Abstract
Background
There is a high prevalence of short video addiction (SVA) among Chinese nursing students. This study was aimed at establishing a risk prediction model for SVA among this population.
Methods
Two rounds of cross-sectional survey were performed, with 620 nursing students from Jinzhou included in T1 (July 2024) survey to perform model establishment and internal validation, and 293 nursing students from Guangzhou included in T2 (February, 2025) survey to perform external validation. Participants were invited to complete a panel of questionnaires to measure SVA using the Short Video Addiction Scale (SVAS) and 22 candidate SVA predictors. A visual nomogram was plotted and its performance was evaluated using area under the receiver operating characteristics curve (AUC), calibration curve and decision curve analysis (DCA). A sex-based subgroup analysis was also conducted.
Results
A 31.3% SVA prevalence was revealed among Chinese nursing students. Academic stress, social interaction anxiety, sleep disorders, depressive symptoms, anxiety symptoms, and whether from a single-parent family were identified as significant factors for SVA nomogram construction. The nomogram performed well in the training cohort, internal validation cohort, and external validation cohort as evidenced by the AUC (0.809, 0.831, and 0.839, respectively), calibration curves and DCA curves. Subgroup analysis showed that this model performed well in both male and female nursing students.
Conclusion
Our present study developed a predictive nomogram for SVA among Chinese nursing students via integration of six salient predictors, including academic stress, social interaction anxiety, sleep disorders, depressive symptoms, anxiety symptoms, and whether from a single-parent family. This nomogram is potentially useful for universities and educators in identifying nursing students more likely to develop SVA and in developing tailored preventive interventions to reduce the prevalence of this condition.
Keywords: Short video addiction, Nursing students, Risk prediction model, Nomogram, Influencing factors
Introduction
Recent years have seen an explosive growth of short video industry and the overwhelming popularity of short videos among people in China, especially the population that is open-minded and always ready to accept new creations such as college students [1]. Unlike traditional social media, short-video platforms such as TikTok adopt a personalized recommendation system powered by big data and push algorithms to accurately target the audience of videos, greatly meeting the diverse psychological needs of users and making them prone to addictive behavior towards short video applications [2]. In addition, short-videos are much more entertaining than traditional social media. All these features of short-videos make them gain greater popularity, especially among youth than any other traditional social media [3, 4]. Social media use (SMU) including the short-video application use (SVU) such as TikTok and Kuaishou have multiple functions that entail but not limited to information access, self-express, self-fulfillment, entertainment, and enhancing social interaction [5]. Although many previous empirical studies confirmed that moderate SMU or SVU was intrinsically innocuous and had positive effect in a connected society, as it would promote the users’ sense of pleasure and subjective well-being [4, 6, 7], excessively watching short videos, on the flip side, would result in a phenomenon referred to as short-video addiction (SVA), which refers to a chronic or cyclical state of fascination resulting from the repeated use of short video applications, with strong and persistent feelings of craving and dependence [8, 9].
Alongside the increasing popularity and spread of short videos among the youth in China, its potential negative influence, cause, mechanism, and intervention of short video use related problems such as SVA have attracted increasing attention [10]. As a newly identified type of internet-related additions, SVA would consequently result in a variety of unfavorable outcomes such as academic procrastination, decrease in learning motivation and well-being, and mental health disorders among school students [11–13]. Studies have shown that SVA would induce depression in the youth via multiple mechanisms, including strengthened attention to negative information and weakened attention to positive information [14], self-objectification and self-dissatisfaction due to upward social comparison [15–18]. Based on a recent study, the SVA ratio has reach up to 30.2% among nursing undergraduates in China [19], much higher than the 21.6% prevalence among general Chinese university students [20]. Nursing students are the candidate nurses and thus their mental health and behavioral performance deeply affects the nursing and healthcare industry in the whole country. It is worth noting that problematic use of smartphone would distract the attention of nursing students in their internship, which would potentially bring about unnecessary medical errors [21]. In addition, high emotional regulation skills are also important professional qualities in nursing education and career [22, 23]. From this perspective, it is also imperative to addressing SVA in nursing students, considering the unfavorable influence of SVA on emotional disorders [11]. Therefore, there is an urgent need to establish a risk prediction model for nursing students’ SVA in order to identify and screen key populations at an early stage and conduct preventive interventions.
From the perspective of preventive interventions, it is important to determine the potential risk factors so as to facilitate the screening SVA high-risk individuals and design tailored intervention regimens. To date, Several publications regarding SVA contributors have shown that this condition was closely related the following factors, including personality trait, mental disorders (depression, anxiety, loneness, and stress), social support, core self-evaluation, social identity, academic pressure as well as its neural-transcriptomic underspinning [1, 8, 9, 24–27]. In spite of this, a systemic collection of a panel of variables directed to a risk prediction and preventive intervention of this condition is lacking, not to mention that among nursing students. Thus the present was to unearth the critical risk factors from four aspects as operated in a recent study constructing predictive model of suicidality among Chinese secondary school students [28], so as to establish a risk prediction nomogram model for nursing students’ SVA and evaluate its effectiveness. After going through the underlying risk factors triggering SVA, the predictors were collected from individual characteristics (e.g., age and mental status), family atmosphere (e.g., only-child status and whether from a single-parent family), school atmosphere (e.g., academic stress and academic performance), and unhealthy lifestyle (e.g., smoking and drinking). This model is expected to be tangible for the early identification and prevention of SVA in nursing students.
Methods
Study procedure and participants
This study adopted a cross-sectional design and was conducted in two different samples of nursing students at T1 (July 2024) in Jinzhou, a city located in Northeast China and T2 (February 2025) in Guangzhou, a city located in South China. The T1 sample was used for model training and internal validation and the T2 sample was for external validation. Three well-trained research assistants were designated for questionnaire collection. They distributed online self-administered questionnaires to the conveniently sampled nursing students through the Questionnaire Star platform by the aid of Wechat. Participants should be 18 years and above and enrolled as a nursing student in a college or university.
