What this is
- This research investigates the relationship between bedtime smartphone use and anxiety symptoms among Chinese residents.
- It examines how the duration of smartphone use before sleep correlates with anxiety levels and explores gender differences in this association.
- The study also assesses the impact of () on anxiety related to bedtime smartphone use.
Essence
- Bedtime smartphone use is linked to increased anxiety symptoms, especially in women. Longer usage correlates with higher anxiety severity, while exacerbates this effect.
Key takeaways
- Using smartphones for more than one hour before bed is associated with a 9.1% higher likelihood of experiencing anxiety (OR = 1.091).
- Individuals with both bedtime smartphone use over one hour and have a 276.2% higher likelihood of experiencing anxiety.
- Bedtime smartphone use duration is positively correlated with anxiety severity (β = 0.116, P < 0.001).
Caveats
- The study's findings may not be generalizable beyond the Chinese population, limiting broader applicability.
- Self-reported data may introduce biases, affecting the accuracy of anxiety and smartphone use assessments.
- As a cross-sectional study, it cannot establish causality between smartphone use and anxiety symptoms.
Definitions
- Problematic Internet Use (PIU): Compulsive internet use leading to negative impacts on work, studies, and social life.
- Generalized Anxiety Disorder 7 (GAD-7): A self-assessment scale measuring anxiety symptoms with scores ranging from 0 (no anxiety) to 21 (severe anxiety).
AI simplified
Introduction
With the rapid development of network technology and the dawn of the intelligent age, smartphones have become increasingly prevalent, rising to the status of an indispensable necessity in many people’s lives. With smartphones owned by 71% of the global population [1] and 104.7 million smartphone internet users in China alone [2]. A survey found that more than 80% of Chinese respondents use smartphones before bedtime [3]. Notably, bedtime smartphone usage has emerged as a widespread behavior pattern in China.
While the proliferation of smartphones has accelerated all areas of society and shortened social distances, using smartphones before bedtime has been associated with a higher risk of anxiety among adolescents [4, 5, 6]. Anxiety itself can exacerbate the risk of mortality and significantly affect interpersonal relationships and physical health [7, 8, 9]. However, the association between the duration of bedtime smartphone use and anxiety symptoms remains inadequately explored across the general population. Currently, there are multiple possible explanations for the detrimental impact of using smartphones before bedtime on anxiety. The Stimulus-Organism-Response (SOR) theory suggests that stimuli in the environment can trigger changes in an individual’s internal state, which then elicits emotional responses. Based on this theory, scholars believe that factors such as information overload, privacy invasion, and the information cocoon effect brought about by the use of digital technologies like smartphones, as environmental stimuli, can alter an individual’s psychological state, leading to anxiety and other negative emotions [10–11]. In particular, the blue light radiation emitted by smartphones when used before bed, as an environmental stimulus, has been shown to affect the user’s physiological state by suppressing the secretion of melatonin, continuously activating the sympathetic nervous system, and disrupting sleep structure, thereby exacerbating anxiety [12–13]. Moreover, the Social Upward Comparison Theory posits that individuals tend to compare themselves with others in certain aspects when using digital technologies, and this upward comparison typically triggers anxiety, inferiority, and other negative emotions. Therefore, when users browse social media on their phones before bed, they may experience anxiety due to the comparison impulses triggered by seeing the lives or achievements of others [14]. At the same time, the Social Replacement Hypothesis suggests that excessive use of smartphones, leading to immersion in the virtual world and a lack of face-to-face communication and social support in reality, limits the development of traditional interpersonal relationships and social skills, which in turn can lead to alienation from others and exacerbate anxiety [15, 16, 17]. Although there is currently no longitudinal study systematically investigating the impact of the time spent using smartphones before bed on anxiety, existing research has shown that staying up late and sleep deprivation caused by the blue light emitted by smartphones increase the risk of stress, depression, and other psychopathological conditions [18–19]. Given the constant rise in smartphone users across all age groups [1], the association between bedtime smartphone use and anxiety symptoms is likely a common concern. Therefore, it is necessary to explore this association in more detail across all age groups to provide a basis for developing early anxiety management strategies.
