A cross-sectional descriptive analysis of technology addiction in adolescents: associations with food addiction, emotional eating, and body weight status

Abstract

Background

This cross-sectional descriptive study aimed to determine the prevalence and risk factors of technology addiction (TA) in adolescents, as well as investigate the association of TA with food addiction and emotional eating by body weight status.

Methods

Adolescents (n = 1388) completed a questionnaire that featured socio-demographic characteristics, the Game Addiction Scale (GAS), the Social Media Disorder Scale (SMD), the dimensional Yale Food Addiction Scale for Children 2.0 (dYFAS‐C 2.0), and the Emotional Eating Scale Adapted for Use in Children and Adolescents (EES-C). The body mass index (BMI) Z-score was classified according to the World Health Organization.

Results

TA was present in one-fifth of adolescents, and boys were four times more likely to develop a digital game addiction (p < 0.001). A 1-point increase in the GAS score is associated with a 1.08-point increase in the dYFAS-C-2.0 score and a 0.5-point increase in the total EES-C score (p < 0.001). A 1-point increase in the SMD score was also related to an increased of 1.21 and 1.26, respectively (p < 0.001). All of these positive associations were significant in adolescents with overweight (p < 0.001).

Conclusions

Given the rapidly increasing prevalence of TA among adolescents, its association with food addiction, emotional eating, and body weight status is worrisome, and our findings shed light on the relevance of developing strategies to reduce the prevalence of TA in this population.

Introduction

Today, there are 5.16 billion internet users in the world; this is an indication that 64.4 percent of the total population is now online [1]. The rapid development of information technology has led to the widespread adoption of versatile portable media tools, including smartphones and tablets, particularly among young individuals [2]. Adolescents' extensive use of the internet for entertainment, socialization, and information has led to increased online time, raising concerns about potential consequences [3, 4]. The increasing utilization of technology can result in issues like technology addiction (TA) [5]. While there remains a lack of agreement concerning the existence of standardized diagnostic guidelines for TA, healthcare institutions have begun to recognize that problematic digital device utilization leads to adverse effects on physical and mental health, significantly stronger for adolescents [2, 6]. The most frequently encountered mental issues include behavioral and emotional disorders [7]. For instance, studies have shown that social media use has negative effects on depression and mental health in adolescents [8, 9]. Another study revealed that appearance-related social media addiction affected three-quarters of young women, leading to higher levels of body surveillance, low body esteem, body comparison, and depressive symptoms [10]. In addition, prolonged computer usage can result in emotional overeating and obesity due to physical inactivity, visual and musculoskeletal concerns, as well as issues related to social development [11].

Eating disorders have attracted global attention, particularly among adolescents, due to their connection with social, cultural, and psychological factors influencing eating attitudes and behaviors [12]. These mental health conditions, marked by maladaptive eating behaviors and body image concerns, have risk factors such as early dieting, extreme weight loss practices, depression, body dissatisfaction, and social media influences [13, 14]. The use of image-focused social media platforms is associated with eating disorder pathology as it promotes heightened comparison of individuals' physical appearances to others, potentially resulting in amplified body dissatisfaction and compromised eating habits [15, 16]. For example, social media use may influence emotional eating [17], a coping mechanism in adolescents that involves eating in response to negative emotions like depression, fear, anxiety, or stress [11].

Moreover, digital game addiction contributes to a negative effect on the psychological states of adolescents [18]. A recent study concluded that digital game addiction leads to aggression and high levels of anger in adolescents [19]. Another study found that digital game addiction and obesity increased social anxiety in adolescents [20]. Adolescents who play digital games can spend extended periods in front of the screen, a behavior that is also associated with decreased dopamine levels [21]. The decrease in dopamine levels hurts psychological well-being, causing low mood and increased emotional eating behaviors [21, 22]. Emotional eating, which occurs as a reaction to negative emotions and is defined as overeating, often leads individuals to consume high-calorie foods as a coping mechanism, resulting in subsequent regret and weight gain [22]. Prolonged screen use can compound this issue by contributing to obesity due to physical inactivity [11].