In the T1 survey, a total of 641 questionnaires were distributed and recovered, with 620 considered to be effective, with an effective recovery rate of 96.7%, whereas in the T2 survey, a total of 306 questionnaires were distributed and recovered, with 293 considered to be effective, with an effective recovery rate of 95.8% according to the preset removal criteria of questionnaires submitted within 90 s and/or having something that does not add up.
According to the 30.2% SVA prevalence in Chinese nursing students as reported in a recent study [19], the minimum required sample size, as per a common rule of thumb of Events Per Variable (EPV) of 10, which is widely used for determination of the least sample size in the smaller of the outcome group when performing a binary logistic regression based risk prediction model construction [29, 30], should be approximately 290 in the training cohort, considering a 20% nonresponse rate. The sample size of the 290 were calculated based on the formula of 7 × 10/30.2%/(1–20%), in which 7 represented the number of variables subjected to a multivariate logistic regression analysis, as seen later in the results section.
This survey was anonymous and subjects voluntarily participate in it. They had the right to withdraw this survey at any time. Informed consent was obtained from all the participants. The study protocol was in accordance with relevant guidelines and regulations and was approved by the Ethics Committee of Jinzhou Medical University (JZMUL20240724).
Measures
General information questionnaire
The general information of the participants was collected from four aspects: (1) individual characteristics, including age, gender, grade, education, body mass index (BMI, normal: 18.5 ≤ BMI < 24 Kg/m2), living residence; (2) family atmosphere, including only-child status, whether from a single-parent family, relationship with father, relationship with mother; (3) school atmosphere, including relationship with classmates, number of close friends, academic performance, academic stress; (4) negative lifestyle, including smoking (during the past month, smoking ≥ 1 day) alcohol drinking (during the past month, alcohol drinking ≥ 1 day); physical inactivity (during the past week, unable to reach the level of exercising ≥ 1 h/day for > 3 days/week).
Short video addiction scale (SVAS)
The SVAS developed by Qin et al. [31] was used in this study to evaluate among nursing students. The SVAS was adapted from the Mobile Phone Addiction Scale [32] via having the “mobile phone use” in the original items placed by “short video use”. The SVAS consists of 14 items divided into four dimensions: withdrawal (5 items), escapism (3 items), loss of control (4 items), and inefficiency (2 items). Each item was rated on five-point Likert scale from 1 (totally inconsistent) to 5 (totally consistent), with higher SVAS total indicating more SVA symptoms. To perform primary diagnosis of SVA, this scale drew on the internet addiction screening criteria proposed by Yong [33], where an individual would be regarded an addict when he/she gives affirmative responses to any five of the eight items with diagnostic features (items 3, 5, 6, 7, 8, 10, 15, 16). Likewise as well, in the SVAS, affirmative responses (≥ 4 points) to any four of the seven items with diagnostic features (items 2, 3, 4, 5, 7, 12, 13) mean one being a short video addict [34]. According to the study by Qin et al. [31], the SVAS had good validity and reliability among college students. For structure validity, the fitting indexes were as follows: χ2 = 251.28, df = 71, χ2/df = 3.54, RMSEA = 0.08, CFI = 0.94, and TLI = 0.93. As for reliability, the total scale Cronbach’s α being 0.91, with its four subscales ranging from 0.76 to 0.89. The applicability of this scale among the Chinese youth was also further confirmed by the subsequent reports [1, 19, 24, 26, 35]. In this study, the Cronbach’s α of the scale was 0.921 in the T1 survey, and 0.924 in the T2 survey, indicating high levels of reliability.
Depression, anxiety, and stress scale (DASS-21)
The mental health status including depression, anxiety, and stress of the participants was evaluated by DASS-21 [36]. This scale has a total of 21 items used to evaluate the three emotions above, with each having seven items. Each item was rated on a four-point Likert scale from 0 (not applied at all) to 3 (applied very much). The standard total score for each emotion was obtained by multiplying the original total score by 2. The higher standard total score means higher severe emotional response. The cut-off total standard scores for depression, anxiety, and stress of 10, 8 and 15 respectively were used for discern between emotionally normal and non-normal individuals. The Chinese version of DASS-21 has been confirmed to be psychometrically sound in college students [37]. In this study the Cronbach’s α of this scale was 0.903 in the T1 survey, and 0.960 in the T2 survey, indicating high levels of reliability.
Pittsburgh sleep quality index (PSQI)
The PSQI was used in this study to identify sleep disturbance during the past month [38]. The scale has 19 self-rated items, which further generate 7 components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Each of components was weighted equally on a four-point Likert scale from 0 (very good) to 3 (very poor). The total PSQI score ranged from 0 to 21, with higher score indicating more severe sleep disturbance. The Chinese version of this scale was demonstrated to be valid and reliable, with a global score of > 7 indicates presence of sleep disturbance [39]. In this study, the Cronbach’s α was 0.864 in the T1 survey, and 0.848 in the T2 survey, indicating high levels of reliability.
Social interaction anxiety scale (SIAS)
The SIAS was originally developed by Mattick and Clarke [40] and its Chinese version [41] has satisfactory psychometric properties when used for evaluation of anxiety and fear that individuals are confronted with din a social interaction setting. This scale consists of 19 items, with each rated on a five-point Likert scale from 0 (not at all characteristic or true of me) to 4 (extremely characteristic or true of me). The total score ranges from 0 to 76, with higher SIAS total score indicating more severe SIAS symptoms. The optimal cut-off score was 22 [42]. In this study, the Cronbach’s α for the SIAS was 0.933 in the T1 survey, and 0.961 in the T2 survey, indicating high levels of reliability.
