1. Background
There has been a dramatic increase in the use of electronic devices, such as computers and smartphones, which, while they provide obvious and extraordinary benefits for their users, can also have negative health effects if used excessively (1). This lack of control over the use of the Internet is considered to be maladaptive behavior (2). There are several types of addictions, and internet addiction is one of them. It is defined as excessive or poorly controlled preoccupations, urges, or behaviors related to computer use and internet access that result in impairment or distress (3). The problem of internet addiction has become a global health concern (1, 4). Internet addiction is reported to be prevalent in Iran at a rate of 20% (5) and at a rate of 23% (6) in another study. There is a prevalence of internet addiction among male and female students in Iran at 33.8% and 20%, respectively (7). Sixty-eight point eighty-three percent and 26.95% of students and graduates in the fields of epidemiology, clinical sciences, and basic sciences in Iran were moderate and mild internet users, respectively, indicating a high risk of severe addiction (8). Meanwhile, Iranian internet users have increased from zero percent in 1990 to 70% in 2018 (9). Furthermore, the rate of internet growth in Iran is estimated to be 20 percent between 2006 and 2015 (5). Moreover, the increasing popularity of mobile phones and technological advancements have resulted in using electronic devices (10). In particular, young people use mobile phones before sleeping at night, leading to sleep disorders and late sleep (11). Internet use as one of the most important factors influencing sleep quality has become a major concern. One of the consequences of internet addiction is its effect on sleep quality. Sleep quality is defined by Kline as one’s satisfaction with the sleep experience, incorporating aspects such as sleep initiation, sleep maintenance, sleep quantity, and refreshment upon waking (12). Studies have been conducted in Iran and around the world about the relationship between internet addiction and sleep quality which have reported a relationship between the Internet and sleep quality (13, 14). A study by Nayak et al. in 2021 among medical students in India found that people with internet addiction had lower sleep quality (15). The study conducted by Wang et al. on 3,738 students in China shows that internet addicts have poorer sleep conditions than normal internet users (4). Another study by Arzani-Birgani et al. in Iran also substantiates this claim (16). Sleep quality can significantly impact a person’s physical and mental performance (17). Also, poor sleep quality can cause a decrease in students’ learning and academic performance (18). By investigating the relationship between internet addiction and sleep quality, we may be able to improve students’ academic performance.
2. Objectives
In this study, we will try to establish a reliable source to minimize inappropriate use of the Internet to facilitate the improvement of sleep quality among students in order to examine: (1) The relationship between sleep quality and internet addiction in students, and (2) the changes in sleep quality among students with different levels of internet addiction.
3. Methods
3.1. Type of Study Design and Selection of Participants
In this study, a cross-sectional design was used. Students at the Isfahan University of Medical Sciences in Iran have been selected as the intended community for this study. The questionnaires were completed between August and September of 2021. It is important to note that the participants were fully informed at the beginning of the questionnaire that they had the complete authority to fill out or not fill out the questionnaire. It was necessary to include participants who were (1) studying at the undergraduate and general doctoral levels (basic sciences) at Isfahan University of Medical Sciences and Medical Services, (2) attempting to fill out the questionnaire, and (3) completing the questionnaire completely. In accordance with self-reports, individuals with different levels of psychiatric disorders have been excluded from the study. Ethical approval was obtained from the Ethics Committee of Isfahan University of Medical Sciences (IR.MUI.RESEARCH.REC.1400.139). A total of 562 cases participated in this study. Multi-stage sampling was used in this study. In this case, colleges were selected as categories, and a college was located at each level. Then, in the next stage, the degrees of education in different educational sections (including undergraduate courses, Basic medical students) were placed in the next layer. In the following, different entries were placed in each section. After that, different entries were placed in each section. Finally, a number of last-layer students with an emphasis on maintaining gender balance were selected as the final sample. This sample size was determined by the correlation between sleep quality score and internet addiction score in the same literature as 0.406, with an alpha of 1% and a beta of 10% and the average cluster size of 20 individuals, and the coefficient of 0.3 and 10% drop was 562.
