SEARCH

SEARCH BY CITATION

Keywords:

  • adolescents;
  • excessive daytime sleepiness;
  • Internet overuse;
  • Korea;
  • prevalence

Abstract

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

Aim:  The purpose of this study was to examine the association of Internet overuse with excessive daytime sleepiness (EDS).

Methods:  A total of 2336 high school students in South Korea (boys, 57.5%; girls, 42.5%) completed the structured questionnaire. The severity of Internet addiction was evaluated using Young's Internet addiction test.

Results:  The proportions of boys who were classified as Internet addicts and possible Internet addicts were 2.5% and 53.7%, respectively. For girls, the corresponding proportions were 1.9% and 38.9%, respectively. The prevalence of EDS was 11.2% (boys, 11.2%; girls, 11.1%). When Internet addicts were compared with non-addicts, they consisted of more boys, drank alcohol more, and considered their own health condition as poor. But smoking was not related with Internet addiction. The prevalence rate of EDS for Internet addicts was 37.7%, whereas that for possible Internet addicts and non-addicts was 13.9% and 7.4%, respectively. The prevalence of insomnia, witnessed snoring, apnea, teeth grinding, and nightmares was highest in Internet addicts, middle in possible addicts, and lowest in non-addicts. With adjustment for duration of Internet use, duration of sleep time, age, gender, smoking, taking painkillers due to headache, insomnia symptoms, witnessed apnea, and nightmares, the odds of EDS were 5.2-fold greater (95% confidence interval [CI]: 2.7–10.2) in Internet addicts and 1.9-fold greater (95%CI: 1.4–2.6) in possible Internet addicts compared to non-addicts.

Conclusion:  Internet addiction is strongly associated with EDS in adolescents. Clinicians should consider examining Internet addiction in adolescent cases of EDS.

IN GENERAL, TEENS have been found to use the Internet for more hours than adults.1 Given that the Internet widely used among adolescents, Internet addiction is now occurring. Researchers perceive Internet addiction as a social phenomenon that needs to be analyzed and explained in psychiatric or psychological terms. Internet addiction has been reported to be related to loneliness, depression, harm avoidance, anxiety symptoms, impulsivity, and attention-deficit–hyperactivity disorder.2–11

Students who spend more time using the Internet have less sleeping time and feel higher levels of tiredness.12 Studies, however, on the relationship between Internet overuse and physical health or sleep problems with independent empirical support have been rarely performed.13–15 Excessive daytime sleepiness (EDS) has been associated with risk of drowsy driving,16–19 injuries in the workplace,20,21 and poor school performance.22–24 It has become a major international health concern.23–26 Even though Internet overuse (or possible addictive aspects of the Internet) and EDS were widespread problems among adolescents, the effect of Internet overuse on EDS has not been studied as yet.

The purpose of the present study was to examine the association of Internet overuse with EDS in adolescents.

METHOD

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

Subjects

A cross-sectional study was performed for a sample of students from senior high schools in Gyeonggi province in the northwestern part of South Korea and Gwangju city in its southwestern part. An objective was established to obtain data from 0.5% of the total 445 430 students (231 868 boys and 213 562 girls) in both regions and the questionnaire was distributed to 0.6% of total students, considering that there will be no response or there will be incomplete answers from some questionnaire recipients. With the help of officials who were responsible for statistics at the Office of Education of each region, five clusters in Gyeonggi province and four clusters in Gwangju city were formed, maintaining sociodemographic similarity within each cluster. One school was randomly selected from each cluster. Institutional approval for the survey was obtained from each school and the survey was performed from September to November 2006. Investigators visited schools and explained the purpose of the study to students. They first distributed a short form of the questionnaire containing questions on sleep-related problems such as snoring, apnea and teeth grinding, which can be identified only by family members or room-mates. Investigators asked students to take the questionnaire home and have it completed by their parents, brothers or sisters, or other persons. Three days later, investigators visited the school again, distributed a full questionnaire to each student this time, and asked them to complete it, especially for sleep-related problems, based on the family member-completed short form of the questionnaire. A total of 2694 students participated in the survey and 2336 individuals completed the questionnaire, which resulted in an 86.7% response rate and collection of data from 0.52% of students.