Statistical analysis
All data analyses were performed using SPSS 26.0 and R version 4.3.2. The sample in the T1 survey was randomly divided into two parts of training cohort and internal validation cohort at a ratio of 7:3, whereas the T2 sample was used for external validation. The training cohort was used for screening effective predictors and model establishment. Internal and external cohorts were used for evaluation of model validation. Continuous data in this study were confirmed to be normally distributed and thus were presented as mean±SD (standard deviation), and categorical data as frequency and percentage. χ2 test and t test were used for rudimentary screening of predictive factors. Subsequently, the statistically significant variables in the univariate analysis were included into the least absolute shrinkage and selection operator (LASSO) regression analysis was used to screen the most suitable independent predictors. Following this, the candidate predictors were further selected using a logistic regression analysis. Finally a visual nomogram was plotted using the selected predictors. Model performance was evaluated in the training, internal, and external cohort using three indexes, including the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA). Additionally, a subgroup analysis was performed to assess the performance of the model in both genders. A two-tailed p less than 0.05 suggested statistically significant.
Results
Sample characteristics
In the T1 survey, a total of 620 nursing students from Jinzhou, a city located in Northeast China were included for model training and internal validation, whereas in the T2 survey for external validation, the effective sample size was 293 nursing students recruited from Guangzhou, a city located in South China. The distribution and level of the candidate predictors in the overall sample (n = 913), training cohort (n = 434), internal cohort (n = 186), and external cohort (n = 293), were summarized in detail in Table 1.
Table 1.
Sample characteristics of the participants
| Variable | Overall N = 913 |
Training cohort, N = 434 | Internal cohort N = 186 | External cohort N = 293 |
|---|---|---|---|---|
| Age [(mean (SD)], year | 21.13 (2.66) | 20.97 (2.65) | 21.20 (2.22) | 21.23 (3.10) |
| Sex, female | 712 (78.0%) | 344 (79.2%) | 151 (81.1%) | 217 (74.0%) |
| Grade | ||||
| First year | 391 (42.8%) | 207 (47.6%) | 89 (47.8%) | 95 (32.4%) |
| Second year | 267 (29.2%) | 111 (25.5%) | 58 (31.3%) | 98 (33.4%) |
| Third year | 166 (18.2%) | 69 (15.8%) | 27 (14.5%) | 70 (23.8%) |
| Fourth year | 89 (9.7%) | 47 (10.8%) | 12 (6.4%) | 30 (10.2%) |
| Residence | ||||
| Urban areas | 453 (49.6%) | 214 (49.3%) | 88 (47.3%) | 151 (51.5%) |
| Rural areas | 460 (50.4%) | 220 (50.6%) | 98 (52.6%) | 142 (48.4%) |
| Only child, yes | 448 (49.1%) | 207 (47.6%) | 91 (48.9%) | 150 (51.5%) |
| From a Single-parent family, yes | 168 (18.4%) | 75 (17.2%) | 32 (17.2%) | 61 (20.8%) |
| BMI, normal | 636 (69.7%) | 310 (71.4%) | 119 (63.9%) | 207 (70.6%) |
| Education | ||||
| Junior college | 396 (43.4%) | 194 (44.9%) | 82 (44.0%) | 120 (40.9%) |
| Undergraduate | 370 (40.5%) | 188 (43.3%) | 77 (41.3%) | 105 (35.8%) |
| Postgraduate | 146 (16.0%) | 51 (11.7%) | 27 (14.5%) | 68 (23.2%) |
| Academic stress | ||||
| Average and below | 418 (45.8%) | 196 (45.1%) | 79 (42.4%) | 143 (48.8%) |
| Relatively high | 399 (43.7%) | 198 (45.6%) | 88 (47.3%) | 113 (38.5%) |
| Very high | 96 (10.5%) | 40 (9.2%) | 19 (10.2%) | 37 (12.6%) |
| Academic performance | ||||
| Bad | 69 (7.6%) | 32 (7.3%) | 13 (6.9%) | 44 (15.0%) |
| Average | 628 (68.8%) | 314 (72.3%) | 125 (67.3%) | 189 (64.5%) |
| Good | 196 (21.5%) | 88 (20.2%) | 48 (25.8%) | 60 (20.4%) |
| Social interaction anxiety, yes | 359 (39.3%) | 180 (41.4%) | 70 (37.6%) | 109 (37.2%) |
| Depressive symptoms, yes | 201 (22.0%) | 95 (21.8%) | 44 (23.6%) | 62 (21.2%) |
| Anxiety symptoms, yes | 405 (44.4%) | 192 (44.2%) | 92 (49.4%) | 121 (41.2%) |
| Stress, yes | 220 (24.1%) | 105 (24.1%) | 41 (22.0%) | 74 (25.2%) |
| Relationship with mother | ||||
| Good | 815 (89.3%) | 398 (91.7%) | 169 (90.8%) | 248 (84.6%) |
| Average | 87 (9.5%) | 34 (7.8%) | 16 (8.6%) | 37 (12.6%) |
| Poor | 11 (1.2%) | 2 (0.4%) | 1 (0.5%) | 8 (2.7%) |
| Relationship with father | ||||
| Good | 676 (74.0%) | 341 (78.5%) | 137 (73.6%) | 198 (67.5%) |
| Average | 209 (22.9%) | 80 (18.4%) | 42 (22.5%) | 87 (29.6%) |
| Poor | 28 (3.1%) | 13 (2.9%) | 7 (3.7%) | 8 (2.7%) |
| Relationship with classmates | ||||
| Good | 729 (79.8%) | 347 (79.9%) | 149 (80.1%) | 233 (79.5%) |
| Average | 172 (18.8%) | 83 (19.1%) | 36 (19.3%) | 53 (18.0%) |
| Poor | 12 (1.3%) | 4 (0.9%) | 1 (0.5%) | 7 (2.3%) |
| Number of close friends | ||||
| 0 | 44 (4.8%) | 10 (2.3%) | 6 (3.2%) | 28 (9.5%) |
| 1 ~ 2 | 233 (25.5%) | 113 (26%) | 56 (30.1%) | 64 (21.8%) |
| 3 ~ 5 | 438 (48.0%) | 220 (50.6%) | 72 (38.7%) | 146 (49.8%) |
| ≥6 | 198 (21.7%) | 91 (20.9%) | 52 (27.9%) | 55 (18.7%) |
| Sleep disorders, yes | 179 (19.6%) | 78 (17.9%) | 39 (20.9%) | 62 (21.1%) |
| Smoking | 132 (14.5%) | 54 (12.4%) | 23 (12.3%) | 55 (18.7%) |
| Drinking | 208 (22.8%) | 105 (24.1%) | 42 (22.5%) | 61 (20.8%) |
| Physical inactivity, yes | 444 (48.6%) | 222 (51.1%) | 90 (48.3%) | 132 (45.0%) |
*SD standard deviation
Of the 913 nursing in the total sample, 286 (31.3%) were self-report SVA addicts. The vast majority of them were female (78.0%), with a mean (SD) age of 21.13 (2.66) years. 42.8% of them were freshmen students. The proportion of residence in urban and rural areas was almost half and half. Approximately half of them were only children. 18.40% of them were from a single-parent family. Most are Undergraduate and Junior college students. Over half of the students self-reported relatively high and very high academic stress. Most had an average academic performance. Prevalence of social interaction anxiety, depressive symptoms, anxiety symptoms, and perceived stress were 39.3%, 22.0%, 44.4%, and 24.1%, respectively. The vast majority had a good relationship with their mother (89.3%), father (74.0%), and classmates (79.8%). Most had 3 ~ 5 close friends. 19.6% students had self-report sleep disorders. 14.5%, 22.8%, and 48.6% students had smoking, alcohol drinking, and physical inactivity lifestyle.
