Prevalence and Pattern of Internet Addiction Among Adolescents in Ibadan, Nigeria: A Cross-Sectional Study

Background and objective Despite the potential benefits the Internet offers, it is prone to excessive and uncontrolled use, thus resulting in a condition called Internet addiction (IA). This study aimed to describe the prevalence of Internet addiction (IA) among in-school adolescents in Ibadan, Southwest Nigeria. Materials and methods A descriptive cross-sectional study was conducted among 632 adolescents using a two-stage sampling technique. IA was assessed using the 20-question Internet addiction test (IAT). Responses to each question ranged from “0” (i.e., “never”) to “5” (i.e., “always”). Cumulative IAT scores > 50% suggested the presence of IA. Chi-square tests were conducted to determine the association between adolescents’ characteristics and IA. Statistically significant variables were pooled into the binary logistic regression model. P-values < 0.05 were statistically significant. Results The mean age of the adolescents was 16.03 ± 1.26 years, and 347 (54.9%) were males. A total of 284 (44.9%) adolescents had IA: 174 (42%) accessed the Internet in both home and school settings (ᵡ2 = 4.103, p = 0.043), and 174 (42%) accessed the Internet at home only (ᵡ2 = 5.003, p = 0.025). Adolescents who accessed the Internet from both home and school settings had higher odds of developing IA (adjusted odds ratio (AOR) = 1.408, 95% confidence interval (CI) = 0.986-2.012, p = 0.060), as well as those who accessed the Internet from home settings only (AOR = 1.404, 95%CI = 1.010-1.953, p = 0.043). Adolescents who gained four to six hours of Internet connection weekly had two times odds of developing IA (AOR = 1.404, 95%CI = 1.010-1.953, p = 0.043), and those who gained more than six hours of Internet connection had more than three times odds of developing IA (AOR = 3.424, 95%CI = 1.937-6.053, p = 0.043). Conclusion To prevent IA, adolescents should develop self-control skills and self-regulation of Internet use. Likewise, Internet access should be restricted from both home and school settings, and adolescents’ Internet access should be monitored and regulated from both home and school settings.


Introduction
Despite the potential benefits the Internet offers, it is prone to excessive and uncontrolled use, thus resulting in a condition called Internet addiction (IA) [1]. IA disorder is defined as too much computer use that contradicts daily activities in ways that harm daily function [2]. Many pieces of literature have reported a strong association between the onset of IA and the duration spent on the Internet. Nakayama et al. defined IA as weekly Internet use that exceeds 12 hours [3]. A study conducted in Turkey reported that the daily use of the Internet for over two hours is related to IA and mental disorders [4].
IA can occur among all age groups; however, adolescents are particularly vulnerable to IA because of the mental, emotional, and social developments associated with this period [5]. IA has generally been strongly associated with reduced sleeping time, a tendency for postponing sleep, insomnia, excessive tiredness, anxiety, depression, suicide, and attention deficit hyperactivity disorder [6][7][8]. In lieu of this, the Australian government has developed guidelines for a maximum screen time of two hours daily for schoolchildren and adolescents [9]. Also, the Chinese government has begun to ban the opening of new Internet cafes as a government campaign to clamp down on IA. They have also banned the entry of children below 16 years to Internet cafes to improve students' concentration on academic work [10].
Many studies have been conducted to identify the home as a risk factor for the development of IA among inschool adolescents [5][6][7][8]. However, there is limited literature that assesses Internet exposure in schools as a risk factor for IA among adolescents. In addition, the underlying factors responsible for the outplay of IA among these adolescents are relatively unknown. Neglect of IA would result in an increased vulnerability of adolescents to IA and related disorders, thereby resulting in comorbidities for which more healthcare costs would be incurred on their treatment on the family and nation. A study in this regard would be helpful in planning appropriate interventions among present-day adolescents to ensure that their physical and mental health is not adversely affected. This study, therefore, aimed to describe the pattern (access type, access location, frequency, and motive) of Internet use, as well as the prevalence and factors associated with IA among in-school adolescents in Ibadan, Southwest Nigeria.