It is worth noting that existing studies have shown that anxiety sensitivity exhibits gender differences, as fluctuations in sex hormones (such as estradiol and progesterone) affect how individuals of different genders perceive and respond to external stimuli, which in turn influences the onset and manifestation of anxiety [20–21]. Therefore, compared to men, women may be more sensitive to the emotional impacts of behavior and are more prone to experiencing anxiety [21–22]. Research shows that women are more likely than men to use social media on smartphones more frequently. While women often face greater social media pressure, they have fewer coping mechanisms, making them more susceptible to anxiety from smartphone use [23]. Additionally, studies indicate that women are more vulnerable to the negative effects of bedtime smartphone use on sleep quality, which may increase their likelihood of experiencing anxiety due to smartphone use before bed [24–25]. Although pre-sleep smartphone use has been shown to be associated with an increased risk of anxiety, the differences in the relationship between smartphone use before sleep and anxiety symptoms across genders have not been fully explored. The gender differences in anxiety susceptibility may result in different anxiety experiences when individuals of different genders use smartphones before bedtime. Focusing on the potential relationship between pre-sleep smartphone use and anxiety symptoms in different gender groups is important for a deeper understanding of the underlying mechanisms and for formulating targeted strategies.
Previous studies have often treated the duration of bedtime smartphone use and anxiety as independent factors, without fully exploring the underlying mechanisms and their relationship [4, 5, 6]. This lack of comprehensive analysis has left the specific association between bedtime smartphone use and anxiety symptoms inadequately explored. Network analysis offers a valuable method to address this gap. By conceptualizing anxiety as a result of interacting symptoms like nervousness, difficulty relaxing, etc [26]., this approach can visualize how these variables influence each other. It reveals detailed relationships between factors, helping identify the most effective targets for intervention [27]. Therefore, identifying the core symptoms of anxiety in the network of bedtime smartphone use is crucial for formulating effective intervention strategies.
Problematic internet use (PIU) refers to the compulsive use of the internet for excessive amounts of time, leading to negative consequences in a person’s work, studies, and social life [28]. The widespread adoption of the internet has resulted in a growing number of individuals facing PIU. Estimates suggest the global prevalence of PIU varies between 20.0% and 44.6%, depending on the region [29–30]. Longitudinal studies have shown that PIU increases the risk of anxiety [31, 32, 33]. Zimmermann et al. found through longitudinal tracking data analysis that PIU is significantly associated with an increased risk of anxiety in college students [31]. Gansnerd et al.‘s cohort study, based on data collected through an app-based ecological momentary assessment protocol, revealed that PIU significantly increases and is associated with elevated anxiety risk [32]. Wartberg et al.‘s longitudinal study, analyzing data from 586 middle school students in the Northeast U.S., found that PIU is linked to an increased risk of social anxiety [33]. This growing prevalence has heightened concerns about PIU’s adverse effects on internet users’ psychological well-being. Both bedtime smartphone use and PIU are associated with a higher risk of anxiety, suggesting a potential interaction between them [4, 5, 6, 31, 32, 33]. Strengthening our understanding of how various risk factors interact can help identify variables that can modulate these risks. This knowledge can then be used to develop more precise risk interventions tailored to different population characteristics [34]. Therefore, further investigation is needed to explore whether reducing PIU can alleviate the association between bedtime smartphone use and anxiety.
