet [32]) addictions. Both a cross-sectional study and a meta-analysis also found that alexithymia was a significant predictor of MPA symptoms in the general adolescent and student population [33, 34]. In addition, Bonnet et al. found that alexithymia may mediate the relationship between negative emotionality and substance abuse [35]. Some studies have also reported that alexithymia increased the risk of internet addiction in adolescents with traumatic experiences [36, 37]. Based on the above, this study hypothesised that alexithymia may play a mediating role in the associations between depressive and MPA symptoms in adolescents with MDD.

Although several studies have investigated the associations between depressive and MPA symptoms, the underlying mechanisms are unclear in adolescents with MDD. Therefore, this study aimed to examine (1) the socio-demographic and clinical correlates of MPA symptoms, (2) the associations between depressive and MPA symptoms, and (3) whether the associations are mediated by alexithymia among adolescents with MDD.

Methods

Study design and participants

This cross-sectional study was conducted from January to July 2021 in three general hospitals and four psychiatric hospitals across the northern, central, and southern regions of Anhui Province, China. Participants were included, if they met the following criteria: (1) aged between 12 and 18 years; (2) diagnosed with MDD by two senior psychiatrists using a structured clinical interview for the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5); (3) had the ability to understand and complete the content of the assessment. Participants with other psychiatric disorders, intellectual disability, neurological or serious physical diseases (e.g., severe infections, cancer and thyroid dysfunction) were excluded.

The study protocol was approved by the Medical Ethics Committee of Chaohu Hospital of Anhui Medical University (202009-kyxm-04). All participants and their guardians signed an informed consent form after fully understanding the purpose and content of this study. All research procedures were strictly in line with the principles of the Helsinki Declaration.

Data collection and measurements

Socio-demographic and clinical data were collected by a predesigned questionnaire, including age (years), sex (male and female), age of onset (years), duration of illness (months), one child family (a reply of “yes” or “no” to the question “Are you the only child in your family?”), parental marriage (normal, separation, divorce or others), study pressure (low, fair, or high), academic performance (good, fair, or poor), personal health status perception (good, fair, or poor), and family financial status perception (good, fair, or poor). Body mass index (BMI) was calculated as weight (kg)/height (m)².

MPA symptoms were assessed using the Mobile Phone Addiction Scale (MPAS), which was derived from the Young’s Internet Addiction Test [38, 39]. Each item is scored on a six-point scale, ranging from 1 (do not agree) to 6 (completely agree), with higher total scores indicating more severe MPA symptoms [40]. In addition, the Center for Epidemiologic Studies of Depression Symptom Scale (CES-D) was used to assess the patients’ depressive symptoms in the past week [41]. The CES-D consists of 20 items that are scored on a 4-point scale ranging from 0 (rarely or none of the time) to 3 (most or all of the time), with a higher total score indicating more severe depressive symptoms. Finally, the 20-item Toronto Alexithymia Scale (TAS-20) was used to assess patients’ alexithymia symptoms [44], which consists of three subscales: difficulty identifying feelings (DIF), difficulty describing feelings (DDF), and externally oriented thinking (EOT) subscales.

Statistical analysis

Statistical analyses were conducted using Statistical Product and Service Solutions (SPSS) version 23.0 (SPSS Incorporated, Chicago, Illinois, United States of America). The continuous variables were described as mean ± standard deviation (SD), and the normal distribution was measured using the Kolmogorov-Smirnov test. The categorical variables were described as frequency distributions (%). The associations between the MPAS score and socio-demographic characteristics were examined with independent samples t-test and analysis of variance (ANOVA) as appropriate. Pearson or Spearman correlation analyses were used to examine the correlations of the severity of MPA symptoms with depressive symptoms and alexithymia. Then multivariate linear regression models were used to examine any significant correlations (P < 0.05) in univariate and correlation analyses. The mediation effect was tested using the SPSS macro PROCESS (model 4). Indirect effects were estimated using non-parametric weights for a bootstrap sample of 5000. The 95% confidence interval (CI) did not include zero, indicating that the indirect effects were significant. The P-values were set as two-tailed α = 0.05.

Results

Participant characteristics

A total of 300 adolescents with MDD were invited to participate in this study. Out of these, 286 participants completed the assessment and were included in the statistical analyses, giving a response rate of 95.3%. In this study, the mean age of the patients was 15.28 years (SD = 1.68), and about one third (27.3%) were male. More than half (53.5%) of the adolescents reported high study pressure, and 20.2% had poor academic performance (Table 1).

Table 1.