Integration of modern technological devices into daily life has become increasingly common, simplifying various aspects of life while also posing challenges due to their misuse [11]. In light of the growing awareness of the potential for technology misuse, studies have shown that excessive use of the internet [23] and game addictions [11, 24] have some adverse effects on eating behavior in adolescents, but there is a scarcity of studies examining the potential effects of problematic social media use. Moreover, the lack of a holistic approach and comprehensive evaluations of the relationships between TA (including both digital games and social media addictions), impaired eating behaviors (including both food addiction and emotional eating), and obesity in a large sample of adolescents stands out. As a result, this study aimed to investigate the association of TA with food addiction and emotional eating by body weight status in a large sample of Turkish adolescents. The study's objectives were (1) to determine the prevalence of TA (digital games and social media addictions) in adolescents; (2) to explore risk factors of TA (i.e., age, sex, economic status, parental data, etc.); (3) to investigate the association of TA with food addiction and emotional eating; and (4) to examine whether the relationship between TA and impaired eating behaviors varies by body weight status (in normal-weight and overweight adolescents).

Methods

Participants and procedure

This cross-sectional descriptive study involved 1647 healthy adolescents in secondary schools. We informed administrators, school teachers, and parents about the research. We obtained written informed consent forms from the adolescents and their parents, in addition to verbal consent. We used face-to-face interview techniques and had students fill out the questionnaire in the classroom. In this study, we excluded those with food allergies or chronic diseases that could affect food intake, those on medication, and those on any diet. After excluding 259 participants due to not meeting eligibility requirements, 1388 adolescents, aged between 10 and 14 years, attended the study from seven secondary schools in Kayseri. Table 1 summarized the demographic characteristics of the participants. All procedures were conducted by the Declaration of Helsinki, and this study has been approved by the Erciyes University Social and Humanities Science Ethics Committee (approval number: 113).

Table 1 General participant characteristics in the whole sample and groups by the digital games and social media addiction

Measures

The questionnaire consisted of anthropometric measurements, demographic (i.e., age, sex, economic status, and parental data) and other health questions, the Game Addiction Scale (GAS), the Social Media Disorder Scale (SMD), the dimensional Yale Food Addiction Scale for Children 2.0 (dYFAS-C 2.0), and the Emotional Eating Scale Adapted for Use in Children and Adolescents (EES-C). We measured body weight, height, and waist circumference using previously described methods [25]. Body mass index (BMI-kg/m²) Z-score, which considers the common growth according to age and sex, was used. The BMI Z-score was classified according to the World Health Organization [26] as normal-weight if it was − 2 SD to > + 1 SD and as overweight > + 1 SD. The BMI Z-scores < − 2 SD were underweight/thin and excluded from the current study.

Game Addiction Scale The 7-item GAS was developed to assess pathological digital game addiction based on DSM-5 (salience, tolerance, mood modification, withdrawal, relapse, conflict, and problems) by Lemmens et al. [27]. The responses range from 1 (never) to 5 (very often), and a higher score indicates an increased risk of digital game addiction. Two formats were suggested to identify game addiction: a monothetic format (if all items score three or more) and a polythetic format (if four out of seven items score three or more) [27]. Yalcin Irmak and Erdogan [28] conducted a validity and reliability study for the Turkish version of GAS (α_Cronbach = 0.72).

Social Media Disorder Scale The 9-item scale was coined to measure social media addiction by Van den Eijnden et al. [29]. According to the DSM-5 definition [30], social media disorder, which is considered a specific form of internet addiction, is diagnosed if a person meets five or more of the nine criteria (preoccupation, tolerance, withdrawal, persistence, displacement, problems, deception, escape, and conflict) of SMD over 12 months. This scale has good reliability (α_Cronbach = 0.76 for the original scale and α_Cronbach = 0.86 for the Turkish version), utilizing a dichotomous (yes/no) response format and a well-defined diagnostic cut-off point [29, 31].

Dimensional Yale Food Addiction Scale for Children 2.0 The development of the 16-item dYFAS-C 2.0, which measures food addiction characteristics in children and adolescents based on the DSM-5 criteria, involved adapting the items from the adult YFAS 2.0 [32]. The dYFAS-C 2.0, like its adult form, guides participants to consider foods rich in refined carbohydrates or fats when responding to its questions, as these foods are most closely associated with addictive-like eating behaviors. The dYFAS-C 2.0 rates each item on a 5-point Likert scale, ranging from never to always, with a higher total score indicating a more significant presence of food addiction behaviors [33]. The high internal consistency (α_Cronbach = 0.90 for the original scale and α_Cronbach = 0.90 for the Turkish version) indicates strong reliability for the 16-item scale [33, 34].