Based on the above research background, this study proposes the following hypotheses: (1) Bedtime smartphone use increases the risk of anxiety in users. (2) The impact of bedtime smartphone use on anxiety symptoms differs by gender, with the effect being more pronounced in females than in males. (3) PIU interacts with the impact of bedtime smartphone use on anxiety. Based on our research hypotheses, the study conducted the following analyses: (1) Logistic regression and linear regression models were used separately to analyze the relationship between bedtime smartphone use and anxiety risk, as well as the severity of anxiety in the general population. Using network analysis to further explore the underlying mechanisms of the association between bedtime smartphone use and anxiety symptoms. (2) Network comparison testing (NCT) was used to explore gender differences in the association between bedtime smartphone use and anxiety symptoms. (3) Interaction analysis was conducted using logistic regression models to examine the interaction effect of PIU on the association between bedtime smartphone use and anxiety symptoms. It is worth noting that the use of electronic devices at night can adversely affect users’ sleep, thereby increasing their risk of anxiety [12–13]. To mitigate the detrimental effects of bedtime smartphone use on sleep quality, Gradisar et al. recommended avoiding smartphone usage for one hour before bedtime [35]. Therefore, in exploring the association between bedtime smartphone use and anxiety risk and the severity of anxiety, this study referred to Gradisar et al.‘s research and transformed the duration of bedtime smartphone use into a binary variable using a one-hour threshold for analysis [35]. Furthermore, following Demetrovics et al.‘s research, participants were grouped based on a cutoff PIU score of 15, separating individuals into those with and without PIU [36]. This variable handling strategy simplifies the variable structure, enhances the interpretative validity of the interaction analysis, and provides clear classification criteria for developing targeted intervention strategies. By gaining a preliminary understanding of these associations, this study aimed to promote healthy smartphone habits among users and lay the groundwork for future longitudinal research.
Materials and methods
Data source and participants
This study utilized data from the Chinese Psychological and Behavioural Study of the Population (PBICR), a nationwide, multicenter, large-sample cross-sectional survey conducted in China [37–39]. PBICR provides a comprehensive and systematic dataset on the mental health and health behaviors of the Chinese population, offering valuable support for research in these areas. The survey was conducted from June 20, 2022, to August 31, 2022, utilizing a stratified and quota sampling method. It encompassed 148 cities, 202 districts/counties, 390 townships/streets, and 780 communities/villages across 23 provinces, 5 autonomous regions, and 4 municipalities directly under China’s Central Government (excluding Hong Kong, Macao, and Taiwan). Trained investigators administered the questionnaire face-to-face to participants. The questionnaire covered eight aspects: demographics, individual health status, family information, social environment, psychological assessment scales, behavioral assessment scales, and attitudes toward societal hot topics. This study was registered with the China Clinical Trial Registry (ChiCTR) (ChiCTR2200061046) [40, 41].
Definition of variables
Bedtime smartphone use
In this study, bedtime smartphone use was recorded as a single item in the database, where participants were asked to report the average number of minutes per day they spent using a smartphone before going to bed over the past week. The measurement options for this item were set in nine gradient intervals, each with a 15-minute range: 0–15 min, 16–30 min, 31–45 min, 46–60 min, 61–75 min, 76–90 min, 91–105 min, 106–120 min, and over 120 min. This interval design, with 15-minute increments, aims to reduce cognitive load for participants when reporting, minimize recall bias, and improve the accuracy of time estimates [42–43]. Participants were required to select the option that best matched their actual usage.
Anxiety status
Participants’ anxiety status was assessed using the Generalized Anxiety Disorder 7 (GAD-7) scale recorded in the database. GAD-7 is a self-assessment scale comprising seven items of anxiety symptoms developed by Spitzer et al. [44]. The details of the seven symptom program are as follows: GAD1: feeling nervous, anxious, or on edge, GAD2: inability to stop or control worrying, GAD3: worrying too much about a variety of things, GAD4: difficulty relaxing, GAD5: Inability to sit still due to restlessness, GAD6: becoming annoyed or easily irritated, GAD7: feeling scared because something terrible seems to be about to happen. The scale aims to measure the severity of anxiety symptoms in participants over the past two weeks and their impact on daily functioning. The Likert four-point scoring (0–3) is used to score each symptom item, with a total score ranging from 0 to 21. Different scores represent varying levels of anxiety: 0–4 (no anxiety), 5–9 (mild anxiety), 10–14 (moderate anxiety), and ≥ 15 (severe anxiety). In this study, patients with a score of ≤ 4 were defined as “not anxious” and those with a score of > 4 were defined as“anxious” [45]. The scale has demonstrated good reliability. In previous studies, the GAD-7 showed a Cronbach’s alpha of 0.87 [46], while in the current study, the GAD-7 exhibited a Cronbach’s alpha of 0.84.