Socio-demographic and clinical characteristics of adolescents with major depressive disorder

Variables	Total sample (N = 286)
	N (%)
Male	78 (27.3)
One Child family	116 (40.6)
Parental marriage
Normal	212 (74.1)
Separation, divorce or others	74 (25.9)
Family financial status perception
Good or fair	246 (86.0)
Poor	40 (14.0)
Study pressure
Low	10 (3.5)
Fair	123 (43.0)
High	153 (53.5)
Academic performance
Good	120 (42.0)
Fair	108 (37.8)
Poor	58 (20.2)
Personal health status perception
Good	26 (9.1)
Fair	189 (66.1)
Poor	71 (24.8)
	Mean (SD)
Age (years)	15.28 (1.68)
BMI (Kg/m2)	20.87 (3.78)
Age of onset (years)	13.74 (2.01)
Duration of illness (months)	19.83 (17.66)
MPAS score	32.72 (12.86)
CES-D score	36.46 (12.92)
TAS−20 total score	59.66 (8.49)
DIF subscale score	18.55 (4.03)
DDF subscale score	18.03 (3.26)
EOT subscale score	23.08 (3.90)

BMI body mass index, MPAS Mobile Phone Addiction Scale, CES-D Center for Epidemiologic Studies Depression Scale, TAS−20 20-item Toronto Alexithymia Scale, DIF difficulty identifying feelings DDF difficulty describing feelings, EOT externally-oriented thinking, SD standard deviation

Factors associated with the severity of MPA symptoms

The unadjusted associations between socio-demographic characteristics and the severity of MPA symptoms were shown in Table 2. We found that the severity of MPA symptoms was significantly associated with parental marriage (P = 0.019) and academic performance (P = 0.001). As shown in Table 3, correlation analyses showed that the severity of MPA symptoms was positively associated with total score of CES-D (r = 0.242, P < 0.001), total score of TAS-20 (r = 0.250, P < 0.001), and subscale scores of DIF (r = 0.168, P = 0.004) and EOT (r = 0.320, P < 0.001).

Table 2.

Univariate analyses of factors associated with severity of mobile phone addiction

Variables	MPAS score	Statistics
	Mean (SD)	t/F	P
Sexa		0.280	0.780
Male	32.37 (13.12)
Female	32.85 (12.79)
One Child familya		−0.107	0.915
Yes	32.65 (12.95)
No	32.82 (12.79)
Parental marriagea		−2.357	0.019
Normal	31.67 (12.59)
Separation, divorce or others	35.73 (13.24)
Family financial status perceptiona		−1.117	0.265
Good and fair	32.38 (12.71)
Poor	34.83 (13.72)
Study pressureb		1.811	0.165
Low	28.50 (13.44)
Fair	31.51 (12.00)
High	33.97 (13.42)
Academic performanceb		7.750	0.001
Good	29.41 (10.77)
Fair	34.35 (13.49)
Poor	36.53 (14.16)
Personal health status perceptionb		1.759	0.174
Good	29.31 (13.68)
Fair	32.47 (12.31)
Poor	34.65 (13.83)

MAPS Mobile Phone Addiction Scale, a Independent-samples t-test, b Analysis of Variance (ANOVA) test, SD standard deviation

Bolded P values < 0.05

Table 3.

Correlations between severity of mobile phone addiction with depressive symptoms and alexithymia

Variables	MPAS score
	r	P
Age (years)	−0.037	0.529
BMI (Kg/m2)	−0.014	0.815
Age of onset (years)	0.038	0.527
Duration of illness (month)a	−0.028	0.635
CES-D score	0.242	< 0.001
TAS−20 total score	0.250	< 0.001
DIF subscale score	0.168	0.004
DDF subscale score	0.062	0.296
EOT subscale score	0.320	< 0.001

MPAS Mobile Phone Addiction Scale, BMI body mass index, CES-D Center for Epidemiologic Studies Depression Scale, TAS−20 20-item Toronto Alexithymia Scale, DIF difficulty identifying feelings, DDF difficulty describing feelings, EOT externally-oriented thinking

 Bolded P values < 0.05, a Spearman correlation analysis

Independent correlates of the severity of MPA symptoms

As shown in Table 4, multivariate linear regression analyses (model 1: TAS-20 total score, or model 2: DIF and EOT subscale scores involved in the models, respectively) indicated that the severity of MPA symptoms was positively associated with fair (model 1: β = 0.158, P = 0.010; model 2: β = 0.140, P = 0.020) and poor (model 1: β = 0.179, P = 0.004; model 2: β = 0.144, P = 0.019) academic performance, total score of CES-D (model 1: β = 0.135, P = 0.042; model 2: β = 0.156, P = 0.020), total score of TAS-20 (β = 0.151, P = 0.024) and EOT subscale score (β = 0.252, P < 0.001).

Table 4.