Study design and area
This was a descriptive cross-sectional study conducted in Ibadan. Ibadan is the capital city of Oyo state and the third-largest city in Nigeria using population projections from the 2006 Nigerian population census [11].
Ibadan is located 7°23′47″N and 3°55′0″E, and it has a total area of 3,080 km 2 [12]. It is 128 km northeast of Lagos and 345 km southwest of Abuja, the Federal Capital Territory [12]. Internet service providers such as Global Communications Limited and Mobile Telecommunications Network have improved Internet access through the erection of masts in public areas [13].

Study population and sample size
The study was conducted among currently enrolled high school students in selected public secondary schools in Ibadan Southeast Local Government Area (LGA) between April 29 and May 3, 2021, shortly after the suspension of school closure in Nigeria due to the COVID-19 pandemic. Public schools are composed of a homogeneous population of students.
The sample size of in-school adolescents was determined using the sample size formula for cross-sectional studies [14]. A prevalence of 32% [15] was used at a 99% confidence interval (CI) and an allowable error of 4.5%. Thus, the minimum sample size obtained after sample size calculation was 675 in-school adolescents.

Sampling technique and data collection
The selection criterion for the study was the possession of computer laboratories in schools. A two-stage sampling technique was used to enroll participants.

Stage One
From the list of schools obtained from the Zonal Chairman of the Teaching Service Commission, three (one boys-only, one girls-only, and one mixed) schools were selected out of the eight schools with computer laboratories using simple random sampling (balloting method).

Stage Two
Of the 12 arms present in each school, four classes were selected from each through a simple random sampling technique (balloting method).
A semi-structured interviewer-administered questionnaire was used for data collection. The questionnaire was divided into four sections: section A: sociodemographic characteristics, which included age, sex, school, family type, parent's educational qualification, parent's age, and parent's occupation; section B: pattern of Internet use, which included questions on the current ownership of an Internet-enabled device, type of Internet-enabled device(s) owned, the motive for Internet access, the pattern of Internet social network type ever used, site(s) of Internet connection, the pattern of restriction to Internet access, and the duration of Internet connection (during school hours and during weekends); section C: causes, effects, and preventive measures for Internet addiction, in which one structured question was used to elicit information on the causes of IA (including boredom, loneliness, poor physical relationships, and idleness), one structured question was used to elicit information on the effects of IA, and one structured questionnaire was used to elicit information on the preventive measures for IA; and section D: Internet addiction test (IAT), an addiction test comprising 20 questions [16]. The IAT was used to determine the prevalence of IA among adolescents. IAT scores were computed with "0" assigned for "never happened," "1" for "once weekly," "2" for "twice weekly," "3" for "three times weekly," "4" for "four times weekly," and "5" for "five times weekly or more."

Statistical analysis
The collected data were input into the computer daily. Data were analyzed using SPSS version 25.0 (IBM Corporation, Armonk, NY, USA) [17]. Descriptive statistics were summarized using frequency tables, percentages, mean, standard deviation, and charts. The number of hours with daily Internet access and age of onset of Internet use were computed using mean and standard deviation. The average time spent online per day during school hours and after school hours on weekdays and during weekends was computed using frequencies and percentages. The overall average time spent on the Internet (either at home or in school) was categorized into three: "three hours or less," "four to six hours," and "more than six hours." For the IAT, the cumulative score in the 20 domains yielded a maximum score of 100 points. IAT scores ranging between 20 and 49 points suggested complete control over Internet use. A range of scores between 50 and 70 points was suggestive of occasional IA, while IAT scores greater than 70 points implied significant IA. Occasional and significant IA were categorized together as the presence of IA, while healthy (complete control over) Internet use was categorized as the absence of IA.
Chi-square tests were used to determine the association between sociodemographic characteristics and IA. Variables that were statistically significant at a 5% level of significance in the bivariate analysis were pooled into the binary logistic regression, which was conducted (on variables such as age, school, the current ownership of an Internet-enabled device, Internet connection (at home only, in school only, or both at home and in school settings), self-perception of problematic Internet use, and average time spent on the Internet weekly) to determine the predictors of IA.