Problematic internet use (PIU)
PIU was assessed using the PIUQ-SF-6 scale in this study. The PIUQ-SF-6 is a brief scale consisting of six items designed to assess problematic internet use. It evaluates three dimensions: obsession, neglect, and control impairment. Obsession (refers to the obsessive thinking about the internet and the mental symptoms caused by the withdrawal of internet use), neglect (refers to the neglect of basic needs and everyday activities), and control disorder (refers to the difficulties in controlling internet use). Each subscale have two items, and each item was scored with five responses, rated from 1 (never) to 5 (always/almost always). Participants respond using a five-point Likert scale (1–5) for each item. Scores range from 6 to 30, with higher scores indicating a greater risk of PIU [47]. The scale has demonstrated good reliability. In previous studies, the PIUQ-SF-6 showed a Cronbach’s alpha of 0.77 [36], while in the current study, the PIUQ-SF-6 exhibited a Cronbach’s alpha of 0.89.
Statistical analysis
Descriptive analysis, comparison of general information, logistic regression analysis, linear regression analysis, and interaction analysis were performed using SPSS 26.0. Continuous variables were normally distributed. Descriptive statistics (mean, standard deviation) were used for quantitative data, and frequency and proportion were used for categorical data. Between-group comparisons were made using t-tests and one-way ANOVA. Pearson correlation analysis was used to assess the relationship between two continuous variables. The r value is the Pearson correlation coefficient, used to measure the linear relationship between two continuous variables.
Logistic regression and linear regression models were used separately to analyze the relationship between bedtime smartphone use and anxiety risk, as well as the severity of anxiety in the general population. Additionally, logistic regression was used to examine the interaction effect of PIU on the association between bedtime smartphone use and anxiety symptoms.
Network analysis and network comparison tests (NCT) were conducted using R 4.3.1 software. Network analysis to further explore the underlying mechanisms of the association between bedtime smartphone use and anxiety symptoms. Network comparison testing (NCT) was used to explore gender differences in the association between bedtime smartphone use and anxiety symptoms. Network comparison tests (NCT) were conducted using the NetworkComparisonTest package [21]. The “bootnet” package was used to estimate the network and calculate Pearson correlation coefficients between variables. Least Absolute Shrinkage and Selection Operator (LASSO) regularization and Extended Bayesian Information Criterion (EBIC) were employed to optimize network estimation and reduce noise [26, 48]. The Fruchterman-Reingold algorithm was used for network layout [49]. Within the network, marginal weight values represent the strength of bedtime smartphone use duration’s influence on anxiety symptoms. Strength, a highly stable and replicable centrality index, was calculated using the centrality_plot function to evaluate each node’s overall importance [50]. The network structure’s robustness was evaluated using a bootstrapped case-dropping method. The Correlation Stability (CS) coefficient value represents the maximum proportion of cases that can be removed while maintaining network stability. A CS value greater than 0.25 indicates a stable and reliable network structure [51].
We adopted a two-tailed approach for all statistical tests conducted in this study, with a p-value of less than 0.05 considered statistically significant.
Results
Baseline characteristics
This study included 30,504 participants from the PBICR 2022 cross-sectional survey. Among these participants, 15,098 (49.50%) reported experiencing anxiety, with a mean anxiety score of 4.99 (SD = 4.77). Among the anxiety symptoms assessed by the GAD-7 scale, the mean scores (standard deviation, SD) were as follows: “Feeling nervous, anxious, or on edge (GAD1)” scored 0.81 (SD = 0.77), “Inability to stop or control worrying (GAD2)” 0.70 (SD = 0.79), “Worrying too much about a variety of things (GAD3)” 0.78 (SD = 0.81), “Difficulty relaxing (GAD4)” 0.74 (SD = 0.81), “Inability to sit still due to restlessness (GAD5)” 0.63 (SD = 0.76), “Becoming annoyed or easily irritated (GAD6)” 0.75 (SD = 0.79), and “Feeling scared because something terrible seems to be about to happen (GAD7)” 0.58 (SD = 0.77) (Table 1).