Assessments

The questionnaire included queries on Internet addiction, EDS, durations of Internet use and night sleep, time that the subjects went to bed, and other sleep-related problems such as insomnia symptoms, witnessed apnea, snoring, teeth grinding, and nightmares. Demographic data collected included age, gender, self-cognitive health level, smoking status, alcohol consumption, school performance, and taking painkillers for headache.

The severity of Internet addiction was assessed using Young's Internet addiction test.27 We used the Korean version, which had been translated into Korean by Kim.28 This scale examines the degree of preoccupation, compulsive use, behavioral problems, emotional changes, and impact on life with regard to Internet usage. The 20 items of the Internet addiction scale are calibrated with scores ranging from 1 to 5 (total score ranging from 20 to 100), with a higher score reflecting a greater tendency toward addiction. Three types of Internet user groups were identified in accordance with the original scheme of Young: Internet addicted, possibly addicted, and non-addicted, whose scores on the Internet addiction scale were ≥70, 40–69, and ≤39, respectively. In the current sample Cronbach's α coefficient was 0.92.

Daytime sleepiness was measured using the Epworth Sleepiness Scale (ESS), a frequently used subjective sleepiness scale consisting of eight items.29 Possible ESS scores range from 0 to 24. A higher score on the ESS represents a greater propensity for sleepiness. In the present study EDS was defined as ESS >10. Snorers were defined as students who snored ≥3 days/week. Students were classified as having apnea, teeth grinding, or nightmares if they experienced each respective problem ≥3 times a week. Questions about insomnia during the previous month included: (i) ‘Do you have any difficulty in falling asleep at night?’ (difficulty in initiating sleep); (ii) ‘Do you wake up during the night after you have gone to sleep and have a difficulty in getting back to sleep?’ (difficulty in maintaining sleep); and (iii) ‘Do you wake up too early in the morning?’ (early morning awakening). The presence of difficulty in initiating sleep, difficulty in maintaining sleep, or early morning awakening was defined as occurrence ≥3 times a week. The presence of insomnia was defined as occurrence of insomnia subtypes.

Subjects were classified as smokers if they were currently smoking cigarettes, and non-smokers if they have never smoked or if they had smoked but quit smoking. They were also grouped into alcohol drinkers or non-drinkers. Self-rated academic performance was categorized into three levels: high, middle, and low. Self-perceived health status was also classified into three levels: good, fair and poor. Students were asked whether they had taken painkillers for headache during the last month. If painkillers had been taken one or more times, the subject was regarded as taking painkillers for headache.

Statistical analysis

The data are summarized as mean (±SD) for continuous variables, and as percentage for categorical variables. The comparison of means among groups was made through analysis of variance. The comparison of proportions was based on χ2 test. Logistic regression models were used to assess the independent relationship between EDS and Internet addiction and between EDS and demographic and health-related factors.

RESULTS

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

A total of 1343 boys (57.5%) and 993 girls (42.5%) completed the questionnaire. The mean age of the study subjects was 16.7 ± 1.0 years and the mean Internet usage was 2.0 ± 1.4 h/day (Table 1). The proportions of boys who were classified as Internet addicts and possible Internet addicts were 2.5% and 53.7%, respectively. For girls, the corresponding proportions were 1.9% and 38.9%, respectively. The prevalence of EDS was 11.2% (boys, 11.2%; girls, 11.1%).

Table 1.  Characteristics of study subjects (mean ± SD)
 Boys (n = 1343)Girls (n = 993)All
  • *

    Significantly different from girls (P < 0.001).

  • EDS, excessive daytime sleepiness.