Variable screening and model construction based on training cohort
As shown in Table 2, in the training cohorts, those from a single-parent family, having a higher academic stress, social interaction anxiety, depressive, anxiety, and perceived stress symptoms, less close friends, and sleep disorders had higher risk to develop SVA. A distinct distribution of academic performance profile was also observed between SVA and non-SVA groups. No significant difference was found in the other candidate variables between these two groups. As a result, a total of nine variables were included in the original model. Subsequently, these nine variables were further reduced to seven following LASSO regression analysis (Fig. 1). These seven variables were further subjected to a multivariate logistic regression analysis. As a result, it was shown that a total of six predictors were independently associated with SVA. These six predictors were academic stress, social interaction anxiety, sleep disorders, depressive symptoms, anxiety symptoms, and whether from a single-parent family (Table 3). Based on these predictors, a user-friendly nomogram was created (Fig. 2). The total point represented a continuum of SVA risk score and a corresponding risk to develop SVA (the bottom line).
Table 2.
Univariate analysis of factors associated with SVA in the training cohort
| Variable | SVA Group N = 119 |
Non-SVA group N = 315 |
p-value |
|---|---|---|---|
| Age, year | 21.18 (2.33) | 20.89 (2.13) | 0.228 |
| Sex, female | 95 (79.8%) | 249 (79.0%) | 0.858 |
| Grade | 0.216 | ||
| First year | 48 (40.3%) | 159 (50.5%) | |
| Second year | 32 (26.9%) | 79 (25.1%) | |
| Third year | 22 (18.5%) | 47 (14.9%) | |
| Fourth year | 17 (14.3%) | 30 (9.5%) | |
| Residence | 0.777 | ||
| Urban areas | 60 (50.4%) | 154 (48.9%) | |
| Rural areas | 59 (49.6%) | 161 (51.1%) | |
| Only child, yes | 56 (47.1%) | 151 (47.9%) | 0.871 |
| From a single-parent family, yes | 37 (31.1%) | 38 (12.1%) | < 0.001 |
| BMI, normal | 91 (76.5%) | 219(69.5%) | 0.154 |
| Education | 0.270 | ||
| Junior college | 46 (38.7%) | 149 (47.3%) | |
| Undergraduate | 57 (47.9%) | 131 (41.6%) | |
| Postgraduate | 16 (13.4%) | 35 (11.1%) | |
| Academic stress | < 0.001 | ||
| Average and below | 35 (29.4%) | 161 (51.1%) | |
| Relatively high | 64 (53.8%) | 134 (42.5%) | |
| Very high | 20 (16.8%) | 20 (6.3%) | |
| Academic performance | 0.005 | ||
| Bad | 14 (11.8%) | 18 (5.7%) | |
| Average | 73 (61.3%) | 241 (76.5%) | |
| Good | 32 (26.9%) | 56 (17.8%) | |
| Social interaction anxiety, yes | 70 (58.8%) | 110 (34.9%) | < 0.001 |
| Depressive symptoms, yes | 48 (40.3%) | 47 (14.9%) | < 0.001 |
| Anxiety symptoms, yes | 81 (68.1%) | 111 (35.2%) | < 0.001 |
| Stress, yes | 47 (39.5%) | 58 (18.4%) | < 0.001 |
| Relationship with mother | 0.394 | ||
| Good | 107 (89.9%) | 291 (92.4%) | |
| Average | 12 (10.1%) | 22 (7.0%) | |
| Poor | 0 (0%) | 2 (0.6%) | |
| Relationship with father | 0.081 | ||
| Good | 86 (72.3%) | 255 (81.0%) | |
| Average | 30 (25.2%) | 50 (15.9%) | |
| Poor | 3 (2.5%) | 10 (3.1%) | |
| Relationship with classmates | 0.555 | ||
| Good | 93 (78.2%) | 254 (80.6%) | |
| Average | 24 (20.2%) | 59 (18.7%) | |
| Poor | 2 (1.7%) | 2 (0.6%) | |
| Number of close friends | 0.027 | ||
| 0 | 8 (6.7%) | 8 (2.5%) | |
| 1 ~ 2 | 37 (31.1%) | 73 (23.2%) | |
| 3 ~ 5 | 56 (47.1%) | 161 (51.1%) | |
| ≥6 | 18 (15.1%) | 73 (23.1%) | |
| Sleep disorders, yes | 43 (36.1%) | 35 (11.1%) | < 0.001 |
| Smoking | 15 (12.6%) | 39 (12.4%) | 0.950 |
| Drinking | 35 (29.4%) | 67 (21.2%) | 0.075 |
| Physical inactivity, yes | 63 (52.9%) | 159 (50.5%) | 0.648 |
*SVA short video addiction
Fig. 1.