Males comprised 56.63% (17,275) of the participants. The median age of the participants was 36.08 years (SD = 18.12). Additionally, 27.8% of participants reported using their smartphones for more than one hour before bedtime. Furthermore, 1,081 individuals (35.40%) met the criteria for PIU (Table 2).
| Variables | Mean/N | SD/% | |
|---|---|---|---|
| Anxiety | Yes | 15,098 | 49.5 |
| No | 15,406 | 50.5 | |
| Anxiety Symptoms (scores) | Total | 4.99 | 4.77 |
| GAD1: Feeling nervous, anxious, or on edge | 0.81 | 0.77 | |
| GAD2: Inability to stop or control worrying | 0.7 | 0.79 | |
| GAD3: Worrying too much about a variety of things | 0.78 | 0.81 | |
| GAD4: Difficulty relaxing | 0.74 | 0.81 | |
| GAD5: Inability to sit still due to restlessness | 0.63 | 0.76 | |
| GAD6: Becoming annoyed or easily irritated | 0.75 | 0.79 | |
| GAD7: Feeling scared because something terrible seems to be about to happen | 0.58 | 0.77 |
| Variables | Mean/N | SD/% | t/F/r | P | |
|---|---|---|---|---|---|
| Gender (%) | Male | 17,275 | 56.63 | -2.712a | 0.007 |
| Female | 13,229 | 43.37 | |||
| Age (years) | 36.08 | 18.12 | -0.103c | <0.001 | |
| BMI (kg/m)2 | ≥ 24 | 6567 | 21.5 | -6.817a | <0.001 |
| <24 | 23,937 | 78.5 | |||
| Residence (%) | Cities and towns | 8444 | 27.68 | -3.683a | <0.001 |
| Rural | 22,060 | 72.32 | |||
| Marriage Status (%) | Married | 16,355 | 53.62 | -19.485a | <0.001 |
| Others | 14,149 | 46.38 | |||
| Monthly per capita family income (yuan) | ≥ 4000 | 15,263 | 50.04 | 0.035a | 0.972 |
| < 4000 | 15,241 | 49.96 | |||
| Family Type (%) | Nuclear family | 15,885 | 52.08 | 34.432b | <0.001 |
| Couple family | 6163 | 20.2 | |||
| Others | 8456 | 27.72 | |||
| Suffering from chronic illness (%) | Yes | 23,693 | 77.67 | 13.966a | <0.001 |
| No | 6811 | 22.33 | |||
| Drinking (%) | Yes | 21,312 | 69.87 | 11.838a | <0.001 |
| No | 9192 | 30.13 | |||
| Smoking (%) | Yes | 25,693 | 84.23 | 7.21a | <0.001 |
| No | 4811 | 15.77 | |||
| Drinking tea (%) | Yes | 14,899 | 48.84 | 9.901a | <0.001 |
| No | 15,605 | 51.16 | |||
| Drinking coffee (%) | Yes | 22,629 | 74.18 | 18.004a | <0.001 |
| No | 7875 | 25.82 | |||
| Duration of bedtime smartphone use (hours) | > 1 | 8490 | 27.8 | 24.048a | <0.001 |
| ≤ 1 | 22,014 | 72.2 | |||
| Length of sleep (hours) | > 6 | 22,480 | 73.7 | 29.046a | <0.001 |
| ≤ 6 | 8024 | 26.3 | |||
| Sleep quality (%) | Good | 5261 | 17.25 | -41.407a | <0.001 |
| Poor | 25,243 | 82.75 | |||
| PIU Total Scores | 12.31 | 5.61 | 0.454c | <0.001 |
The relationship between the bedtime smartphone use and risk of anxiety
After adjusting for covariates in all three models, the logistic regression results revealed that, compared to participants who used their smartphones for one hour or less before bedtime, using smartphones for more than one hour before bedtime was associated with a 9.1% higher likelihood of experiencing anxiety (OR = 1.091) (Table 3).
| Model Adjustment | β | SE | Wald χvalue2 | -valueP | OR(95%CI) |
|---|---|---|---|---|---|
| Model 1 | 0.569 | 0.026 | 481.987 | <0.001 | 1.767(1.680,1.859) |
| Model 2 | 0.49 | 0.027 | 323.236 | <0.001 | 1.633(1.548,1.722) |
| Model 3 | 0.087 | 0.031 | 7.978 | 0.005 | 1.091(1.027,1.158) |
Relationship between the duration of bedtime smartphone use and the severity of anxiety
A multiple linear regression model was used to analyze the association between the duration of bedtime smartphone use and the severity of anxiety (eTable 1 in Supplement). Even after controlling for covariates, the duration of bedtime smartphone use was positively correlated with anxiety severity (β = 0.116, P < 0.001).