Age (years)16.8 ± 1.0*16.5 ± 1.016.7 ± 1.0
Duration of Internet use (h)2.1 ± 1.5*1.9 ± 1.42.1 ± 1.4
Duration of night sleep (h)6.1 ± 1.46.1 ± 1.36.1 ± 1.3
Level of Internet addiction (%)   
 Non-addicted43.8*59.250.3
 Possibly addicted53.738.947.4
 Addicted2.51.92.3
EDS (%)11.211.111.2

Table 2 lists the baseline subject characteristics by level of Internet addiction. When Internet addicts were compared with non-addicts, they were a little older, consisted of more boys, drank alcohol more, and considered their own health condition as poor. The prevalence rate of drinking in possible Internet addicts was lower than the rate in the non-addicted group. The association between Internet addiction and smoking, school performance, or taking painkiller for headache was less clear.

Table 2.  Baseline characteristics vs level of Internet addiction
VariablesNon-addictedPossibly addictedInternet addictedχ2 (P)
(n = 1176)(n = 1107)(n = 53)
  • F.

Age (years), mean ± SD16.7 ± 1.016.6 ± 1.016.9 ± 1.12.4 (0.0915)
Gender (%)    
 Boys50.065.164.154.4 (<0.0001)
 Girls50.034.935.9 
Duration of Internet use (h), mean ± SD1.7 ± 1.22.3 ± 1.53.5 ± 2.483.5 (<0.0001)
Smoking (%)    
 Yes11.49.19.43.2 (0.1988)
 No88.690.990.6 
Drinking (%)    
 Yes20.417.932.17.8 (0.0202)
 No79.682.167.9 
School performance (%)    
 High grade33.730.732.14.2 (0.3842)
 Moderate grade37.037.830.2 
 Low grade29.331.537.7 
Self-cognitive health level (%)    
 Fair24.626.136.715.5 (0.0037)
 Good66.262.342.9 
 Poor9.211.620.4 
Taking painkiller for headache (%)    
 Yes16.815.126.45.4 (0.0662)
 No83.284.973.6 

The prevalence rate of EDS for Internet addicts was 37.7%, whereas that for possible Internet addicts and non-addicts was 13.9% and 7.4%, respectively (P < 0.0001, Fig. 1). When durations of sleep and Internet use, age, gender, smoking, taking painkillers due to headache, insomnia symptoms, witnessed apnea, and nightmares were adjusted, the prevalence rate of EDS decreased to 28.6% in Internet addicts but increased slightly in possible Internet addicts and non-addicts (13.5% and 7.9%, respectively). There was still a strong positive relationship, however, between severity of Internet addiction and prevalence of EDS. Table 3 presents the duration of night sleep, time that subjects went to bed and prevalence rates of other sleep problems by level of Internet addiction. Internet addicts significantly slept less than non-addicts or possible Internet addicts. The prevalence of every sleep problem was highest in Internet addicts, middle in possible addicts, and lowest in non-addicts, demonstrating an increased rate of sleep problems among more severely Internet-addicted students. Insomnia symptoms were prevalent among high school students. Even in non-Internet-addicted students, 19.3% of them complained about difficulty in initiating sleep or maintaining sleep, or waking up too early. Of possible Internet addicts and Internet addicts, 26.1% and 35.9%, respectively, had insomnia symptom(s).

image

Figure 1. (□) Crude and (inline image) adjusted prevalence of excessive daytime sleepiness (EDS) vs Internet addiction level. Adjusted variables: duration of Internet use, duration of sleep time, age, gender, smoking, taking painkillers due to headache, insomnia symptoms, witnessed apnea, and nightmares.

Download figure to PowerPoint

Table 3.  Sleep-related characteristics vs level of Internet addiction
VariablesNon-addictedPossibly addictedInternet addictedχ2 (P)
(n = 1176)(n = 1107)(n = 53)
  • EDS: yes, ESS > 10; no, ESS ≤ 10; for other sleep problems: Yes, ≥3 days/week; No, 0–2 days/week.