Results of Lasso regression. (A) Lasso Regression Coefficient Path Plot; (B) Lasso Regression CrossValidation Plot
Table 3.
Multivariate logistic regression analysis for SVA
| Characteristic | N | Event N | OR | 95%CI | p-value |
|---|---|---|---|---|---|
| Academic stress | |||||
| Average and below | 196 | 35 | — | — | |
| Relatively high | 198 | 64 | 1.726 | 1.001, 2.976 | 0.049 |
| Very high | 40 | 20 | 3.755 | 1.574, 8.961 | 0.003 |
| Social interaction anxiety | |||||
| No | 254 | 49 | — | — | |
| Yes | 180 | 70 | 2.595 | 1.574, 4.278 | < 0.001 |
| Sleep disorders | |||||
| No | 356 | 76 | — | — | |
| Yes | 78 | 43 | 3.642 | 2.013, 6.589 | < 0.001 |
| Depressive symptoms | |||||
| No | 339 | 71 | — | — | |
| Yes | 95 | 48 | 2.098 | 1.193, 3.687 | 0.010 |
| Anxiety symptoms | |||||
| No | 242 | 38 | — | — | |
| Yes | 192 | 81 | 2.802 | 1.674, 4.690 | < 0.001 |
| From a single-parent family | |||||
| No | 359 | 82 | — | — | |
| Yes | 75 | 37 | 2.407 | 1.323, 4.378 | 0.004 |
*SVA short video addiction, OR odds ratio, CI confidence interval
Fig. 2.
Nomogram prediction model for short-video addiction
Model performance assessment
The model performance based on the three cohorts (training, internal validation, and external validation) were evaluated based on model AUC, calibration curve and DCA curve. As shown in Fig. 3, the AUC for these three cohorts were 0.809, 0.831, and 0.839, indicating high discriminative power of our model. The calibration curves of this model were all extremely close to the ideal curve (Fig. 4), with the good model fit as reflected by Hosmer-Lemeshow test (P = 0.475, 0.294, and 0.335 in the three cohorts, respectively), indicating a high consistence between the predicted results and actual results. The DCA curves suggested the satisfactory clinical usefulness of our model (Fig. 5). The subgroup analysis revealed that this model had high discriminative power in both sexes, with AUC in males being 0.863, 0.833, and 0.870, and in females being 0.793, 0.822, and 0.840.
Fig. 3.
ROC curves of the nomogram prediction model
Fig. 4.
Calibration curve of the nomogram prediction model for the (A) training cohort, (B) internal validation cohort, and (C) external validation cohort
Fig. 5.
Decision curve analysis of the nomogram for the (A) training cohort, (B) internal validation cohort, and (C) external validation cohort
Discussion
In the present study, we for the first time have constructed a nomogram to predict the SVA risk among Chinese nursing students using a total of six variables, which stood up from the initial 22 candidate variables determined from the perspective of individual characteristics, family atmosphere, school atmosphere, and unhealthy lifestyle. According to our results, academic stress, sleep disorders, depression, anxiety, being from a single-parent family, and social anxiety were significant predictors for SVA among Chinese nursing students. This study has contributed to the existing literatures regarding the related factors of SVA risk by adding new evidence and provided a visual nomogram graph with satisfactory accuracy and discrimination to incorporated intangible risk factors and convert them into tangible and quantifiable scores.
Compared to the 21.6% prevalence of SVA among general Chinese college students based on the SVAS and its diagnostic criteria [20], a much higher SVA prevalence of 31.3% in our present study and 30.2% elsewhere [19] were seen among Chinese nursing students using the same scale and diagnostic criteria. This might be primarily accounted for by the gender difference in cognitive style, as compared with general college students, females make up the majority of nursing students such as 78.0% in our total sample. Generally, females tend to exhibit higher attentional bias toward negative information [14] when compared with males, who are relatively more easily to block negative stimuli via effortful control [43]. In line with this, it was revealed that nursing students were more likely to perceive more stress and anxiety during the COVID-19 pandemic than general adults or students [44]. Studies have shown that emotional disorders are critical etiology of SVA [1, 24]. Another explanation regarding this high prevalence of SVA among nursing students might be due to their relatively lower core self-evaluation due to their involuntarily choosing and pursuing nursing profession [26, 45], as students with low core self-evaluation tend to develop SVA [26].
Our study showed that out of the six salient predictors of SVA, four (depression, anxiety, social interaction anxiety, and academic stress) were of emotional disorders, suggestive of the importance of mental health problems in the development of SVA among nursing students. This is in line with existing studies showing that SVA was generated by emotional distress [24, 46, 47]. For example, a recent study by Yang et al. showed that depression and anxiety mediated the influence of parental phubbing on adolescents’ SVA [47]. Depression is even considered to be the common basis for diverse addictive behaviors [48]. Meanwhile, addictive users of TikTok self-reported a higher level of academic stress compared with the moderate and non-users [4]. Social anxiety was also confirmed to be a non-negligible factor for internet addiction in young adults [49]. In compatible with this, a prior study showed that classmate relationship was negative related to SVA [50]. It is worth noting that there might exist a vicious cycle between SVA and negative emotions, as evidenced by the bidirectional causality between these two phenomena [46] and the co-morbidity between anxiety, depression and SVA [51]. Social anxiety was also found to be a mediator between SVA and sleep disorders [52]. The complex interaction between anxiety, depression, academic stress, and social anxiety, and SVA might be accounted for by the combination of the Interaction of Person-Affect-Cognition-Execution (I-PACE) Model [53], compensatory internet use (CIU) theory [54] and he model of self-regulatory failure (MSF) [55]. Specifically, it is reckoned that high mentally distressed individuals may hold cognitive bias in expectation on virtual platform (I-PACE model) and are thus inclined to resort virtual social media platforms to avoid face-to-face awkwardness, embarrassment, loneliness as well as other negative emotions in the real life (CIU theory), this ostrich strategy, however, is destined to be ineffective and to make matters worse, even harmful, as it would result in addictions to the virtual platforms, especially the more addictive short videos, and further aggravate diverse negative emotions, forming a vicious cycle.