Network analysis of the duration of bedtime smartphone use and anxiety symptoms
The duration of bedtime smartphone use showed a negative correlation with GAD-5 (inability to sit still due to restlessness) and a positive correlation with the remaining six anxiety symptoms (Fig. 1). Details on network accuracy and stability are presented in the supplemental eResults.
Among all anxiety symptoms in the network, GAD2 (inability to stop or control worrying) exhibited the highest strength, followed by GAD3 (worrying too much about various things) (eFigure 1). Interestingly, the strongest marginal weight value was observed between bedtime smartphone use duration and GAD6 (becoming easily annoyed/irritated) (eTablein Supplement). Detailed results regarding network accuracy and stability are presented in the supplemental materials (eResults, eFigures–). 2 2 4
Network Of The Duration Of Bedtime Smartphone Use And Anxiety Symptoms. GAD1: feeling nervous, anxious, or on edge. GAD2: inability to stop or control worrying. GAD3: worrying too much about a variety of things. GAD4: difficulty relaxing. GAD5: inability to sit still due to restlessness. GAD6: becoming annoyed or easily irritated. GAD7: feeling scared because something terrible seems to be about to happen. phone: the duration of bedtime smartphone us
Network analysis of the duration of bedtime smartphone use and anxiety symptoms in different genders
Figures 2 and 3 depict the separate networks for males and females, respectively, revealing the associations between bedtime smartphone use duration and anxiety symptoms. Notably, the female network exhibited a higher strength for GAD4 (inability to relax) compared to the male network (males = 0.948, females = 0.986, P = 0.007) (eFigure 6 in Supplement). Conversely, the male network showed a higher strength for GAD7 (feeling scared of something terrible happening) compared to the female network (males = 0.948, females = 0.986, P = 0.007) (eFigure 7 in Supplement). Detailed results regarding network comparison, accuracy, and stability for different genders are presented in the supplemental materials (eResults, eFigs. 5, 6, 7, 8, 9, 10, 11, 12 and 13).
The analysis revealed significant gender differences in the association between bedtime smartphone use duration and anxiety symptoms across three connections within the networks (phone-GAD1, phone-GAD5, phone-GAD6; P = 0.003, 0.001, and 0.04, respectively). For these connections, female participants exhibited higher marginal weight values compared to males (eTable 3 and eTable 4 in Supplement).
Network Of The Duration Of Bedtime Smartphone Use And Anxiety Symptoms In Males. GAD1: feeling nervous, anxious, or on edge. GAD2: inability to stop or control worrying. GAD3: worrying too much about a variety of things. GAD4: difficulty relaxing. GAD5: inability to sit still due to restlessness. GAD6: becoming annoyed or easily irritated. GAD7: feeling scared because something terrible seems to be about to happen. phone: the duration of bedtime smartphone us
Network Of The Duration Of Bedtime Smartphone Use And Anxiety Symptoms In Females. GAD1: feeling nervous, anxious, or on edge. GAD2: inability to stop or control worrying. GAD3: worrying too much about a variety of things. GAD4: difficulty relaxing. GAD5: inability to sit still due to restlessness. GAD6: becoming annoyed or easily irritated. GAD7: feeling scared because something terrible seems to be about to happen. phone: the duration of bedtime smartphone us
The interaction between PIU and bedtime smartphone use
Individuals who reported both bedtime smartphone use of more than 1 h and PIU were associated with a 276.2% higher likelihood of experiencing anxiety. Those who reported bedtime smartphone use of more than 1 h and did not have PIU were associated with a 35.3% lower likelihood of experiencing anxiety (OR = 0.647) (Table 4).
