Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Nov 18.
Published in final edited form as: Addict Behav. 2013 Jan 16;38(4):2022–2025. doi: 10.1016/j.addbeh.2012.12.024

Substance abuse precedes internet addiction

Young Sik Lee a, Doug Hyun Han a,*, Sun Mi Kim a, Perry F Renshaw b
PMCID: PMC4651446  NIHMSID: NIHMS736752  PMID: 23384457

Abstract

The purpose of the current study was to evaluate possible overlapping substance abuse and internet addiction in a large, uniformly sampled population, ranging in age from 13 to 18 years. Participants (N=73,238) in the current study were drawn from the 6th Korea Youth Risk Behavior Web-based Survey (KYRBWS-V) for students from 400 middle schools and 400 high schools in 16 cities within South Korea. Of adolescent internet users, 85.2% were general users (GU), 11.9% were users with potential risk for internet addiction (PR), and 3.0% were users with high risk for internet addiction (HR). There was a difference in the number of students with alcohol drinking among the GU, PR, and HR groups (20.8% vs 23.1% vs 27.4%). There was a difference in the number of students who smoked among the GS, PR, and HR groups (11.7% vs 13.5% vs 20.4%). There was a difference in the number of students with drug use among the GU, PR, and HR groups (1.7% vs 2.0% vs 6.5%). After adjusting for sex, age, stress, depressed mood, and suicidal ideation, smoking may predict a high risk for internet addiction (OR=1.203, p=0.004). In addition, drug use may predict a high risk for internet addiction (OR=2.591, p<0.001). Because students with a high risk for internet addiction have vulnerability for addictive behaviors, co-morbid substance abuse should be evaluated and, if found, treated in adolescents with internet addiction.

Keywords: Internet use, Adolescence, Alcohol, Smoking, Drug, Co-morbidity

1. Introduction

Several studies have reported that the risk of internet addiction is associated with an increased prevalence of substance dependence (Bakken, Wenzel, Gotestam, Johansson, & Oren, 2009; Padilla-Walker, Nelson, Carroll, & Jensen, 2010). However, the subjects in most published studies have been adults and the number of subjects surveyed not been typically large enough to demonstrate significant co-morbidity of substance and internet addiction. Adolescence is a critical period which is characterized by risk-taking behavior, increased levels of novelty seeking and exploration, and intimate social interaction (Spear, 2000). In addition, 12% of 8th graders (13–14 years old) and 22% of 10th graders in USA students had episodes of heavy alcohol drinking within past two years (Johnston, O’Malley, Bachman, & Schulenberg, 2004).

The overlap between internet addiction and substance abuse and dependence may be due to similar characteristics predisposing toward, and brain regions responding to, internet use or substances. Individuals with internet addiction and substance addiction share similar temperaments. In a study of 686 high school students, Cho, Kim, Kim, Lee, and Kim (2008) reported that adolescents with problematic internet use showed higher self-directedness and cooperativeness and lower scores in novelty seeking and self-transcendence on a junior temperament and character inventory. In a study of 166 high school students, Lee et al. (2008) reported that adolescents with excessive internet use showed higher harm avoidance, relative to healthy comparison adolescents. In 88 patients with alcohol dependence, harm avoidance was negatively associated with the duration of abstinence (Ando et al., 2012) In addition, low novelty seeking and high self-directedness were also reported in adults with alcohol dependence (Anghelescu et al., 2010). Individuals with internet addiction and substance addiction may also share similar vulnerable brain regions including dorsolateral and orbitofrontal cortices (Crockford, Goodyear, Edwards, Quickfall, & el-Guebaly, 2005; Han, Hwang, & Renshaw, 2010; Ko et al., 2009).

Based on studies consistently reporting overlap and shared characteristics in individuals with internet and substance addiction, the purpose of the current study was to evaluate possible overlapping substance addiction and internet addiction in a large uniformly sampled population, ranging in age from 13 to 18 years.