Nursing students are prone to sleep disorder [56]. Our results also identified sleep disturbance as another independent predictors for SVA. This was consistent with the phenomenon of concurrence of social media use (including short video addiction) and sleep problems [52, 57]. On the one hand, online activities and problematic social media use would impair individuals’ offline activities, including sleep [58, 59]. Specifically, it was shown that SVA would lead to a decrease of sleep quality among adolescents and college students via social anxiety, physical activity, and procrastination behavior [35, 52]. On the other hand, sleep problem might be a potential risk for SVA, considering that sleep problem is a salient contributor to diverse adverse emotion responses [60] and SVA is highly related to emotional problems [24]. In addition, affected core self-evaluation of nursing students might be another mechanism underlying the effect of sleep disorders on SVA, given the negative effect of sleep disorder on core self-evaluation in nursing students [45] and the contribution of low core self-evaluation on the development of SVA [26]. Our results along with others suggest that sleep disorders and SVA are frequently concurrent and the influence might be mutual, which underscores the critical necessity to simultaneous interventions targeting to both so as to block their vicious interaction.
Whether from a single-parent family also played an important role in SVA predictive models. As far as we know, this is the first report regarding the adverse effect of single-parent on SVA among the Chinese youth, which adds new insight into the understanding the etiology of SVA. Although there is a lack of similar researches for comparison, the phenomenon that those from single-parent families have a higher risk of internet addiction [61] supports our view. Sing-parent family represents one type of diverse forms of family disruptions, which are important stressors for the family and its members [62, 63]. Generally, individuals from divorced family tend to report high emotional distress [64]. In addition, single-family family is associated with increased parent-child conflicts [65]. As argued by interpersonal acceptance-rejection theory [66], behaviors that destroy parent-child relationship would result in a series of negative mental outcomes among the offsprings such as anxiety and depression, and unhappiness, thereby inducing SVA [67]. Therefore, more attention should be paid to those from single-parent families when performing SVA prevention and intervention, which is particularly important in contemporary China which has witnessed a surge in single-parent families due to continuous divorce rate [68].
Nomogram is now widely used in clinical research to predict the probability of certain clinical outcomes. We in this study for the first time have developed the predictive nomogram for SVA. This graphical model was visual and user-friendly, and the variables incorporated in the nomogram are easily obtainable, intervenable, and/or modifiable, which would facilitate its application in clinical settings. In addition, the model performance AUC, calibration curve and DCA curve indicate that the constructed model has high discriminative power, accuracy and clinical usefulness. All these properties would assist healthcare providers or educators to carry out early screening and intervention of SVA high-risk nursing students.
This study had several limitations. First, the cross-sectional design prevents from drawing a clear-cut causality among variables in this study. A prospective study design is warrant to confirm the findings in this study. Second, convenient sampling method would affect the representativeness of the study sample. A random probability sampling strategy is needed to explore this issue in the future. Third, although this study has endeavored to find the most relevant predictors via systemic screening among a host of candidate variables, some potentially salient variables would be inevitably omitted, e.g., personality, which was found to be significant correlated with SVA in a previous study [26]. Fourth, the use of self-administered questionnaires would lead to potential information bias. Finally, the model construction and validation was performed based on the respondents from two cities, which would limit the extrapolation our findings in the present study.
Conclusion
In summary, our present study developed a predictive nomogram for SVA among Chinese nursing students via integration of six salient predictors, including academic stress, sleep disorders, depression, anxiety, whether from a single-parent family, and social interaction anxiety. The model performed well in its accuracy, discriminative power, and clinical usefulness. It is potentially useful for universities and educators in identifying nursing students more likely to develop SVA and in developing tailored preventive interventions to reduce the prevalence of this condition.
Acknowledgements
Not applicable.
Author contributions
LYT, WFX, MJC, and HLD, conceptualization; LYT and WFX, methodology; LYT, WFX, and MJC, formal analysis; WFX and MJC, investigation; LYT, WFX, and MJC, writing-original draft preparation; LYT, HLD, and WFX, writing-review and editing; HLD and MJC, supervision. All authors have read and approved the final manuscript.
Funding
Not applicable.
Data availability
The datasets used during the current study were available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This survey was anonymous and subjects voluntarily participated in it. They had the right to withdraw this survey at any time. This study was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all the participants. The study protocol was in accordance with relevant guidelines and regulations and was approved by the Ethics Committee of Jinzhou Medical University (JZMUL20240724).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Zhao Z, Kou Y. Effects of loneliness on short video addiction among college students: the chain mediating role of social support and physical activity. Front Public Health. 2024;12:1484117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Montag C, Yang H, Elhai JD. On the psychology of TikTok use: A first glimpse from empirical findings. Front Public Health. 2021;9:641673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Smith T, Short A. Needs affordance as a key factor in likelihood of problematic social media use: validation, latent profile analysis and comparison of TikTok and Facebook problematic use measures. Addict Behav. 2022;129:107259. [DOI] [PubMed] [Google Scholar]
- 4.Chao M, et al. TikTok use and psychosocial factors among adolescents: comparisons of non-users, moderate users, and addictive users. Psychiatry Res. 2023;325:115247. [DOI] [PubMed] [Google Scholar]
- 5.Shao, Guosong. Understanding the appeal of user-generated media: a uses and gratification perspective. Internet Res Electron Netw Appl Policy. 2009;19(1):7–25.