3.2. Research Instrument
In order to collect data, the structural questionnaire consists of three sections. In addition to demographic information, two Persian versions of the Pittsburgh Sleep Quality Index (PSQI) and Internet Addiction Test (IAT) standard questionnaires were provided to students of Isfahan University of Medical Sciences and Health Services via the Internet using online questionnaires (Google Forms). The first part of the questionnaire included demographic information such as age and gender as well as other information such as using a smartphone before going to bed (19). The second part includes the PSQI, which has been selected as the indicator of student sleep quality assessment. This questionnaire consists of 18 items consisting of 7 components. Scores greater than 5 indicate poor sleep quality (20) in the range of 0 to 21. Based on the seven component scores of the PSQI, the overall reliability coefficient (Cronbach’s alpha) was 0.83, and in Iran, it was 0.77, indicating a high degree of internal consistency (20, 21). The third part of the questionnaire was done using the IAT with 20 items and six subgroups with a score of 0 to 100 to measure internet addiction, which Yang et al. first evaluated (11). A score of 0 - 19 indicates a lack of dependence on the Internet, while a score of 20 - 39 indicates a mild degree of internet addiction. On the other hand, 40 - 69 points indicate moderate internet addiction (at risk), while 70 to 100 points indicate severe internet addiction (22-24). In Iran, Cronbach’s alpha coefficient of 0.917 was the acceptable level (25), indicating the validity and reliability of the internet addiction questionnaire.
3.3. Statistical Analysis
SPSS statistical software version 25 of the Windows operating system was used throughout the study. Mean (
4. Results
4.1. Demographic Information, Internet Addiction Levels, and Sleep Quality
Demographic information, as well as the internet addiction score for sleep quality, is shown in Table 1. Participants were 18 to 29 years old with a mean age of (21.41 ± 1.87). Regarding gender distribution, there were 370 (65.8%) female students and 192 (34.2%) male students. The sample distribution between undergraduate and public doctoral students was 328 (58.4%) and 234 (41.6%), respectively. Four hundred seventy-seven participants (84.9%) lived with their families, and 85 (15.1%) lived alone. Of these, 509 (90.6%) were single. In the case of using a smartphone, 517 people (92%) used smartphones against 45 people (8%) before sleep. In accordance with the internet addiction scores, 77.4 percent of the participants were classified as mild or moderately addicted to the Internet.
Table 1.Demographic Information and Internet Addiction Levels for Sleep Quality a
| Variables | Sleep Quality | |||
|---|---|---|---|---|
| Total (n = 562) | Poor (n = 294) | Good (n = 268) | Sig | |
| Age | 21.41 ± 1.87 | 21.68 ± 1.88 | 21.11 ± 1.81 | < 0.001 |
| Gender | 0.04 | |||
| Woman | 370 (65.8) | 182 (61.9) | 188 (70.1) | |
| Men | 192 (34.5) | 112 (38.1) | 80 (29.9) | |
| Grade | 0.679 | |||
| Undergraduate | 328 (58.4) | 174 (59.2) | 154 (57.5) | |
| General Doctorate | 234 (41.6) | 120 (40.8) | 114 (42.5) | |
| Live with family | ||||
| Yes | 477 (84.9) | 250 (85) | 227 (84.7) | 0.912 |
| Marital status | ||||
| Single | 509 (90.6) | 267 (90.8) | 242 (90.3) | 0.834 |
| Married | 53 (9.4) | 27 (9.2) | 26 (9.7) | |
| Smartphone use before sleep | ||||
| Yes | 517 (92) | 283 (3.96) | 234 (3.87) | < 0.001 |
| Internet addiction test level | < 0.001 | |||
| Normal | 117 (20.8) | 27 (9.2) | 90 (33.6) | |
| Mild | 262 (46.6) | 144 (49) | 118 (44) | |
| Moderate | 173 (30.8) | 120 (40.8) | 53 (19.8) | |
| Severe | 10 (1.8) | 3 (1) | 7 (2.6) | |
a Values are expressed as mean ± SD or No. (%).
The ANOVA analysis among internet addiction degrees was carried out for variables of total sleep quality and seven components, whose results are visible in Table 2. A significant (P < 0.001) increase in the average scores of sleep quality and its components was observed, and internet addiction exhibited an upward trend among users. This did not apply only to the level of severe Internet addiction and was different from other subgroups. In all three levels of normal internet users, users with mild and moderate internet addiction had lower sleep quality scores, habitual sleep efficiency, use of sleeping medication, and daytime dysfunction than students with severe internet addiction. Tukey’s post hoc test, which shows a significant difference between the groups, is shown in Table 2.