Emotional Eating Scale Adapted for Use in Children and Adolescents The emotional eating status of adolescents was evaluated using the 25-item EES-C [35], providing a total emotional eating score as well as scores of three subscales: eating in response to anxiety, anger, and frustration (EES-C-AAF), depressive symptoms (EES-C-DEP), and feeling unsettled (EES-C-UNS). The response options ranged from one (“I never want to eat”) to five (“I want to eat a lot”). Higher scores indicate higher levels of emotional eating, or its subscales. The validity and reliability of EES-C were determined by Bektas et al. [36] in Turkey (α_Cronbach = 0.90).

Statistical analyses

Power analysis Sample power was calculated using the statistical software G*Power (version 3.1) for primary outcomes. The sample size of 1385 participants provided 99.9% power at an alpha level of 0.05 for the association between the GAS score with dYFAS‐C 2.0 and EES-C total scores of adolescents adjusted for potential confounders obtained by linear multiple regression. In addition, when sample power was estimated using the SMD score about the dYFAS‐C 2.0 and EES-C total scores adjusted for potential confounders, the sample size of 479 participants provided more than 95% power for both parameters (α = 0.05).

Data analysis Statistical analysis was performed using the IBM SPSS Statistics software (version 27.0). Data were expressed as the number (n) and percentage (%) for qualitative variables, and mean and standard deviation for quantitative variables. Continuous variables were examined for normal distribution and also skewness and kurtosis. When required, log-transformed before analysis and reported back-transformed geometric mean and standard error. According to scores from the GAS and SMD scales, participants were divided into two groups addicted and non-addicted. To assess the differences between groups, the chi-square test and Student’s t-test were used. Multivariable logistic regression analyses were performed to assess the risk factors of digital games or social media addiction, and the non-addicted group was considered the reference group. Reference categories of predictors were the girl for sex, employed for mother's job, retired for father's job, and moderate-income status. Age, BMI Z-score, and parents’ ages were a continuous variable. Odds ratios (OR) and 95% confidence intervals (CI) were reported. To investigate the association of the GAS and SMD scores (independent variables) with the dYFAS‐C 2.0 and EES-C scores (dependent variables), multivariable linear regression analyses were used, and regression models were adjusted for potential confounders. In the first adjusted model, age and sex were controlled. The further model was also adjusted for BMI Z-score, parents’ age and job, and income status. As the dYFAS‐C 2.0 score had a skewed distribution, it was log-transformed before regression analyses, and back-transformed β-coefficients were reported. The analyses were repeated by splitting according to body weight status. Both regression analyses were performed with the backward stepwise technique and started with a model consisting of all potential explanatory variable. Then any variables having a significant Wald test at a level of 0.25 were selected for the multivariate analysis and others with insignificant relationships were removed one-by-one. The overall model's fit was evaluated by comparing the modification of respective associations substantially and the Nagelkerke or adjusted R² values with and without each explanatory variable. Model assumptions were checked for any potential multicollinearity concerns, and no violation was found. For all statistical analyses, p < 0.05 was considered statistically significant.

Results

Participant characteristics

This study was conducted with 1388 adolescents, aged 12.6 ± 1.1 years, and half of them (53.4%) were boys. More than one-third of participants (41.7%) were overweight, and one-fifth of them exhibited digital games (20.5%) and social media (21.5%) addiction. The mean scores of the GAS, SMD, dYFAS‐C 2.0, and EES-C scales and sociodemographic characteristics describing the whole sample are presented in Table 1.

Table 1 also depicts the characteristics of participants based on sex and TA. While the percentages of overweight (47.1% vs. 35.5%) and digital game addiction (30.7% vs. 8.8%) were higher in boys, the emotional eating score in response to depressive symptoms was higher in girls (17.4 ± 5.1 vs. 18.0 ± 5.3) (p < 0.05). Participants with digital games or social media addictions had a higher age, BMI, dYFAS-C 2.0 score, and EES-C score than those with non-addiction (p < 0.05).

The factors affecting technology addictions

Factors (independent variables) associated with digital games and social media addictions (dependent variables) were constructed using multivariable logistic regression, and predictors of addiction are presented in Table 2. While higher age (p = 0.005 for both addictions) and BMI Z-score (p = 0.004 for digital game addiction and p = 0.006 for social media addiction) were associated with an increased risk of both addictions, the odds of digital game addiction were enhanced four times in boys (p < 0.001).

