Risk factors and psychosocial characteristics of potential problematic and problematic internet use among adolescents: A cross-sectional study

Particularly among adolescents, the internet is observed to be increasingly adopted as a readily accessible means for information retrieval, entertainment, and socialization [1, 2]. Excessive internet use may instigate potential adverse effects upon the psychosocial development of adolescents [3]. While both the adoption of excessive internet use, as well as related adverse psychosocial effects may be attributed to the compromised psychosocial wellbeing prior to the initiation of internet use [4], the likelihood of developing problematic patterns of behavior during adolescence is eminent [5, 6]. Consequently, as adolescents allocate ever-increasing time periods for internet use, the risk for developing maladaptive internet use (MIU), including potential problematic internet use (PIU) and PIU, is inherent.

While PIU has received increasing research attention [7], a consistent definition of this construct has not been currently applied [8]. PIU has been proposed as a novel entity of dysfunctional behavioral patterns similar to those identified within the spectrum of impulse control disorders [9]. The proposed criteria for PIU initially included: (1) an uncontrollable use of the internet, (2) internet use which is markedly distressing, time consuming or resulting in social, occupational or financial difficulties; and (3) internet use not solely present during hypomanic or manic clinical episodes [10]. Hence, PIU is conceptualized as an individual's inability to control his/her use of the internet, thus causing marked distress and/or functional impairment [11, 12]. Potential PIU is defined as internet use which fulfils some, but not all, of the above criteria [9, 12, 13].

Worldwide, the prevalence of PIU among adolescents and young adults has been observed to range between 0.9% [14] and 38% [15]. In particular, among European adolescents, the prevalence of PIU has been observed to range between 2% and 5.4% [6, 16‐18]. In Greece, the prevalence of PIU is observed to range between 1.0% [19] and 8.2% [20] among adolescents residing in rural and urban areas, respectively. Hence, PIU is markedly elevated among Greek adolescents as compared to their counterparts in other European countries.

Both excessive and PIU have been associated with a plethora of adverse psychosocial and mental health conditions. Specifically, the adoption of excessive internet use has been associated with social isolation [21] and related adversities [22]. Moreover, PIU has been associated with hostile behavioral patterns [23], impaired social skills [24], attention deficit hyperactivity disorder [14], and depression and/or suicidal ideation [25‐27]. However, to date, evidence does not exist regarding the differential determinants and psychosocial implications of potential PIU and PIU among adolescents.

The primary objective of the present investigation is to evaluate the determinants of PIU and potential PIU among adolescents. The secondary objective is to evaluate the psychosocial characteristics and implications associated with PIU among the study population.

The present study is the first of its kind to assess the internet use characteristics associated with both potential PIU and PIU among adolescents. Moreover, it is also the first of its kind to evaluate both the solitary and differential psychosocial implications associated with PIU among adolescents according to the degree of maladaptive behavioral patterns adopted.

The study findings indicated that the prevalence rate of PIU among adolescents was 1.5%. The observed prevalence rate is within the lower range of those reported both in Greek rural areas and in other European countries [6, 16, 18, 20, 34] and may be attributed to the limited penetration of computer/internet access among urban Greek youth [35]. However, marked international variances regarding the prevalence rates of PIU may also be attributed to a measurement bias incurred by a lack of international consistency regarding both the definition and assessment of PIU [8]. Furthermore, among the study population examined approximately one fifth (19.4%) of adolescents were identified with potential PIU. It is upheld that such internet users are at an enhanced risk for developing PIU.

The majority of adolescents with either potential PIU or PIU were male. Similar gender differences regarding the frequency and nature of internet use have been previously reported [36]. The gender differences observed may be attributed to the potential confounding effect of the differential frequency of internet utilization between genders. Specifically, since adolescent boys utilize the internet both more frequently and extensively than adolescent girls [19], the average weekly hours of internet utilization may serve as a potential confounder for the development of PIU, particularly among adolescent males.

The study findings indicated that potential PIU and PIU were independently associated with utilizing the internet for the purposes of retrieving sexual information, socialization, and entertainment, including interactive game-playing. Moreover, it is noteworthy that potential PIU was inversely associated with utilizing the internet for educational purposes. Previous reports indicate that more than one quarter of frequent internet users utilize the internet for accessing sexual information and education [19, 37, 38]. Both frequent internet use and accessing the internet for the purposes of sexual education have been found to be significant predictors of pornographic internet site use [39, 40] and consequent PIU [41]. Hence, it is proposed that PIU may develop and/or manifest secondary to the specific content of internet sites accessed, rather than to the internet per se.