2. Methods

2.1. Participants and data collecting

Seventy three thousand two hundred thirty eight participants in the current study were drawn from the 6th Korea Youth Risk Behavior Web-based Survey (KYRBWS-V) for students from 400 middle schools and 400 high schools in 16 cities within South Korea (KCDCP, 2011). The response rate for students was 97.7% (N=73,238). The IRB of Chung Ang University Hospital approved data analysis without informed consent considering that the KYRBWS-V cohort is a nationally representative group and the survey results did not contain protected health information.

2.1.1. Questionnaire of survey

The survey consisted of 128 questions in 14 fields including demographics, smoking, alcohol, drugs, obesity, eating behaviors, physical activity, prevention of trauma, sexual behaviors, psychiatric illness, oral hygiene, individual hygiene, atrophy/asthma, and internet use. Among these 14 fields, we extracted information from 6 fields including demographics, smoking, alcohol, drugs, psychiatric illness, and internet use for the purpose of our study. In a 2 week interval test-retest reliability assessment of the KYRBWS-V questionnaire, all of the self-reported health risk behavior indices had kappas ≥0.41(Bae et al., 2010).

The students were asked to check the Korean self-reporting internet addiction scale short form (KS-scale) developed by Kim, Jeng, Lee, Kim, and Joe (2008). The KS-scale consists of 20 items scored with a four-point Likert scale (1: “not at all” and 4: “always”). According to the scores of KS-scale total and subscales, students were classified into high risk for internet addiction, potential risk for internet addiction, and general user groups. As a result of reliability analysis of the KS-scale, the Cronbach’s alpha for middle school student scores was 0.909 (Kim et al., 2008). Alcohol use, smoking, and drug use were assessed by the questionnaires which were described in supplementary.

2.2. Statistical analysis

All the data from students in the current survey is presented as an average and standard deviation. The comparison of substance use characteristics between groups was analyzed with chi-square analysis and one way ANOVA. For confirmation that the results were not due to large sample size and for multiple comparisons instead of a substantive relationship between variables, Cramer’s V coefficient of less than 0.3 in chi-square testing and Bonferroni collection p<0.05/5 (internet use, alcohol, smoking, drug, and multi-substances) in one way ANOVA were applied. The correlations of the risk between internet addiction and substance addiction were conducted by binary logistic regression analyses after adjusting for age, sex, mental stress, depression, and suicidal ideation.

3. Results

3.1. Demographic characteristics

Eighty-two percent of Korean teenager students reported use of the internet in the current survey. Of internet users, 85.2% were general users, 11.9% were users with potential risk for internet addiction, and 3.0% were users with high risk for internet addiction. Twenty one point two percent of students reported drinking alcohol. The rate of alcohol drinking in male students (23.5%) was higher than the rate (18.7%) in female students (χ2=256.9, p<0.01). Twelve point two percent of students were smokers. One point nine percent of students reported drug use (Table 1).

Table 1.

Demographic data.

