Introduction

In recent years, the worldwide percentage of smartphone owners and users has increased steadily [1]. With features including, among others, communication, Internet, and multimedia [2], smartphones—not least because of their ease of access [3]—have several advantages such as productivity enhancement, facilitated information seeking [3], and heightened pleasure via social interactions [4].

Despite these benefits, however, a growing body of literature hints at negative consequences and possible dangers associated with smartphones [3, 5, 6]. These include excessive use [3], increasingly uncontrollable behaviors such as constantly checking for notifications [4], mental health problems such as depression and anxiety [3, 7], and physical problems [8]. It has been argued that problematic smartphone use can be viewed as a form of behavioral addiction like gaming addiction or Internet addiction [2, 4, 9]. Symptoms commonly associated with behavioral addictions, such as tolerance, withdrawal, mood dysregulation, cravings, and loss of control, have also been found to be related to problematic smartphone use [10]. Based on these findings, and similar to Internet addiction, Demirci et al. [2] have suggested that smartphone addiction is characterized by an overuse of smartphones that interferes with the users’ daily functioning.

Nevertheless, smartphone addiction is included neither in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [11] nor in the upcoming International Classification of Diseases 11th Revision (ICD-11) [12], although the DSM‑5 now lists diagnostic criteria for Internet gaming addiction with the need for further research [11], and the ICD-11 has included the diagnosis of (Internet) gaming disorder in its preliminary online version [12]. On the one hand, this is interpreted as an increasing awareness concerning the existence of smartphone addiction by some [9]. On the other hand, a recent review [6] concludes that, to date, evidence is not sufficient to support the existence of smartphone addiction.

Instead, the authors suggest the terms problematic or maladaptive smartphone use, which pertain to an excessive behavior with lower levels of impairment than addiction [6]. Excessive use is sometimes measured as duration of usage and usage frequency (e.g., see Bae [13]), and problematic use is considered an uncontrolled behavior leading to negative consequences in everyday life [14]. The terms problematic smartphone use and smartphone addiction seem to be used synonymously, based on the researchersʼ understanding of the underlying construct. Researchers who assume the observed behaviors meet addiction criteria seem to choose the term smartphone addiction (e.g., see Yen et al. [15]), whereas researchers who do not consider addiction criteria met choose to refer to problematic smartphone use [3, 6, 14].

Given this inconclusive terminology, synthesizing existing research regarding problematic smartphone use in children and adolescents is a challenge [16]. For instance, reported prevalence rates of children and adolescents with problematic smartphone use vary widely, from 5% [17] to about 50% [15]. This may be interpreted mainly as a result of the different operationalizations used in different studies. These include, among others, problematic phone use [15, 17], extensive use of mobile phones (e.g., see Sánchez-Martínez and Otero [18]), smartphone addiction risk (e.g., see Lee et al. [5] and Cha and Seo [9]), and smartphone addiction [19,20,21].

Similarly, research on risk factors for problematic smartphone use in children and adolescents has, to date, yielded inconclusive results. Overall, minors seem to be particularly vulnerable [5, 9], which could be related to difficulties in self-regulation [4] and immature control competencies [22]. Other factors possibly influencing maladaptive usage in children and adolescents include age [23,24,25], gender [10, 24, 25], social factors [24, 26], and personality [23, 25, 27, 28].

Despite the growing body of data, comprehensive reviews synthesizing key findings are still scarce. A meta-analysis [20] focuses only on India, and Park and Park [24] propose a model of smartphone addiction but without comparing different results and without considering age groups other than early childhood. Furthermore, as smartphone technology is advancing rapidly, new developments have arisen since 2014. Hence, a review focusing solely on mobile phone addiction seems to fall short. Therefore, this review sets out to provide an overview of studies on problematic or addictive smartphone use in children and adolescents, with the focus particularly on factors that increase the risk of problematic smartphone use.

Methods

Search strategy

In order to increase the likelihood of including studies that focus on smartphones instead of older kinds of mobile phones (without Internet access), only publications from 2008 onward were searched, as done by Elhai and colleagues [3] in their review on problematic smartphone use and anxiety and depression. A literature search was conducted in Google Scholar and PubMed regarding papers published between January 2008 and May 2019. Search parameters were PROBLEMATIC/MALADAPTIVE/EXCESSIVE/PATHOLOGICAL/DYSFUNCTIONAL in combination with PHONE/SMARTPHONE/SMART PHONE/CELLPHONE/CELL PHONE/MOBILE PHONE and ADDICTION/USE as well as ADOLESCENTS/CHILDREN/YOUTH. Google Scholar alerts were enabled to ensure the inclusion of accepted articles and articles in preprint status. The title, abstract, and main text of each study were examined independently by the authors, and exclusions of studies occurred at each stage of the process (see Fig. 1). Additionally, a reference search strategy was used to identify other relevant articles.