- 6.Wang YH, Gu TJ, Wang SY. Causes and Characteristics of Short Video Platform Internet Community Taking the TikTok Short Video Application as an Example. in 2019 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-TW). 2020.
- 7.Ko HC, Kuo FY. Can blogging enhance subjective well-being through self-disclosure? Cyberpsychol Behav. 2009;12(1):75–9. [DOI] [PubMed] [Google Scholar]
- 8.Guo J, Chai R. Adolescent short video addiction in china: unveiling key growth stages and driving factors behind behavioral patterns. Front Psychol. 2024;15:1509636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gao Y, et al. Neuroanatomical and functional substrates of the short video addiction and its association with brain transcriptomic and cellular architecture. NeuroImage. 2025;307:121029. [DOI] [PubMed] [Google Scholar]
- 10.Liao M. Analysis of the causes, psychological mechanisms, and coping strategies of short video addiction in China. Front Psychol. 2024;15:1391204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhu C, et al. The relationship between short-form video use and depression among Chinese adolescents: examining the mediating roles of need gratification and short-form video addiction. Heliyon. 2024;10(9):e30346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Xie J, et al. The effect of short-form video addiction on undergraduates’ academic procrastination: a moderated mediation model. Front Psychol. 2023;14:1298361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ye JH, et al. Effects of short video addiction on the motivation and Well-Being of Chinese vocational college students. Front Public Health. 2022;10:847672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wu Y, et al. Gender differences in the relationship between Short-Form video addiction and adolescent depression: the mediating role of attentional Bias. Cyberpsychol Behav Soc Netw. 2025. [DOI] [PubMed]
- 15.Kim H, et al. The contributions of social comparison to social network site addiction. PLoS ONE. 2021;16(10):e0257795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bissonette Mink D, Szymanski DM. TikTok use and body dissatisfaction: examining direct, indirect, and moderated relations. Body Image. 2022;43:205–16. [DOI] [PubMed] [Google Scholar]
- 17.Pan W, et al. Female users’ TikTok use and body image: active versus passive use and social comparison processes. Cyberpsychol Behav Soc Netw. 2023;26(1):3–10. [DOI] [PubMed] [Google Scholar]
- 18.Liang X, Guo X, Zhao T. The impact of short-video use on depression of female undergraduate: A chain mediating model of self objectification and body satisfaction. 2020;43(05):1220–6.
- 19.Guo M, et al. Mediating effect of loneliness between short video addiction and sleep quality among nursing undergraduates. J Nursing(China). 2023;30(7):20–4. [Google Scholar]
- 20.Li X, et al. Relationship between short video addiction symptom and personality trait among college students. Chin Ment Health J. 2021;35(11):725–8. [Google Scholar]
- 21.Zarandona J, et al. With a smartphone in one’s pocket: A descriptive cross-sectional study on smartphone use, distraction and restriction policies in nursing students. Nurse Educ Today. 2019;82:67–73. [DOI] [PubMed] [Google Scholar]
- 22.Ciydem E. The relationship between difficulties in emotion regulation and solution-focused thinking in nursing students. Nurse Educ Pract. 2024;79:104088. [DOI] [PubMed] [Google Scholar]
- 23.Weurlander M, et al. How do medical and nursing students experience emotional challenges during clinical placements? Int J Med Educ. 2018;9:74–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zhang L, et al. The relationship between personality and short video addiction among college students is mediated by depression and anxiety. Front Psychol. 2024;15:1465109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hu H, Huang M. How stress influences short video addiction in china: an extended compensatory internet use model. Front Psychol. 2024;15:1470111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ding J, et al. The relationships between short video addiction, subjective well-being, social support, personality, and core self-evaluation: a latent profile analysis. BMC Public Health. 2024;24(1):3459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Feng T, et al. The relationships between mental health and social media addiction, and between academic burnout and social media addiction among Chinese college students: A network analysis. Heliyon. 2025;11(3):e41869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yan J, et al. Establishment and validation of a nomogram for suicidality in Chinese secondary school students. J Affect Disord. 2023;330:148–57. [DOI] [PubMed] [Google Scholar]
- 29.Pavlou M, et al. Review and evaluation of penalised regression methods for risk prediction in low-dimensional data with few events. Stat Med. 2016;35(7):1159–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Peduzzi P, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373–9. [DOI] [PubMed] [Google Scholar]
- 31.Qin H, et al. The preliminary development of short video addiction scale among college students. Psychol China. 2019;1(8):586–98. [Google Scholar]
- 32.Leung L et al. Leisure Boredom, Sensation Seeking, Self-esteem, Addiction Symptoms and Patterns of Mobile Phone Use. 2007.
- 33.Young K. Internet Addiction: The emergence of a new clinical disorder. In the 104th annual meeting of the American Psychological Association in Toronto, Canada. August 20, 1996. 1996.
- 34.Qin H. The study on effect mechanism and intervention of short-video addicition among college students. Jiangxi Science and Technology Normal University. 2022.
- 35.Zhao Z, Kou Y. Effect of short video addiction on the sleep quality of college students: chain intermediary effects of physical activity and procrastination behavior. Front Psychol. 2023;14:1287735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the depression anxiety stress scales (DASS) with the Beck depression and anxiety inventories. Behav Res Ther. 1995;33(3):335–43. [DOI] [PubMed] [Google Scholar]
- 37.Cao CH et al. The psychometric properties of the DASS-21 and its association with problematic internet use among Chinese college freshmen. Healthc (Basel). 2023;11(5). [DOI] [PMC free article] [PubMed]
- 38.Buysse DJ, et al. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. [DOI] [PubMed] [Google Scholar]
- 39.Liu XC, Tang MQ. Reliability and validity of the Pittsburgh sleep quality index. Chin J Psychiatry. 1996;29(2):29103–7. [Google Scholar]
- 40.Mattick RP, Clarke JC. Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behav Res Ther. 1998;36(4):455–70. [DOI] [PubMed] [Google Scholar]
- 41.Ye D, et al. Revision of social interaction anxiety scale and social phobia scale. Chinese Journal of Clinical Psychology. 2007.