Table 2.Sleep Quality Among Participants with Varying Degrees of Internet Addiction a
| Pittsburgh Sleep Quality Index | Internet Addiction Test | P-Value | Tukey’s Post Hoc | |||
|---|---|---|---|---|---|---|
| Normal (A) | Mild (B) | Moderate (C) | Severe (D) | |||
| Total score Pittsburgh Sleep Quality Index | 4.41 ± 2.20 | 6.60 ± 3.43 | 8.41 ± 4.58 | 4.20 ± 2.57 | < 0.001 | C > B > A; C > D |
| Subjective sleep quality | 0.73 ± 0.62 | 1.12 ± 0.73 | 1.35 ± 0.86 | 0.60 ± 0.51 | < 0.001 | C > B > A; C > D |
| Sleep latency | 0.98 ± 0.66 | 1.43 ± 0.91 | 1.6 ± 1.01 | 1.00 ± 1.05 | < 0.001 | C > A; B > A |
| Sleep duration | 0.31 ± 0.60 | 0.65 ± 0.87 | 1.07 ± 1.14 | 0.70 ± 1.05 | < 0.001 | C > B > A |
| Habitual sleep efficiency | 0.56 ± 0.88 | 0.82 ± 1.05 | 1.32 ± 1.24 | 0.40 ± 0.96 | < 0.001 | C > B; C > A; C > D |
| Sleep disturbances | 1.02 ± 0.13 | 1.08 ± 0.27 | 1.22 ± 0.44 | 1.10 ± 0.31 | < 0.001 | C > B; C > A |
| Use of sleeping medication | 0.04 ± 0.20 | 0.21 ± 0.49 | 0.32 ± 0.69 | 0.00 ± 0.00 | < 0.001 | C > B > A |
| Daytime dysfunction | 0.77 ± 0.74 | 1.29 ± 0.89 | 1.49 ± 1.01 | 0.40 ± 0.69 | < 0.001 | C > A; B > A; B > D |
a Values are expressed as mean ± SD.
Table 3 illustrates the relationship between the total score of internet addiction and sleep quality based on the logistic regression. Sleep quality was significantly predicted by the total score of internet addiction, with an odds ratio of 1.035 with a confidence interval of (1.023, 1.048). It is important to note that when the score of Internet addiction is more than one, it means that the increase in the odds ratio of Internet addiction decreases the quality of sleep. (Code 1: poor sleep quality, code 0: good sleep quality).
Table 3.Relationship Between Overall Scores of Internet Addiction and Sleep Quality by Logistic Regression
| β | Standard Error | Exp (β) | 95% CI for Exp (β) | P-Value | ||
|---|---|---|---|---|---|---|
| Lower | Upper | |||||
| Internet Addiction Test total score | 0.006 | 0.034 | 1.035 | 1.023 | 1.048 | < 0.001 |
Using logistic regression, Table 4 examines the relationship between internet addiction subgroups and sleep quality. Among the salience subgroup, the odds ratio is 0.95 (CI 95% = 0.90, 0.99) and less than one, meaning that as salience scores increase, the odds of poor sleep quality decrease, and this relationship is significantly predicted (P = 0.038). These results were shown after controlling for the effect of independent variables. The odds ratio in the subgroup of Excessive use of the Internet (OR = 1.02, CI 95% = 1.05, 1.18) had a direct and significant relationship (P < 0.001) with sleep quality. Moreover, this positive and significant relationship (P < 0.001) was also seen in the subgroup of Anticipation and Sleep Quality (OR = 1.40, CI 95% = 1.23, 1.52). Other subgroups of neglect of work, lack of control, and neglect of social life did not significantly correlate with sleep quality (P-value: 0.46, 0.66, 0.46, respectively). The gender variable did not significantly correlate with sleep quality (P = 0.66). Regarding the age variable, the odds ratio was 1.15 (CI 95% = 1.08, 1.28). A significant inverse relationship was also predicted between using a smartphone before bed and sleep quality (P = 0.007, OR = 0.35, CI 95% = 0.16, 0.75).