| Model Adjustment | SE | Wald χvalue2 | -valueP | OR(95%CI) | ||
|---|---|---|---|---|---|---|
| bedtime smartphone use duration>1 h* with PIU | Model 1 | 1.222 | 0.036 | 1165.088 | <0.001 | 3.394(3.164,3.641) |
| Model 2 | 1.163 | 0.037 | 994.896 | <0.001 | 3.200(2.977,3.440) | |
| Model 3 | 1.016 | 0.038 | 714.658 | <0.001 | 2.762(2.564,2.976) | |
| bedtime smartphone use duration>1 h* without PIU | Model 1 | -0.289 | 0.035 | 69.997 | <0.001 | 0.749(0.700,0.802) |
| Model 2 | -0.367 | 0.035 | 109.124 | <0.001 | 0.693(0.647,0.742) | |
| Model 3 | -0.436 | 0.037 | 142.601 | <0.001 | 0.647(0.602,0.695) |
Discussion
This study found using smartphones before bedtime was associated with a higher likelihood of experiencing anxiety and was positively correlated with the severity of anxiety. Within the network analysis, “inability to stop worrying (GAD2)” and “worrying too much (GAD3)” showed the strongest associations with bedtime smartphone use. Interestingly, bedtime smartphone use negatively correlated with “inability to sit still (GAD5)” but positively correlated with other anxiety symptoms, especially “becoming easily annoyed (GAD6)”, which had the strongest edge weight. Gender differences emerged in the network: women showed stronger associations with “difficulty relaxing (GAD4)”, while men exhibited higher intensity for “fear of something terrible happening (GAD7)”. Additionally, women had stronger edge weights between bedtime smartphone use and “nervousness (GAD1)”, “restlessness (GAD5)”, and “irritability (GAD6)” than men. The study also found that PIU moderated the relationship between bedtime smartphone use and anxiety. Individuals who used smartphones for more than 1 h before bed and had PIU were more likely to experience anxiety, those who used smartphones for more than 1 h and did not have PIU were less likely to experience anxiety.
Bedtime smartphone use has become a prevalent lifestyle habit [3]. Prior research suggests a link between bedtime smartphone use and a higher risk of anxiety disorders [7]. Similar to our findings, previous studies have shown a significant correlation between overall screen time and the severity of anxiety symptoms [52]. Several explanations exist for the detrimental effects of bedtime smartphone use on anxiety. Firstly, Screen time may replace valuable time spent with others, harming social connections and emotional well-being [53]. Secondly, using devices at night exposes users to blue light, which delays melatonin release, increases arousal, disrupts sleep, and raises anxiety levels [12–13]. Thirdly, the content viewed before bed also plays a role; for example, exposure to cyberbullying online can increase anxiety. Moreover, seeing idealized content may trigger upward social comparisons, further contributing to anxiety and negatively affecting mental and social health [54–55].
This study found that “inability to stop or control worrying” (GAD2) and “worrying too much about a variety of things” (GAD3) exhibited the strongest intensity among anxiety symptoms, indicating that worry may be the most significant and prevalent symptom among bedtime smartphone users. Research indicates that using social media before bed is associated with anxiety. Nearly 50% of smartphone users spend more than an hour on social networks before sleeping [56]. Prolonged use can cause social media fatigue, which is closely linked to anxiety and results in various worrying emotions [57]. Interestingly, the network analysis showed a negative correlation between bedtime smartphone use and “inability to sit still due to restlessness” (GAD5). Lepp et al. found that high-frequency smartphone users are generally more sedentary. Since GAD5 reflects anxiety related to movement, it may not fully capture the link between bedtime smartphone use and anxiety [58]. Additionally, bedtime smartphone use can significantly disrupt sleep patterns. Using social media and apps before bed can delay sleep, disrupt melatonin secretion, interfere with the sleep-wake cycle, shorten sleep duration, and increase irritability [59, 60, 61]. This explains why our study found that “becoming easily annoyed or irritated (GAD6)” had the strongest link with bedtime smartphone use.