Ethics
Ethical approval for this study was obtained from the Oyo State Research Ethics Committee (AD 13/479/4130A). Permission to conduct the study was obtained from the Deputy Administrator of each school. Assent was obtained from adolescents below 18 years, while consent was obtained from those aged 18 years or older. The adolescents were informed of the right to opt-out of the study before the administration of the questionnaire. Participation in this study did not expose adolescents to any known form of harm or injury; rather, the study was helpful in the identification of IA among adolescents as a step toward influencing healthy Internet use habits among them.

Sociodemographic characteristics
A total of 632 questionnaires were filled out of the 675 questionnaires distributed. Thus, a response rate of 93.6% was obtained.
The mean age of the 632 in-school adolescents was 16.03 ± 1.26 years. Among them, 433 (68.5%) were older than 15 years; 324 (51.3%) were enrolled at a mixed school, and 162 (25.6%) were enrolled in a girls-only school ( Table 1).

Pattern of Restriction of Internet Access From Home and School Settings
In total, 208 (

Motives for Internet access
Regarding the pattern of Internet social network type used by the adolescents, 494 (78.2%) had used Facebook, 241 (38.2%) had accessed WhatsApp, and 153 (24.2%) had made use of Snapchat (Figure 1).

Pattern of Internet access in the past one week among adolescents
Overall, 334 (52.8%) adolescents had accessed the Internet in the past one week in school settings only through the computer laboratory. Among them, 184 (55.1%) had spent one hour or less, and 87 (26%) had spent between two and three hours online. A total of 414 (65.5%) adolescents had accessed the Internet both at home and in school in the past one week. Among them, 223 (53.9%) had spent one hour or less on the Internet at home after school hours (Monday to Friday), and 166 (40.1%) had spent one hour or less on the Internet at home during the weekend ( Table 2). 2022 Table 3). 2022

Association between the restrictions implemented by parents/guardians and teachers and adolescents' Internet addiction
A total of 165 (44.6%) in-school adolescents whose Internet use was restricted at home by their parents had IA compared with 119 (45.4%) whose Internet use was not restricted by parents (ᵡ 2 = 1.423, p = 0.233

TABLE 4: Association between the restrictions implemented by parents/guardians and teachers and Internet addiction among in-school adolescents in Ibadan Southeast
LGA, 2021

Determinants of Internet addiction among adolescents
Adolescents in the mixed school had two times the odds of developing IA compared with those in the boysonly school (adjusted odds ratio (AOR) = 2.108, 95% confidence interval (CI) = 1.339-3.319, p ≤ 0.001). Inschool adolescents in the girls-only school had nearly three times the odds of developing IA compared with those in the boys-only school (AOR = 2.675, 95%CI = 1.580-4.529, p = 0.001). In-school adolescents who did not access the Internet from both home and school settings had nearly 31% fewer odds of developing IA compared with their counterparts who accessed the Internet via multiple settings (AOR = 0.691, 95%CI = 0.481-0.991, p = 0.044) ( Table 5).

TABLE 5: Determinants of internet addiction among in-school adolescents in Ibadan Southeast
LGA, 2021