Fig. 1
figure 1

PRISMA flow diagram of the literature search

Full size image

Study selection process

Articles were included if they were original articles, written in English, published in peer-reviewed journals not earlier than 2008, and focused primarily on children (1–10 years) and adolescents (11–21 years). Although the search focused on smartphones, publications regarding problematic cell phone or mobile phone use were also included, as it can be assumed that cell/mobile phones were used from 2008 onward at least by some participants. Articles focusing on positive effects of smartphones as well as editorials were excluded. Titles and abstracts found in the search were screened for relevance before full-text articles were reviewed. Studies examining college students were also included if a clear distinction between age groups was possible. Furthermore, risk factors were defined as variables that predict problematic smartphone use/smartphone addiction.

Results

Sample of included studies

The initial search yielded 120 articles. Eighty-two articles were subsequently excluded because they did not or did not exclusively examine children/adolescents, or because they were not original articles, were not peer-reviewed, were not written in English, or only examined Internet addiction or media use in general. Articles focusing on consequences of smartphone addiction were also not considered. For a detailed description of the exclusion process, see Fig. 1.

The final sample consisted of 38 articles. Of the included studies, 42.1% (n = 16) were conducted in South Korea and 13.2% (n = 5) in Turkey. Other studies were from Taiwan, China, Switzerland, Italy (each of these accounting for 5.3%), Germany, India, Poland, and Romania (each of these accounting for 2.6%). Two studies (5.3%) were conducted in Spain and one (2.6%) in the UK, and two additional studies (5.3%) were conducted both in Spain and the UK.

The Smartphone Addiction Scale–Short Version (SAS–SV) [10] or the longer version of the SAS [29] was used in 26.3% (n = 10) of the studies, and the Smartphone Addiction Proneness Scale (SAPS) [30] was used in 21.1% (n = 8) of the included studies. Of these, one study [31] applied both. The Mobile Phone Problem Use Scale for Adolescents (MPPUSA) [32] was used in three studies (7.9%), a shorter version of the Mobile Phone Problem Use Scale (MPPUS) was used in one study (2.6%) [16], and two studies (5.6%) applied a modified version of the Internet Addiction Scale by Young [33]. The remaining 14 studies each used a different measure (see Table 1).

Table 1 Study characteristics and results regarding risk factors of problematic smartphone use
Full size table

Regarding gender, most studies (n = 35, 92.1%) had an almost equal distribution of male and female participants (50% ± 15%). One of the included studies examined only boys [49], and two other studies had a ratio of about 94% boys to 6% girls [50] and 93% boys to 7% girls [41], respectively.

Risk factors

Gender

Several studies identified female gender as a risk factor [5, 21, 31, 36, 52, 56], reporting significant positive associations between female gender and problematic usage in adolescents (13–20 years) [10, 18, 37, 47]. Contrary to this, some studies reported smartphone addiction only in boys [28] or found higher scores in boys than in girls [19, 45]. Finally, no influence of gender was detected in other studies [17, 48, 50]. Additionally, one study [27] showed that boys and girls use their phones for different reasons: Girls spend more time on social media or text messaging, while boys are more interested in video gaming, media sharing, and Internet searches.

Age

Most studies found age to predict problematic usage [48, 52] or to be associated with it [18, 27, 38, 39, 51]. Yet, some found older adolescents [38, 48, 52] or older girls [39] to be at a higher risk, whereas others found a higher prevalence in younger (11–14 years) than in older pupils (15–18 years) [51] or in high school students than in university students [27, 42]. However, others found no predictive value of age [5, 31, 34].

Duration of use

A higher frequency of smartphone use [13, 50], a higher duration of daily usage [5, 9] (on average 33.17 min longer than healthy users, [9]), and a higher habitual use [21] have all been found to be related to problematic usage. Similarly, more time spent online and a higher amount of mobile data traffic [16] were found to be associated with addiction.

Use patterns

Using the smartphone for social networking services (SNS) [43, 46] and the duration of this usage [9] both seem to predict smartphone addiction. Adolescents with problematic use patterns also spent more time on SNS [5] or in online chats and used the smartphone more often for communication [21]. Another study [13], however, failed to find an association between smartphone addiction and SNS or instant messenger use.

In addition, gaming [21, 43, 46, 50] and a stronger denial of game overuse [9] were also found to predict smartphone addiction. Time spent gaming on the smartphone was shown to be positively related to addiction [13]. Furthermore, adolescents with problematic usage engaged in gaming more habitually and more often to achieve targets [9].