This study revealed gender differences in the network structure of bedtime smartphone use and anxiety symptoms. In females, “difficulty relaxing” (GAD4) showed higher centrality in the network, and the connections between bedtime smartphone use and symptoms like “feeling nervous” (GAD1), “inability to sit still” (GAD5), and “becoming easily annoyed” (GAD6) were stronger than in males. These differences may be due to females being more sensitive to emotional impacts, more frequent social media use, and having fewer coping mechanisms for stress [20, 21, 22, 23]. Additionally, females are more affected by the negative impacts of bedtime smartphone use on sleep quality [24–25]. In contrast, males showed a higher intensity for “feeling scared because something terrible seems to be about to happen” (GAD7). Research suggests this could be due to males experiencing a higher fear of missing out (FOMO), which drives a desire to stay connected on social media and fear missing enjoyable experiences [62–63].
Research shows a positive correlation between PIU and anxiety, with PIU individuals scoring higher on anxiety measures than non-PIU individuals [31, 32, 33]. The compensatory internet use theory suggests that socially anxious people may increase their anxiety by using the internet as a substitute for real-life social interactions [64, 65, 66]. Our study found that PIU exacerbates the negative effects of bedtime smartphone use on anxiety. Specifically, individuals with PIU experience a greater increase in anxiety from smartphone use before bed compared to those without PIU. This indicates that reducing PIU could help mitigate the adverse effects of prolonged bedtime smartphone use.
This study has several limitations. First, all participants were Chinese, so the findings may not be generalizable to other populations. Second, the study used self-reported data, which may be subject to misclassification, recall issues, and response biases. Third, as a cross-sectional observational study, it cannot establish causality. While we accounted for many confounding factors, reverse causation and unmeasured confounders may still have influenced the results.
Our research highlights the need to reduce bedtime smartphone use. Individuals can start by being mindful of their nightly phone habits and cutting down on screen time before sleep. On a societal level, public health initiatives can promote healthy digital habits in schools and workplaces. Technology companies can also help by creating features that encourage responsible phone use, such as bedtime reminders or restricted modes. To spread these findings, engaging content on platforms like WeChat and articles in mainstream media can raise awareness and promote positive behavioral change.
Conclusions
This study demonstrates that bedtime was associated with a higher likelihood of developing anxiety and was positively correlated with the severity of anxiety in the general population. Worry-related symptoms showed the strongest association with bedtime smartphone use. Furthermore, the association between bedtime smartphone use and anxiety symptoms was stronger in women than in men. Additionally, our findings revealed that PIU moderated the relationship between bedtime smartphone use and anxiety. Using smartphones for more than 1 h before bedtime, combined with PIU, was associated with a higher likelihood of developing anxiety, while using a smartphone for more than 1 h before bedtime, without PIU, was associated with a lower likelihood of developing anxiety. This finding provides a potential basis for developing interventions that alleviate anxiety symptoms in the general population by targeting PIU reduction. It also provides valuable direction and a foundation for future longitudinal studies to further explore these relationships.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Acknowledgements
Thanks to all the workers who contributed to this article.
Abbreviations
Author contributions
TIAN Zheng and LU Junshuai are the co-first authors of this article. Yibo Wu and Lan Wang are the corresponding authors of this article. Writing-original draft: All authors. Methodology: Tian, Zhang, Liu, Li. Visualization: Tian, Zhang, Li, Wu, Wang. Funding acquisition: Wang. Resources: Wu.
Funding
This work was supported by Tianjin Science and Technology Program (24YDTPJC00300) and funded by Tianjin Key Medical Discipline(Specialty) Construction Project（TJYXZDXK-011A).
Data availability
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
The study was reviewed by the Second Xiangya Hospital of Central South University and Shaanxi (University) Philosophy and Social Science Key Research Base-Health Culture Research Centre. All participants were required to sign an informed consent form. The Ethics Committee of the Second Xiangya Hospital of Central South University is authorized to approve this study (ethical approval number: K050) and Shaanxi (University) Philosophy and Social Science Key Research Base-Health Culture Research Centre (ethical approval number: JKWH-2022–02). The study was conducted in accordance with the Declaration of Helsinki of the World Medical Association. This study was registered for research on 15 June 2022 (ChiCTR2200061046) with the China Clinical Trial Registry (ChiCTR).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Contributor Information
Yibo Wu, Email: bjmuwuyibo@outlook.com.
Lan Wang, Email: wangl0423@tmu.edu.cn.
References
Associated Data
Supplementary Materials
Data Availability Statement
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.