Boys (N=38,391, 52.4%) Girls (N=34,847, 47.6%) Total (N=73,238, 100%) t/χ2, p, Cramer’s V
Age (years) 15.1±1.7 15.1±1.8 15.1±1.7 t=1.5, p=0.14
School gradea χ2=23.7, p<0.01, V=0.02
 1st middle school 6519 (17.0) 5949 (17.1) 12,468 (17.0)
 2nd middle school 6620 (17.2) 5879 (16.9) 12,499 (17.1)
 3rd middle school 6817 (17.8) 5786 (16.6) 12,603 (17.2)
 1st high school 6229 (16.2) 5792 (16.6) 12.021 (16.4)
 2nd high school 6273 (16.3) 5851 (16.8) 12,124 (16.6)
 3rd high school 5933 (15.5) 5590 (16.0) 11,523 (15.7)
Economic statea χ2=383.9, p<0.01, V=0.07
 Rich 12,238 (31.9) 8834 (28.8) 21,072 (28.8)
 Average 17,061 (44.4) 17,192 (49.3) 34,253 (46.8)
 Poor 9092 (23.7) 8821 (25.3) 17,913 (24.5)
Stressa 3.3±1.0 3.5±0.9 3.4±1.0 t=40.5, p<0.01
Depressed mooda
 Yes/No 12,366 (32.2)/26,025 (67.8) 15,007 (43.1)/19,840 (56.9) 27,373 (37.4)/45,865 (62.6) χ2=919.5, p<0.01, V=0.11
Suicidal ideationa
 Yes/No 5855 (15.3)/32,536 (84.7) 8156 (23.4)/26,691 (76.6) 14,001 (19.3)/59,227 (80.7) χ2=785.1, p<0.01, V=0.10
Internet use
 WKD (use/non-use)a 25, 612 (66.7)/12,779 (33.3) 25, 706 (73.8)/9141 (26.2) 51,318 (70.1)/21,920 (29.9) χ2=433.5, p<0.01, V=0.08
 WKD (use time/min)a 123±103 113±87 118±96
 WED (use/non-use)a 28,882 (75.2)/9509 (24.8) 27,574 (79.1)/7273 (20.9) 56,456 (77.1)/16,782 (22.9) χ2=157.1, p<0.01, V=0.05
 WED (use time/min)a 197±131 163±109 181±122
 General users 31,574 (82.2) 30,746 (88.2) 62,320 (85.1) χ2=559.7, p<0.01, V=0.09
 Potential riska 5305 (13.8) 3417 (9.8) 8722 (11.9)
 High riska 1512 (3.9) 2194 (2.0) 2196 (3.0)
Alcohol (use/non-use)a 9040 (23.5)/29,351 (76.5) 6515 (18.7)/28,332 (81.3) 15,555 (21.2)/57,683 (78.8) χ2=256.9, p<0.01, V=0.06
Smoking (use/non-use)a 6482 (16.9)/31,909 (83.1) 2455 (7.0)/32,392 (93.0) 8937 (12.2)/64,301 (87.8) χ2=256.9, p<0.01, V=0.15
Drugs (use/non-use)a 814 (2.1)/37,577 (97.9) 586 (1.7)/34,261 (98.3) 1400 (1.9)/71,838 (98.1) χ2=18.7, p<0.01, V=0.02
Open in a new tab
a

Statistically significant (p<0.01, V<0.3), WKD: week day, WED: weekend, post hoc Cramer’s V coefficient.

3.2. Internet use pattern

The high risk group used the internet 202±149 min/day on weekdays and 318±172 min/day on weekends. The Potential group used internet 150±110 min/day on weekdays and 244±141 min/day on weekends. General users used the internet 110±87 min/day on weekdays and 166±110 min/day on weekends. Controlling for age, sex, depressed mood, and suicidal ideation, reported stress was positively correlated with the risk for internet addiction (B=0.216, S.E.=0.007, beta=0.103, t=30.7, p<0.01). Controlling for age, sex, stress, and suicidal ideation, the number of depressed students was higher in high risk group than in the potential group or among general users (B=0.685, S.E.=0.011, OR=1.944, p<0.001). Controlling for age, sex, stress, and depressed mood, the number of students with suicidal ideation was higher in the high risk group, compared to the potential group and general users (B=0.729, S.E.=0.013, OR=2.072, p<0.001).

3.3. Internet addiction risk vs alcohol use pattern

There was a difference in the number of students with alcohol drinking among the general user group, the potential risk group, and the high risk group (20.8% vs 23.1% vs 27.4%, χ2=73.5, p<0.01). However, alcohol use did not predict an increased risk for internet addiction after adjusting for sex, age, stress, depressed mood, and suicidal ideation (beta=0.06, S.E.=0.05, OR=1.235, p=0.266) (Table 2). Adjusting for sex, age, depressed mood, sadness, and suicidal ideation, the age of first drinking alcohol was earlier in the high risk group, compared to general user and the potential risk groups (beta=0.353, S.E.=0.07, t=5.101, p<0.01). Adjusting for sex, age, depressed mood, sadness, and suicidal ideation, heavy alcohol drinking predicted a high risk for internet addiction (beta=0.037, S.E.=0.09, OR=1.038, p<0.01).