Table 4.Association Between Internet Addiction and Sleep Quality by Logistic Regression
| Variables | β | Standard Error | Exp (β) | 95% CI for Exp (β) | P-Value | |
|---|---|---|---|---|---|---|
| Lower | Upper | |||||
| Salience | -0.051 | 0.024 | 95.0 | 906.0 | 997.0 | 0.038 |
| Excessive use | 0.110 | 0.031 | 1.028 | 1.051 | 1.186 | < 0.001 |
| Neglect work | 0.028 | 0.038 | 1.028 | 0.095 | 1.108 | 0.464 |
| Anticipation | 0.317 | 0.053 | 1.405 | 1.237 | 1.524 | < 0.001 |
| Lack of control | -0.016 | 0.038 | 0.984 | 0.914 | 1.059 | 0.668 |
| Neglect social life | 0.034 | 0.045 | 1.034 | 0.946 | 1.130 | 0.460 |
| Age | 0.145 | 0.052 | 1.156 | 1.043 | 1.282 | 0.006 |
| Gender | 0.088 | 0.202 | 1.092 | 0.735 | 1.624 | 0.663 |
| Use smartphone before sleep | -1.049 | 0.389 | 0.350 | 0.163 | 0.751 | 0.007 |
5. Discussion
This study examined the impact of internet addiction on sleep quality among students. According to the results of this study, 46.6% of users have a mild internet addiction, and 30.8% have a moderate internet addiction. The results of a similar study conducted in Iran are close to those of this study. The same study by Yarahmadi et al. in Iran in 2020 found that 34.6% of the participants had mild internet addiction, and 57.6% had moderate to severe internet addiction (26). A systematic and meta-analysis study conducted by Salarvand et al. in 2020 on internet addiction among 16,585 Iranian students revealed that 31.51% were addicted to the Internet (27). A systematic study carried out by Modara et al. in Iran in 2017 found that internet addiction is prevalent in various parts of the country, ranging from 70% in southern Iran to 15% in western Iran (5). The high prevalence of internet addiction is due to the increasing availability of communication devices such as mobile phones (28) and laptops (29), as well as the development of technology, which affects all aspects of our daily lives. In addition, frequent and increasing internet use may also contribute to the fact that many users have some degree of internet addiction (30). In addition, the outbreak of the COVID-19 disease at the time of this study is another factor contributing to this access and excessive use of the Internet (31). Tahir conducted a similar study among 2749 medical students in 2021, finding that 67.6% were addicted to the Internet (32). Moreover, in a 2014 study by Mak et al. on the epidemiology of internet behavior and addiction among six Asian countries, the Philippines and Japan were found to have the highest rates of internet addiction with 51% and 48%, respectively (33). A systematic study by Kuss et al. in 2021 of 62 studies carried out worldwide (except in Europe) also revealed that internet addiction has a prevalence of between 12.6% to 67.5% (34). In a study by Karki et al. in 2021, 21.5% of participants had an addiction to the Internet, while 13.3% had a lower level of addiction (35). This difference is more justified than ever by attention to different geographical areas. As in Lozano-Blasco et al.’s systematic and meta-analysis study in 2022, internet addiction in countries such as Australia and New Zealand shows a declining trend (36). In comparison with Asian countries (36), there are significant differences. Different measurement expressions have caused this prevalence to vary (37).
According to the present study, 52.3% of the participants reported poor sleep quality. Among students in Iran in 2016, a systematic and meta-analysis study conducted by Ranjbaran reported a prevalence of poor sleep quality of 56% (38). Or a study by Rezaei et al. in 2018 among medical students in Iran showed that 60% had poor sleep quality (39). According to a similar study conducted by Khosravi et al. in 2021, among 4170 Iranian adults, 42.9% reported poor sleep quality (40). Khayat et al. conducted a study in 2018 among Saudi Arabian students aged 18 to 25, which also indicated that 54.4% of the participants had poor sleep quality (41). In the study Gomez-Chiappe et al. conducted in 2020 among Colombian students, 58.9% reported poor sleep quality (42). Based on a systematic and meta-analysis study by Jahrami et al. in 2021 among 54,231 participants from 13 countries, it was determined that 35.7% of the population suffered from sleep problems (43). According to these studies, students report significantly poorer sleep quality than the general population. Sleep problems are also associated with increased health concerns, irritability, depression, fatigue, attention and concentration problems, as well as poor academic performance (43).
According to our study, internet addiction adversely affects sleep quality and impairs sleep quality. In addition, this association was shown in studies conducted by Karimy et al. in 2020 among Iranian students (44) and Gupta et al. in India in 2021 (45). Zhang et al. found that 26.7% of Vietnamese youth addicted to the Internet had sleep problems in a study he conducted in 2017 (46). Other studies have shown that internet addiction is inversely related to sleep quality (32, 41, 47, 48). The same relationship was observed among adolescents (35, 47, 48), youth (27, 32, 46), and adults (49).
This study was limited by its cross-sectional nature, which prevented it from clearly demonstrating the causal relationship between variables. Participants’ self-reports may also be affected by biases such as reminder bias. A further limitation is that samples were selected from a university in Iran, and no samples were collected from other universities in Iran and abroad, which may affect the study’s external validity.
5.1. Conclusions
The study found that 79.2% of participants had mild to severe levels of internet addiction. In addition, 52.3% of the participants reported poor sleep quality. There was a negative association between internet addiction and sleep quality. Also, these results can be a source for universities and all educational institutions to control the adverse effects of using the Internet and improve sleep quality in students.