Independent correlates of severity of mobile phone addiction in adolescents with major depressive disorder

Variables	Model 1				Model 2
	B	β	t	P	B	β	t	P
Parental marriage (ref. normal)
Separation, divorce or others	3.088	0.105	1.873	0.062	3.133	0.107	1.946	0.053
Academic performance (ref. good)
Fair	4.176	0.158	2.580	0.010	3.707	0.140	2.334	0.020
Poor	5.723	0.179	2.921	0.004	4.583	0.144	2.363	0.019
CES-D score	0.135	0.135	2.045	0.042	0.155	0.156	2.346	0.020
TAS -20 total score	0.288	0.151	2.275	0.024	-	-	-	-
DIF subscale score	-	-	-	-	0.044	0.014	0.207	0.836
EOT subscale score	-	-	-	-	0.832	0.252	4.392	<0.001
F	8.184				9.579
R 2	0.128				0.171

Model 1: TAS−20 total score involved in the linear regression

Model 2: DIF and EOT subscale scores involved in the linear regression

CES-D Center for Epidemiologic Studies Depression Scale, TAS−20 20-item TorontoAlexithymia Scale, DIF difficulty identifying feelings, EOT externally-oriented thinking

Bolded P values < 0.05

The mediating effect of alexithymia in the associations between depressive and MPA symptoms

Figure 1 shows the results of the mediation model with alexithymia and EOT as mediators, respectively. After controlling for demographic variables, the bootstrap 95% CI did not include zero, indicating that all indirect effects were statistically significant (95% CI = 0.006–0.161 for alexithymia and 95% CI = 0.016–0.094 for EOT). The total effect of depressive symptoms on MPA symptoms was 0.212. The indirect effects of alexithymia and EOT were 0.079 and 0.051, which accounted for 37% and 24% of the total effect, respectively (Table 5). Furthermore, we investigated whether depressive symptoms mediated the associations between alexithymia and MPA symptoms. The results showed that the 95% CI for the indirect effects included zero, suggesting that there was no significant mediating role of depressive symptoms (Table S1).

Fig. 1.

The estimated coefficients of mediation effects of alexithymia (a) and externally oriented thinking (b) on depressive symptoms and mobile phone addiction. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001

Table 5.

Mediation analyses of alexithymia and externally-oriented thinking between depressive and mobile phone addiction symptoms

Paths	B	SE	95% CI		Ratio
			LLCI	ULCI
CES-D→TAS−20→MPAS
Indirect effect	0.079	0.038	0.006	0.161	37%
Direct effect	0.133	0.066	0.004	0.263	63%
Total effect	0.212	0.056	0.101	0.323
CES-D→EOT→MPAS
Indirect effect	0.051	0.021	0.016	0.094	24%
Direct effect	0.161	0.056	0.052	0.271	76%
Total effect	0.212	0.056	0.101	0.323

CES-D Center for Epidemiologic Studies Depression Scale, TAS−20 20-item Toronto Alexithymia Scale, MPAS Mobile Phone Addiction Scale, EOT externally-oriented thinking, SE standard error, CI confidential interval.

Discussion

To our knowledge, this study is the first to examine the socio-demographic and clinical correlates of MPA symptoms in Chinese adolescents with MDD, and whether alexithymia mediates the associations between depressive and MPA symptoms. Consistent with prior research [5, 43], we found that MPA symptoms were associated with poorer academic performance in adolescents with MDD, which may be due to the fact that they spend a significant amount of time on phones and limited time on coursework. Moreover, excessive mobile phone use may lead to insomnia, fatigue and poor concentration, which can also affect their academic performance [44]. This study also found that patients with abnormal parental marital status were more likely to have MPA symptoms, which is similar to findings in the general adolescent and student population [45, 46]. Unlike adolescents who grow up in intimate families, those who live in dysfunctional families are less likely to develop intimate relationships with others, instead tending to treat mobile phones as attachment objects [45]. However, this association disappeared after adjusting for other demographic and clinical variables. Our interpretation is that parental marital status may be an indicator of MPA symptoms, but is not independent of other variables.

The study also found that depressive symptoms were positively correlated with the severity of MPA symptoms. Similarly, Liu et al. found that MPA symptoms were positively associated with depressive symptoms in college students, and the associations were mediated by sleep disturbances [15]. Another meta-analysis showed that 55.8%−89.9% of mobile phone users had neck and upper back complaints, and that various musculoskeletal problems caused by excessive mobile phone use can also trigger negative emotions such as depression and anxiety [47]. Additionally, adolescents with MDD are more likely to use social media to communicate with others to avoid face-to-face contact, and the reduction in real-world social activities may further exacerbate their MPA symptoms [48]. Therefore, depression may be seen as a cause and consequence of MPA and vice versa, creating a vicious cycle that leads to a poor prognosis.