Table 2 Multivariable logistic regression analysis assessing the risk factors of digital games and social media addiction

Association of technology addictions with food addiction and emotional eating

Regression coefficients (β) and 95% CI for the association of the GAS and SMD scores with the dYFAS-C-2.0 and EES-C scores are given in Table 3. The multivariable linear regression analysis, considering the sociodemographic variables as confounders, confirmed that a greater TA was associated with higher food addiction and emotional eating. A 1-point increase in the GAS score is correlated with a 1.08-point increase in the dYFAS-C-2.0 score and a 0.5-point increase in the total EES-C score. An increase of 1 point in the SMD score also led to an increase in scores of 1.21 and 1.26, respectively. While both TAs showed the highest association with EES-C-AAF among the emotional eating subscales, the EES-C-DEP score had a higher correlation with the SMD score (β 0.501, 95% CI 0.275, 0.727; p < 0.001) than the GAS score (β 0.141, 95% CI 0.092, 0.190; p < 0.001).

Table 3 Multivariable linear regression analysis assessing the associations of the GAS and SMD scores with the dYFAS‐C 2.0 and EES-C scores

On the other hand, regression models were split by body weight status (normal-weight or overweight), and the best cubic equation accounted for the relationship between dependent and independent variables (Fig. 1). While there was no relationship between the SMD score and the EES-C score in normal-weight adolescents (p = 0.465), all positive associations of TAs with food addiction and emotional eating were statistically significant in adolescents with overweight (p < 0.001). Moreover, the relationships were stronger in adolescents with overweight than those with normal body weight, except for the association between the GAS and dYFAS-C-2.0 scores (R² = 0.168 in normal-weight adolescents and R² = 0.143 in adolescents with overweight). In addition, among adolescents with overweight, the SMD and dYFAS-C-2.0 scores had the strongest association and the regression model determined that social media addiction accounted for 22.4% of the variance in their food addiction (p < 0.001).

Fig. 1

Association of the GAS and SMD scores with the dYFAS‐C 2.0 and EES-C scores in groups by body weight status. dYFAS‐C 2.0, dimensional Yale Food Addiction Scale for Children 2.0; EES-C, Emotional Eating Scale Adapted for Use in Children and Adolescents; GAS, Game Addiction Scale; SMD, Social Media Disorder Scale. ^*Values are log-transformed variables

Discussion

The present study provides the first evaluation of the association of TA with food addiction and emotional eating by body weight status in a large sample of Turkish adolescents. The current findings revealed that adolescents with TAs (digital games or social media) had higher age, BMI, food addiction, and emotional eating scores than those with non-addiction. On the other hand, boys had fourfold higher odds of developing digital game addiction. Furthermore, TA was associated with increased food addiction and emotional eating, even after adjustment for potential confounders. All of these positive associations were also significant in adolescents with overweight, whereas there was no relationship between social media addiction and emotional eating in normal-weight adolescents. In addition, social media addiction had the strongest association with food addiction and accounted for 22.4% of its variance among adolescents with overweight. Given the rapidly increasing prevalence of TA among adolescents, its association with food addiction, emotional eating, and body weight status is worrisome, and our findings shed light on the relevance of developing strategies to reduce the prevalence of TA in this population.

The rapid advancement of information and communication technology has led to the widespread availability of multifunctional portable media devices, but this growing reliance on technology can also give rise to issues such as TA among young individuals [2, 5]. Studies revealed that 24.4% of adolescents were addicted to social media [37], while 22.4% were addicted to digital games [38]. Similarly, the present study determined that TA affected one fifth of the adolescents. Furthermore, the study identified high age and male gender as risk factors for TA. Supporting these findings, studies have found increasing age and BMI to be positively associated with TA (digital games or social media addictions) [39,40,41]. These findings suggest that younger adolescents have relatively limited use of products under parental control, such as phones and tablets, and draw attention to the importance of physical activity. The most significant risk factor for digital game addiction is gender [42], and this study found that boys were four times more likely than girls to develop digital game addiction in line with previous studies [19, 43]. The fact that boys use the internet for activities such as playing digital games or browsing websites, while girls prefer to use it for social networking sites, helps explain these results.