With regard to the psychosocial implications of PIU, including potential PIU and PIU, the study findings indicated that such behaviors are associated with an enhanced likelihood of hyperactivity and conduct problems. It is important to note, though, that while the likelihood of conduct problems was more than three times greater among adolescents with PIU as compared to those with potential PIU, the odds of hyperactivity problems was approximately two times greater, respectively. To date, similar findings regarding the likelihood of hyperactivity and conduct problems among adolescent with potential PIU have not been reported.

The evidence provided regarding the concomitant occurrence of conduct problems and PIU corroborates with related findings in the literature indicating that adolescents with PIU tend to be lonelier [42] and adopt more aggressive behaviors [43]. Moreover, previous findings have indicated that conduct problems among youth with PIU may be proximally associated with increased social isolation and impaired communication skills [24]. However, the present study findings indicated that adolescents with either potential PIU or PIU did not present with deterred peer relations and/or social skills. It is posited that adolescents may counteract their possible real world social isolation with increased use of cyber communication and socialization platforms, thus retaining social networks through the internet medium.

The present study indicated that neither potential PIU nor PIU among adolescents were significantly associated with emotional maladjustment. These findings contrast those established in the literature indicating that emotional symptoms, such as depressive and anxiety symptoms, have been associated with PIU [9, 44‐47]. It is posited that the emotional adjustment among adolescents with either potential PIU or PIU may be secondary to a potential population bias introduced by the study sampling applied. Specifically, due to the fact that the study population was recruited from students attending public junior high and high schools, those adolescents with severely impaired functionality, including both severely deterred academic performance to the extent of exclusion and/or expulsion from academic attendance and activities, may have not been included in the study population.

Furthermore, the present study indicated that adolescents with potential PIU or PIU were more than two and eight times, respectively, as likely to have global emotional and psychosocial maladjustment, as assessed by the total SDQ score. A correlation between PIU and compromised psychological well-being has been previously documented [42, 48]. However, differential psychosocial impacts according to the degree of PIU have not been reported. Thus, the present study provides evidence that while adolescents with PIU exhibit marked behavioral and psychosocial maladjustment, adolescents with potential PIU also have a limited, albeit notable, increased risk for manifesting comprehensive emotional and psychosocial impairments.

Therefore, the study findings indicate that both potential PIU and PIU are associated with notable emotional and psychosocial maladjustment among adolescents. It is upheld that such internet behaviors may constitute an escape mechanism for adolescents to temporarily relieve and/or escape from emotional and behavioral difficulties [49]. Therefore, adolescents may use the internet excessively in order to cope with emotional turmoil. Concurrently, PIU has been observed to lead to unsuccessful life-coping mechanisms [50]. It is posited that poorly adjusted adolescents, may suffer more harmful effects following PIU, thus creating a vicious gyration centered upon internet usage and psychosocial maladjustment. Consequently, PIU may compound pre-existing psychosocial symptomatology present among adolescents.

The strengths of the present study include that it is the first of its kind conducted in order to assess both the determinants and psychosocial effects of potential PIU and PIU among adolescents in Greece. Due to the random sampling applied for the selection of the study population, it is upheld that the potential introduction of a selection bias was deterred. The limitations of the study include its inability to decipher the etiological association between PIU and the psychosocial characteristics of adolescents due to the cross-sectional study design applied. In addition, psychiatric conditions and other risk factors could not be assessed in relation to the occurrence and development of maladaptive internet use. Finally, since adolescents belonging to the same class and/or school may potentially utilize internet applications with each other, a clustering effect upon the association between the use of cyber social networking, as well as gaming, in relation to maladaptive internet use may have been introduced. Since a stratified cluster sample was used for the present investigation, both the standard errors and confidence intervals reported may be an underestimation of their true magnitude. Further prospective investigations are necessary in order to assess both such clustering effects and whether the psychosocial characteristics observed among adolescents with potential PIU may constitute potential risk factors for the consequent development of PIU.

Among adolescents the prevalence rates of potential PIU and PIU were observed to be 19.4% and 1.5%, respectively. Multinomial logistic regression indicated that potential PIU and PIU were significantly associated with male gender, as well as utilizing the internet for retrieving sexual information, interactive game playing, and socialization, including chat-room use and email. Adolescents with potential PIU had an increased likelihood of concomitantly presenting with hyperactivity and conduct problems. Moreover, adolescent PIU was significantly associated with hyperactivity and conduct problems, as well as comprehensive psychosocial maladjustment. Therefore, the determinants of potential PIU and PIU include accessing the internet for the purposes of retrieving sexual information, game playing, and socialization. Furthermore, both potential PIU and PIU adversely associated with notable behavioral and social maladjustment among adolescents.