Another predictor of addiction seems to be entertainment [43]. Seeking pleasure [21] and entertainment via smartphones by watching videos [13, 46], listening to music [13], or reading e‑books [13] have all been found to be associated with problematic use. Furthermore, adolescents with problematic smartphone use have been shown to use the phone more for the purpose of stress relief or preoccupation, in cases of conflict [21], to gain peer acceptance [47], and to avoid being left out [21]. Finally, one study reported frequency of information seeking to be a risk factor for smartphone addiction [13].

School

On the one hand, a study identified going to a private school as a predictor for excessive mobile phone use [52]. On the other hand, higher school success [39, 47] and satisfaction with school life [47], as well as reading books [39] and an increase in academic motivation [35] seem to be negatively correlated with addiction rates.

Family factors

Sociodemographic variables including parental educational background and monthly income were found to have no effect on smartphone addiction in one study [34], yet another was able to show a significant positive association between family income and intensive phone use (as defined by the frequency of usage and monthly phone bills) [18]. Furthermore, parental punishment [21], as well as restrictive mediation by parents (e.g., restricting access to apps) [46, 48] all seem to increase the likelihood of problematic use and addiction in children and adolescents, whereas attachment to parents [47] and a democratic parenting style [35] seem to serve as protective factors. A significant effect has also been found for domestic violence, parental addiction (substance abuse or gambling problems) [44], parental neglect [44, 45, 54], psychological abuse [54], and emotionally traumatic experiences, the latter being partially mediated by body image dissatisfaction, social anxiety, and depression [37]. The association between parental neglect and smartphone addiction seems to be partially mediated by dysfunctional relationships with teachers [45], emotional intelligence, and coping styles [54].

Social network

A positive association has been found between problematic smartphone use and larger circles of friends [39]. In contrast, social network variables [26], friendship quality [44], friendship satisfaction [35], better relationships between students [55], and social engagement [26] may constitute protective factors. Finally, including loneliness as a risk factor for problematic smartphone usage produced inconclusive results: While one study [40] found a positive correlation, another did not detect a significant relationship [43].

Personality

The likelihood of being addicted to smartphones seems to be higher in adolescents with lower self-control [43, 44]. Furthermore, low self-esteem [41, 49, 55, 57] as well as depression [48], somatization, interpersonal sensitivity, and hostility [38] seem to be correlated with problematic phone use.

A study examining personality traits [27] found neuroticism, conscientiousness, and openness to be negatively correlated with smartphone addiction. Another study found a significant negative correlation between the Smartphone Addiction Score and emotional stability, but found no significant associations with extroversion, conscientiousness, agreeableness, or openness to experiences [28]. In addition, emotion-focused coping [54] and external-dysfunctional, internal-dysfunctional, and internal-functional emotion regulation strategies [58] have been found to explain smartphone addiction to some extent. Sensation seeking, however, does not seem to predict smartphone addiction [43].

Discussion

Although research regarding problematic smartphone use in children and adolescents covers many potential risk factors, the results are somewhat conflicting. Several aspects might have contributed to these contradictory results.

First, constructs examined by different questionnaires were not the same. The most frequently applied measure [10, 27, 28, 31, 37, 42, 49, 55, 58] was the Smartphone Addiction Scale (SAS–SV [10]). With eight studies referencing it [5, 9, 21, 31, 44, 46,47,48], the Smartphone Addiction Proneness Scale (SAPS [30, 70]) was the second most frequently applied assessment. Both measures assess the construct of smartphone addiction. Other measures pertained to the operationalization “mobile phone use.” Additionally, unrelated measures were adapted by authors for the purposes of their studies (e.g., see Young [33, Kwak et al. 43, and Jeong et al. 45]). The use of such a wide range of questionnaires is merely a symptom of a field of research that has yet to define its key research subject. Hence, operationalizations between questionnaires differ, and in many cases the term “cell phone” is simply substituted with the term “smartphones” [30]. In sum, most studies refer to smartphone addiction, whereas the term “problematic use” seems to be prevalent particularly in studies referencing mobile phones or cell phones. These heterogeneous operationalizations may again be understood as a reflection of the lack of a clear and concise conceptualization of the phenomenon. Hence, the need to reach a mutually accepted comprehensive definition of problematic smartphone use is a conditio sine qua non for further progress in the field.