Discussion
From this study, we identified that nearly 50% of in-school adolescents had IA. The high proportion obtained in this study could be explained as an aftermath of the COVID-19 lockdown restrictions that mandated physical isolation and compulsory homestay for at least five months in Nigeria and other parts of the globe [18,19]. The COVID-19 lockdown lasted from March 2020 to September 2020, and during this period, the only avenue through which many individuals could keep connected to the world was the Internet. Although this study was conducted a few months after all schools had reopened and were bound to observe COVID-19 safety measures including social distancing, many students and older individuals were yet to recover from the impacts of the COVID-19 lockdown. Therefore, such uncontrolled Internet connection during the lockdown could have transcended into IA even among in-school adolescents, long after the restrictions were lifted. The proportion of problematic Internet users found in this study is higher than the results of research conducted in Iran (40.7%), Turkey (9.7%), and Dhaka (24%) [20,21]. Our finding in the present study is similar to the 56.5% prevalence of IA reported from a study conducted among undergraduate students from a Malaysian public university [22].
This study found that Internet access from both school and home settings was a significant predictor of IA among adolescents. Given that adolescents spend more than half of each day in their homes, self-control needs to be practiced by these adolescents. In addition, parental control of adolescents' phones and Internet access is noteworthy for preventing IA among adolescents. However, parents need to be cautious while weighing the different methods to enhance compliance with Internet use among their adolescents. The restrictive parenting approach has been established to have a positive association with IA among adolescents [23,24]. In many instances, the more the Internet control rules, the more likely it is that adolescents will become problematic Internet users. Therefore, the use of the restrictive parenting approach may be counterproductive in the context of IA among adolescents and may prompt adolescents to act in deviant ways while coping with stressful and unpleasant situations.
This study revealed that nearly 78% of in-school adolescents in Ibadan use Facebook more frequently than other social network types. The possible reasons for this observation are not far-fetched. Firstly, Facebook provides user-friendly services at lower costs compared to other social network types [25]. From their research, Ngonso et al. reported that a high proportion of Nigerian youths frequently accessed Facebook than other social network sites [26]. Facebook is the oldest social network site, and this could therefore provide some explanation on its high proportion of coverage. A cross-sectional study conducted among teenagers who are currently enrolled in high school has reported that although Facebook remains the most popular site accessed, teenagers commit to more than one social network platform [27]. Teenagers maintain a "social media portfolio" where top sites such as Facebook, Twitter, and Instagram are visited [27,28]. To prevent a situation where in-school adolescents invest too much time on social network sites to the detriment of their physical and mental health, the controlled use of these sites should be encouraged among adolescents.
The three Ms of parental involvement elucidated by the New York Center for Living would be important to yield optimal results in ensuring healthy Internet use among adolescents [29]. These would include parental modeling, through demonstration of the appropriate use of Internet-enabled devices; parental monitoring, such as the use of lights out and technology-off hours; and parental mentoring. Parental monitoring would entail that parents are engaged in their children's world by building their interest in non-digital affairs and creating a robust offline schedule for exercise, parent/child activities, and family meditation.
This study revealed that Internet access in school could indeed be a potential gateway to the development of IA among in-school adolescents. Because adolescents spend nearly eight hours in school, teachers are required to monitor adolescents' access to the computer laboratory, as well as activities engaged in during computer classes. Restricting Internet access from home and school settings should be undertaken to reduce the likelihood of the development of IA among adolescents. Likewise, adolescents should be discouraged from bringing phones to school, as this distracts them from the primary responsibility of learning.

Limitations
Many schools lack computer laboratories in Ibadan; thus, the sampling frame was restricted to a list of only a few schools. Because it was conducted in only one city, this study could lack some generalizability. The heterogeneous population of students in the schools enlisted in this study provides some generalizability to the general population of in-school adolescents. Despite these potential limitations, this study provides concrete evidence on the pattern of IA among in-school adolescents.

Conclusions
The prevention of IA among in-school adolescents is multifaceted and requires a multidisciplinary approach. For effectiveness and sustainability on a long-term basis, many factors come to the fore: individual (adolescent), parental, school, and media. Therefore, we recommend the adoption of the following measures, especially during lockdown periods. Firstly, self-control skills such as self-monitoring and self-regulation of screen time (time spent on all Internet-enabled devices) should be adopted by adolescents themselves. Secondly, for the effectiveness of monitoring of Internet use, parents are required to act as role models; i.e., when they regulate their own pattern (frequency of access and time spent online) of Internet use, controlled Internet use can be enhanced among adolescents. Parents are also required to actively engage with adolescents during Internet activities such as knowing the different social media platforms that exist; decisions on adaptive Internet use could be jointly made by adolescents and their parent(s). Thirdly, school-based health awareness programs should be conducted on a regular basis on the dangers of IA and the need to regulate adolescents' Internet use. Table 6 shows the questionnaire used for data collection.