The influence of online social networks on adolescents is far more pronounced, as a majority of their social and emotional growth occurs through internet and mobile phone interactions [44]. Adolescents may develop negative attitudes toward body image, nutritional status, and eating behaviors as a result of social media addiction [45]. Previous studies have shown a correlation between heightened social media addiction in adolescents and an elevated susceptibility to eating disorders [45,46,47]. Gumus et al. [45] found that both social media addiction and duration of use were positively associated with eating disorders in adolescents. Some studies of adolescents have also indicated a relationship of emotional eating with social media addiction [48] and digital gaming addiction [11]. Furthermore, research has demonstrated a positive association between digital game addiction and both food addictions and unhealthy eating patterns [24]. Consistent with the literature, we found both TAs to be associated with increased food addiction and emotional eating, even after adjusting for potential confounding factors, with social media addiction, in particular, having the strongest association with food addiction. The cause of this phenomenon can be attributed to the fact that adolescents often use social media and are exposed to nutrition and health advice that lacks scientific validity. Social media posts also have the potential to adversely impact their mood, alter their eating behavior, and increase their desire to consume unhealthy foods.

The problematic use of technologies for communication and information can potentially result in obesity through the indirect consequences of reduced physical activity and altered eating behaviors [49]. Furthermore, obese individuals are more vulnerable to problematic internet usage, characterized by impulsive behavior, dependence, risk-taking, or impairment [50]. Previous studies have demonstrated a significant positive association between digital game addiction and BMI [20, 41] but there is a scarcity of studies examining the potential connection between obesity and problematic use of social media. Jolliff et al. [50] discovered a noteworthy correlation between obesity and social media addiction in young adults. Separate anxiety and depression symptoms influenced this connection. However, there is a lack of research examining the possible association of social media addiction with emotional eating and food addiction in adolescents with overweight. Strikingly, the present study showed that both digital games and social media addictions were positively associated with emotional eating and food addiction in adolescents with overweight, whereas there was no relationship between social media addiction and emotional eating in normal-weight adolescents. One possible explanation for this finding is that hedonic mechanisms may underlie both obesity and addiction [51]. Therefore, it is logical to examine the relationship between problematic technology use, impaired eating patterns, and obesity from a neurobiological perspective.

The study's strengths and limitations

This study stands out for its notable strengths. It is the first study in Turkey to examine the relationship between technological addictions (both digital games and social media addictions), food addiction, and emotional eating in adolescents. Additionally, it boasts a large sample size, setting it apart from previous studies conducted on this topic. Furthermore, these addictions were examined in a specific demographic, such as individuals who are overweight.

However, it is crucial to acknowledge the limitations of the current study. First, the study's cross-sectional design precludes the demonstration of causality among factors. Second, because we only collected data from secondary school students in Kayseri and excluded factors that may alter eating behavior, there may be potential for selection bias, limiting the generalizability of our results to the broader adolescent population. Lastly, we obtained data from adolescents using a self-reported questionnaire, as it allows collecting data from a larger number of participants and is less costly. Despite data cleaning prior to conducting analyses, the possibility of misreporting or recall bias due to self-reports cannot be dismissed. Therefore, our findings should be carefully interpreted.

Conclusion

The current study concludes that one-fifth of adolescents exhibit TA, and that boys are four times more likely to develop a digital game addiction. Another impressive result was the association of TA with food addiction and emotional eating, especially among adolescents with overweight. Considering these findings, adolescents are particularly vulnerable to developing addictions, and irregular eating habits and health-endangering behaviors occur at a higher rate during this period. While digital game addiction is more associated with the male gender, increasing BMI is associated with both social media addiction and digital content addiction. Additionally, female adolescents may be more likely to engage in emotional eating, a coping method.

Consequently, it is essential to create intervention strategies to address disordered eating attitudes and distinct forms of TA, including internet addiction, gaming addiction, and social media addiction, to provide expert assistance to those seeking help. Additionally, parents, educators, health professionals, and policymakers should collaborate on prevention initiatives. Parents should be informed about TA and possible eating disorders in adolescents. They may also impose restrictions on digital gaming, especially for children in early adolescence. Schools should raise adolescents awareness about TAs and disordered eating behaviors, and school counseling services should teach them coping mechanisms for these addictions and negative emotions. Furthermore, providing psychological support to children about addictions in schools may be an effective strategy to prevent eating disorders before they occur. Moreover, adopting practices that promote physical activity could help stop obesity that is the result of technological addictions.

For future research, it is advisable to employ a proactive approach to confirm our findings and develop strategies for improved identification and management of technological addictions and eating disorders in adolescents. We need longitudinal studies that could shed light on the long-term impacts of TA on eating behaviors.