Several potential biological factors may explain the associations. Firstly, an important predisposing factor for MDD is thought to be the abnormal functioning of the brain reward circuitry. Several clinical studies have shown that, compared with healthy controls, patients with MDD have impaired reward-related learning signals in the ventral tegmental area and blunted reward prediction error signals in the striatum [49, 50]. In addition, a systematic review revealed that smartphone addiction was associated with impairments in cognitive control related to reward processing and executive function in adolescents and young adults [51]. This suggests that altered function of the brain reward system may be a common pathophysiological mechanism for both MDD and MPA. Secondly, the development of MPA may be associated with the dopaminergic and serotonergic systems, as well as the hypothalamic neuropeptide oxytocin [52, 53]. Two previous studies found that adolescents with internet addiction had increased dopamine level and decreased 5-hydroxytryptamine (5-HT) level compared with the general population [54, 55]. The homozygous short allelic variant of the serotonin transporter gene was found to be significantly associated with both depression and excessive internet use [56]. Additionally, oxytocin, a significant neuropeptide, has been found to improve internet use disorders by modulating the activity of positive and negative emotional circuits [57].

In this study, alexithymia was positively associated with MPA symptoms in adolescents with MDD, which is consistent with the results of a multicentre study in different countries [58]. Adolescents with alexithymia usually have problems with real interpersonal relationships, so they tend to communicate through mobile phones, which allows them to hide real feelings and show a more comfortable mental state. In addition, individuals with alexithymia may resort to emotional regulation strategies of expression suppression, which may increase impulsive or compulsive behaviors, such as excessive mobile phone use [59]. The current study also found that the EOT subscale score positively predicted MPA symptoms in adolescents with MDD. A network analysis research found that EOT served as a bridging node in the association between alexithymia and cognitive emotion regulation in students with internet addiction [60]. Furthermore, individuals with high EOT scores lack internal awareness and control, which may lead to difficulties in changing coping strategies for excessive mobile phone use [61, 62].

Further analyses showed that alexithymia partially mediated the associations between depressive and MPA symptoms in adolescents with MDD. The role of alexithymia in the associations may be explained by cognitive-behavioral models and neurobiological mechanisms associated with pathological internet use. Firstly, according to the theory proposed by Davis [63], an individual’s pathological internet use is caused by distal contributory causes of psychopathology, including depression and anxiety, and by proximal sufficient causes of maladaptive cognitions, which in this study is alexithymia, characterized by deficits in emotional cognition. And distal factors (e.g. depression) may lead to pathological internet use through proximal factors (e.g. alexithymia). Secondly, neuroimaging studies have found abnormal functional connectivity in the default mode network (DMN) and salience network (SN) in both adolescents with depression and those with problematic smartphone use [64, 65]. Additionally, alexithymia was found to be negatively associated with functional connectivity in the DMN (e.g. pregenual anterior cingulate cortex), and the SN (e.g. right insular cortex and left anterior cingulate cortex) [66, 67]. These findings suggest that abnormal function of the DMN and SN associated with emotion processing may be a common pathophysiological mechanism underlying the associations between depressive and MPA symptoms, as well as alexithymia.

There are several limitations to this study. Firstly, as this study was cross-sectional in design, the results were not able to address the direction of causality between depression and MPA. Secondly, the MPAS used in this study lacks reliable cut-off criteria for “MPA symptoms”. Further empirical validation or psychometric rationale is necessary to determine their diagnostic thresholds. Thirdly, the relatively small sample size of male patients in this study limits the reliability of the results, and future research should prioritize expanding this subgroup. Finally, some of the self-report tools used in this study may have a recall bias, which could be further explored with more objective clinical interviews or measurement tools in the future.

Conclusion

In summary, MPA symptoms are common in adolescents with MDD, particularly in those with poorer academic performance, and higher scores of alexithymia and EOT. The positive associations between depressive symptoms and the severity of MPA symptoms were partially mediated by alexithymia. Given the negative impact of MPA symptoms in adolescents with MDD, interventions focusing on alexithymia could be developed to reduce the emergence of MPA behaviour influenced by depression, including regular screening, cognitive behavioral therapy and mindfulness-based interventions.

Authors’ contributions

HL and LX were responsible for the design and direction of the study. YT, YC, LL, CY, CC, ZL, WL, JW, XL, and XW were responsible for the collection, analysis and interpretation of the data. Drafting of the manuscript was done by YT. HL and LX were responsible for critical revision of the manuscript. All the authors revised the final version for publication.

Funding

This study was supported by the National Natural Science Foundation of China (82401798), the Scientific Research Project of Anhui Higher Education Institutions (2022AH050671), the Research Fund Project of Anhui Translational Research Institute (2022zhyx-B01), and the Anhui Health Research Project (AHWJ2024Aa10004).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study protocol was approved by the Medical Ethics Committee of Chaohu Hospital of Anhui Medical University (202009-kyxm-04). All participants and their guardians signed an informed consent form after being fully understanding the purpose and content of the study. All research procedures were strictly in line with the principles of the Helsinki Declaration.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.