  • F. EDS, excessive daytime sleepiness; ESS, Epworth Sleepiness Scale.

Duration of night sleep (h), mean ± SD6.1 ± 1.36.1 ± 1.35.4 ± 1.67.7 (0.0005)
Bedtime (%)    
 Before or at 11.00 hours15.713.311.38.0 (0.4341)
 11.01–12.00 hours29.728.320.8 
 00.01–01.00 hours27.830.130.2 
 01.01–02.00 hours17.919.524.5 
 At or after 02.01 hours8.98.813.2 
Difficulty in initiating sleep (%)    
 Yes12.015.626.413.1 (0.0014)
 No88.084.473.6 
Difficulty in maintaining sleep (%)    
 Yes7.610.813.27.8 (0.0205)
 No92.489.286.8 
Early morning awakening (%)    
 Yes6.310.813.215.6 (0.0004)
 No93.789.286.8 
Insomnia symptoms (%)    
 Yes19.326.135.920.1 (<0.0001)
 No80.773.964.1 
Snoring (%)    
 Yes8.311.124.517.5 (0.0002)
 No91.788.975.5 
Witnessed apnea (%)    
 Yes1.83.311.320.0 (<0.0001)
 No98.296.788.7 
Teeth grinding (%)    
 Yes3.96.222.637.2 (<0.0001)
 No96.193.877.4 
Nightmares (%)    
 Yes6.07.218.913.4 (0.0012)
 No94.092.881.1 

Table 4 shows the association of Internet addiction with EDS after adjusting for age, gender, durations of sleep and Internet use, and other relevant covariates. With adjustment for age, gender, and durations of sleep and Internet use, the odds of EDS were 1.9-fold greater (95% confidence interval [CI]: 1.4–2.5) in possible Internet addicts and 5.9-fold greater (95%CI: 3.1–11.1) in Internet addicts than in non-addicts (model I). In addition to age, gender, and durations of sleep and Internet use, relevant variables that were selected through the stepwise method were included as independent variables in model II. Compared with model I, the adjusted odds ratio (OR) of EDS for Internet addicts was decreased slightly to 5.2 (95%CI: 2.7–10.2) and that for possible addicts was not changed. In model III, additionally adjusted for alcohol drinking, school performance, self-cognitive health level, snoring, and teeth grinding simultaneously with each variable listed in Table 4, the two ORs were almost the same as those in model II. Besides Internet addiction, taking painkillers for headache, insomnia symptoms, witnessed apnea, and nightmares independently and positively influenced EDS.

Table 4.  Factors independently associated with EDS
VariablesAdjusted OR
Model I OR (95%CI)Model II OR (95%CI)Model III OR (95%CI)
  • *

    P < 0.05.

  • Additionally adjusted for alcohol drinking, school performance, self-cognitive health level, snoring, and teeth grinding simultaneously with each variable listed in Table 4.

  • CI, confidence interval; EDS, excessive daytime sleepiness; OR, odds ratio.

Age1.0 (0.9–1.2)0.9 (0.8–1.1)0.9 (0.9–1.1)
Gender   
 Boys1.1 (0.9–1.5)1.2 (0.9–1.6)1.2 (0.9–1.7)
 Girls111
Duration of Internet use (h)1.2 (1.1–1.3)1.1 (1.0–1.2)*1.1 (1.0–1.2)*
Duration of sleep (h)1.0 (0.9–1.1)1.0 (0.9–1.2)1.0 (0.9–1.1)
Level of Internet addiction   
 Not addicted111
 Possibly addicted1.9 (1.4–2.5)1.9 (1.4–2.6)1.9 (1.4–2.6)
 Addicted5.9 (3.1–11.1)5.2 (2.7–10.2)4.9 (2.5–9.7)
Smoking   
 Current smoker 1.7 (1.2–2.6)1.5 (0.9–2.3)
 None smoker 11
Taking painkiller due to headache   
 Yes 1.8 (1.3–2.5)1.8 (1.3–2.5)
 No 11
Insomnia symptoms   
 Yes 1.5 (1.1–2.1)1.5 (1.1–2.0)
 No 11
Witnessed apnea   
 Yes 3.2 (1.8–5.7)2.7 (1.4–5.2)
 No 11
Nightmares   
 Yes 2.3 (1.5–3.5)2.1 (1.4–3.3)
 No 11