- 42.Mörtberg E, et al. Factor solutions of the social phobia scale (SPS) and the social interaction anxiety scale (SIAS) in a Swedish population. Cogn Behav Ther. 2017;46(4):300–14. [DOI] [PubMed] [Google Scholar]
- 43.Pintzinger NM, et al. Attentional biases in healthy adults: exploring the impact of temperament and gender. J Behav Ther Exp Psychiatry. 2016;52:29–37. [DOI] [PubMed] [Google Scholar]
- 44.Bai W, et al. Network analysis of anxiety and depressive symptoms among nursing students during the COVID-19 pandemic. J Affect Disord. 2021;294:753–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Ning L, et al. The effect of sleep problems on core self-evaluations in undergraduate nursing students and the role of emotion regulation and resilience: A cross-sectional study. J Prof Nurs. 2024;51:64–73. [DOI] [PubMed] [Google Scholar]
- 46.Zhang D, Yang Y, Guan M. A cross-lagged analysis of the relationship between short video overuse behavior and depression among college students. Front Psychol. 2024;15:1345076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Yang C, et al. Association between parental phubbing and short-form video addiction: A moderated mediation analysis among Chinese adolescents. J Affect Disord. 2025;369:523–30. [DOI] [PubMed] [Google Scholar]
- 48.Flores Mosri D. Affective features underlying depression in addiction: Understanding what it feels like. Front Psychol. 2019;10:2318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Weinstein A, et al. Internet addiction is associated with social anxiety in young adults. Ann Clin Psychiatry. 2015;27(1):4–9. [PubMed] [Google Scholar]
- 50.Liu QQ, Yang XJ, Nie YG. Interactive effects of cumulative social-environmental risk and trait mindfulness on different types of adolescent mobile phone addiction. Curr Psychol. 2022.
- 51.Ho RC, et al. The association between internet addiction and psychiatric co-morbidity: a meta-analysis. BMC Psychiatry. 2014;14:183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Jiang L, Yoo Y. Adolescents’ short-form video addiction and sleep quality: the mediating role of social anxiety. BMC Psychol. 2024;12(1):369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Brand M, et al. Integrating psychological and Neurobiological considerations regarding the development and maintenance of specific Internet-use disorders: an interaction of Person-Affect-Cognition-Execution (I-PACE) model. Neurosci Biobehav Rev. 2016;71:252–66. [DOI] [PubMed] [Google Scholar]
- 54.Elhai JD, Levine JC, Hall BJ. The relationship between anxiety symptom severity and problematic smartphone use: A review of the literature and conceptual frameworks. J Anxiety Disord. 2019;62:45–52. [DOI] [PubMed] [Google Scholar]
- 55.Acuff SF, et al. Temporal precedence of self-regulation over depression and alcohol problems: support for a model of self-regulatory failure. Psychol Addict Behav. 2019;33(7):603–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Marta OFD, et al. Gender differences in the relationships between sleep disturbances and academic performance among nursing students: A cross-sectional study. Nurse Educ Today. 2020;85:104270. [DOI] [PubMed] [Google Scholar]
- 57.Scott H, Biello SM, Woods HC. Social media use and adolescent sleep patterns: cross-sectional findings from the UK millennium cohort study. BMJ Open. 2019;9(9):e031161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Espinoza G, Juvonen J. The pervasiveness, connectedness, and intrusiveness of social network site use among young adolescents. Behav Social Netw. 2011;14(12):705–9. Cyberpsychology. [DOI] [PubMed]
- 59.Wong HY et al. Relationships between severity of internet gaming disorder, severity of problematic social media use, sleep quality and psychological distress. Int J Environ Res Public Health. 2020;17(6). [DOI] [PMC free article] [PubMed]
- 60.Gül H, Haney M. Associations of health literacy and sleep quality with depressive symptoms in Turkish adolescents: A descriptive cross-sectional study. J Pediatr Nurs. 2024;79:9–15. [DOI] [PubMed] [Google Scholar]
- 61.Ni X, et al. Factors influencing internet addiction in a sample of freshmen university students in China. Cyberpsychol Behav. 2009;12(3):327–30. [DOI] [PubMed] [Google Scholar]
- 62.D’Onofrio B, Emery R. Parental divorce or separation and children’s mental health. World Psychiatry. 2019;18(1):100–1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Kang Y, Nielson T. Helping postdivorce families enhance Parent-Child relationships: clinical applications to effectively help divorced parents appropriately communicate their new romance with children. J Marital Fam Ther. 2025;51(1):e12764. [DOI] [PubMed] [Google Scholar]
- 64.Laumann-Billings L, Emery RE. Distress among young adults from divorced families. J Fam Psychol. 2000;14(4):671–87. [DOI] [PubMed] [Google Scholar]
- 65.WANG M, et al. Influence of family structure on adolescents’ health behaviors. China Sport Sci. 2012;32(5):34–41. [Google Scholar]
- 66.Rohner RP, Lansford JE. Deep structure of the human affectional system: introduction to interpersonal Acceptance-Rejection theory: interpersonal Acceptance-Rejection theory. J Family Theory Rev. 2017;9(4):426–40. [Google Scholar]
- 67.Mu H et al. Drivers and consequences of Short-Form video (SFV) addiction amongst adolescents in china: Stress-Coping theory perspective. Int J Environ Res Public Health. 2022;19(21). [DOI] [PMC free article] [PubMed]
- 68.Chen M, Rizzi EL, Yip PSF. Divorce trends in China across time and space: an update. Asian Popul Stud. 2021;17(2):121–47. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used during the current study were available from the corresponding author on reasonable request.