Second, while the time frame (2008–2019) for the search was deliberately chosen so that the probability of including papers that focus on smartphones was increased, it is possible that participants were indeed using mobile phones without Internet access. It is not always clear which type of phone is referenced in studies and which specific features these phones had. This may substantially bias the conclusions drawn here. For example, time spent online [16] and on SNS [5, 9, 21, 46] has been identified as a possible predictor for smartphone addiction, yet in phones without Internet access, problematic usage is automatically precluded. Upcoming studies should therefore pay attention to precisely describing the type of phones studied as well as their available and actually used functionalities.

Third, all the studies included in this review focused on adolescents or mixed samples, whereas studies on young children (aged 1–10 years) are considerably scarcer. This is due to a higher prevalence of use and ownership of smartphones in adolescents. Yet in the past years, smartphone usage rates have also considerably increased among preschool children aged 6–10 years (see [71]). Hence, future research should include younger samples, as well as make an effort to not only focus on the role of parents in mediating media use, as done by those studies including younger children (e.g., see Hwang et al. [72]), but to assess child experiences directly.

It may be noticed that most studies (42.1%) that met criteria for inclusion came from South Korea. In comparison to other nations, ownership of smartphones has been found to be highest in South Korea [73], and about 96% of adolescents from South Korea use a smartphone [9]. Furthermore, studies suggest that cultural factors such as individualism vs. collectivism may have a significant influence on technology usage and technology acceptance in general [74] as well as on Internet addiction in particular [75]. For instance, a study evaluating the underlying factor structure of the Internet Addiction Test (IAT) in three collectivistic and individualistic cultures (United States, China, Africa) found the psychometric constructs to differ significantly across cultural, economic, and technological contexts [75]. Translated to the context of problematic smartphone usage, it is likely that the instruments used to assess addiction levels may have also overestimated or underestimated addiction rates as they may not have been designed to appropriately capture culturally shaped behaviors associated with smartphone use (e.g., whether smartphones are used more for mood modification in one culture than another; see Chen and Nath [75]). A more careful consideration of cultural factors in future research is needed to add to a better understanding of the generalizability and validity of the construct of smartphone addiction across cultural contexts.

Among possible predictors of problematic smartphone use, most factors produced contradictory findings. While a longer duration of use seems to be quite clearly associated with higher addiction scores, research on risk factors such as gender, social networks, and patterns of use remains inconclusive. Regarding age, findings suggest positive as well as negative correlations with smartphone addiction. Some findings exist about the influence of the school environment, as well as of family and personality factors, but corresponding data are still scarce.

Regarding gender, more studies seem to support the conclusion that female adolescents are more prone to a higher smartphone addiction risk than male adolescents. This is in line with a study in adults by van Deursen et al. [4], who also found a higher risk for smartphone addiction in women than in men. The authors relate this to the finding that women experience more social stress than men and that this results in gender-specific use patterns. Correspondingly, it has been shown that girls use their smartphones to a higher extent for social reasons than boys do [5, 27, 43, 47]. Boys, in turn, seem to focus more on gaming and media data sharing [27].

These differing usage patterns again fuel the debate about the conceptual validity of the construct at hand. One is inclined to argue that smartphone addiction, as it is currently defined (e.g., see Cha and Seo [9] and Kim et al. [30]), might be too broadly conceptualized. Similar to the dispute on whether the phenomenon of “Internet addiction” exists (e.g., see Widyanto and Griffiths [76]), it is conceivable that smartphone users may not be addicted to the device itself but to the applications provided by it (e.g., SNS, online games, online pornography). Thus, in correspondence to the suggestion made by Griffiths [77], a distinction needs to be made between addictions to the smartphone and addictions on the smartphone. It may, thus, be more promising to focus on specific types of use (e.g., problematic gaming as proposed by the ICD-11 [12]) in future research and assess relevant indicators of behavioral addictions (e.g., salience, mood modification, tolerance, withdrawal, conflict, and relapse [76]) in relation to specific applications rather than merely considering the frequency or duration of using the technical device in general.

Limitations and conclusion

One of the limitations of this review is that the causality of the relation described for the variables in question is not statistically firm across all studies. Most studies included here used correlational research and were cross-sectional. Yet in order to debunk the question of directionality (i.e., whether a postulated risk factor is a contributor or a consequence), longitudinal research is warranted.

Despite interpretational difficulties due to different operationalizations, this review was able to provide an overview of risk factors related to problematic smartphone use or smartphone addiction in children and adolescents. Based on this, the following implications for future research may be postulated: Most importantly, a concise definition of the construct with a standardized terminology and operationalization would enhance the comparability of findings. Developing a comprehensive theoretical framework for this construct is, however, closely related to the question of whether it constitutes a singular entity that is sufficiently distinct from other concepts such as problematic gaming, or whether it is merely a symptom of the latter or of the addiction to specific applications on the smartphone.