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

This study has found that Internet addiction and possible Internet addiction were positively associated with EDS with adjustment for relevant variables. In particular, Internet-addicted students were likely to have 5.2-fold more EDS than non-addicted students.

The concept of Internet addiction has been proposed as an explanation for uncontrollable and damaging use of the Internet. Symptoms of excessive Internet use have been compared to the criteria used to diagnose other addictions such as pathological gambling.30 The present study examined the association of Internet addiction with smoking and drinking. It was found that Internet addiction was not related with smoking habits. This result was consistent with those of previous studies by Jung,31 and by June et al., in which they found that Internet addiction did not influence smoking.32 The prevalence rate of alcohol drinking was significantly high in Internet addicts compared with non-addicts in the present study, but the rate in possible addicts was low and did not show a gradually increasing tendency as severity of Internet addiction increased. Those results from the present study and previous studies suggest that the pathway of addiction with regard to the Internet is different from that in smoking and alcohol drinking. It should not be overlooked, however, that it is easy for adolescents to purchase cigarettes or alcohol over the Internet and to indulge in smoking and drinking.33,34 The fact that possible Internet addicts drink alcohol less than non-addicts might indicate that indulging in Internet reduces alcohol drinking in adolescents. But Internet addicts drink alcohol more than possible Internet addicts and non-addicts. This suggests that it is important to prepare a program to prevent possible Internet addicts from becoming Internet addicts. It is interesting that the prevalence of taking painkillers for headache in non-addicts, possible addicts, and Internet addicts had a J shape (P = 0.0662) as in alcohol drinking. In a separate analysis, taking painkillers for headache was related with alcohol drinking (P = 0.0014).

Although smoking, alcohol drinking, and taking painkillers for headache had a non-significant or significant but J-shaped relationship with Internet addiction level, the stronger the Internet addiction became, the more they perceived their health as poor. The result was similarly shown in the relationship between Internet addiction levels and sleep problems. The percentage of each sleep problem was highest in Internet addicts, middle in possible addicts, and lowest in non-addicts, with an increasing prevalence shown as severity of Internet addiction increased. This indicates that Internet addiction is more directly related with physical health and sleep problems than smoking or alcohol habits. Among sleep problems, we clarified the effect of Internet addiction on EDS because it is relatively common in adolescents and disturbs adolescents' school activities, especially academic performance. It was found that Internet addiction influenced EDS highly and independently from duration of Internet use, duration of night sleep, age, gender, smoking, taking painkillers for headache, and other sleep problems. This suggests that Internet addicts experience a lower quality of sleep and tend to have EDS. Too much exposure to light at night through Internet use or television viewing may curtail the sleep time and cause poor sleep or insomnia.35–37

Usually it has been supposed that Internet addiction is closely related with EDS, but no study had been performed on this relationship using empirical data. The present study examined this association and found that Internet addiction influences EDS in adolescents. Research into treatment for Internet addiction is now being performed.38 Successful treatment for Internet addiction will improve physical health and sleep quality as well as upgrade psychological and social well-being.

CONCLUSION

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

The present study shows that Internet overuse was strongly associated with EDS and other sleep problems in adolescents. Because the number of Internet addicts will continue to grow, clinicians should consider examining Internet addiction in adolescent cases of EDS.

ACKNOWLEDGMENT

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES

This work was supported by the second stage of the Brain Korea 21 Project.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHOD
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. ACKNOWLEDGMENT
  8. REFERENCES