Table 2.

The binary logistic regression analyses of substance use for high risk group for internet addiction comparing general and potential risk group.

Category Beta S.E. OR 95% CI p-values
Alcohol 0.06 0.05 1.235 0.957–1.173   0.266
Smoking 0.185 0.064 1.203 1.060–1.365   0.004
Drugs 0.952 0.104 2.591 2.111–3.180 <0.001
Multi-substances
 One substance 0.122 0.057 1.129 1.011–1.262   0.032
 Two substances 0.101 0.073 1.106 0.958–1.277   0.168
 Three substances 2.225 0.256 9.256 5.600–15.297 <0.001
 Non user Ref
Open in a new tab

S.E.: standard error; OR: odd ratio; CI: confidential interval; adjusting for the covariate variables: sex, age, stress, sadness, and suicidal ideation; one/two/three substances: students using one/two/all of three substances (alcohol, tobacco, and drugs).

3.4. Internet addiction risk vs smoking pattern

There was a difference in the number of students who smoked among the general user group, the potential risk group, and the high risk group (11.7% vs 13.5% vs 20.4%, χ2=73.5, p<0.01). Moreover, smoking may predict a high risk for internet addiction after adjusting for sex, age, stress, depressed mood, and suicidal ideation (beta=0.185, S.E.=0.06, OR=1.203, p=0.004). Adjusting for sex, age, depressed mood, and suicidal ideation, the age of first smoking was earlier in the high risk group, compared to the general user group and the potential risk group (beta=748.9, S.E.=93.8, t=7.983, p<0.01). Adjusting for sex, age, depressed mood, sadness, and suicidal ideation, heavy smoking predicted a high risk for internet addiction (beta=0.410, S.E.=0.114, OR=1.507, p<0.01).

3.5. Internet addiction risk vs substance use pattern

There was a difference in the number of students with drug use among the general user group, the potential risk group, and the high risk group (1.7% vs 2.0% vs 6.5%, χ2=73.5, p<0.01). Moreover, drug use may predict high risk for internet addiction after adjusting for sex, age, stress, depressed mood, and suicidal ideation (beta=0.952, S.E.=0.104, OR=2.591, p<0.001) (Table 2). In addition, all substance use can strongly predict high risk for internet addiction after adjusting for sex, age, stress, depressed mood, and suicidal ideation (beta=2.225, S.E.=0.256, OR=9.256, p<0.001) (Table 2).

4. Discussion

The current study suggests that the risk for internet addiction is associated with smoking and drug use in teenage students. Moreover, the severity of internet use was positively correlated with the severity of alcohol use, smoking, and drug use. In addition, adolescents with earlier ages of onset of substance use and multiple substance use were more likely to be at increased risk of internet addiction. The present study is valuable for its uniform sampling of a large population with 73,000 students, the correlation of patterns between chemical addiction and behavioral addiction, and the inclusion of younger subjects.

Several studies have reported a correlation between internet addiction and substance addiction. The current survey showed that the percentage of students who drank alcohol (23.1%) or who smoked (13.5) in the potential risk group was greater than the percentage observed among general users (alcohol: 20.8%, smoke 11.7%). The high risk group drank alcohol (27.4%) more and smoked more (20.4%) than the potential risk group. In addition, earlier alcohol use and smoking were correlated with a high risk for internet addiction. Smoking, but not alcohol use, can predict a high risk for internet addiction. Similar to these observation, a survey of 2336 Korean high school students by Choi et al. (2009) noted that the prevalence of alcohol use was not linearly correlated with the severity of internet addiction (Choi et al., 2009). The students with internet addiction drank alcohol more (32.1%), compared to the non-addiction group (20.4%). However, the rate of alcohol use in the possible addiction cohort (17.9%) was lower than in the non-addicted group. The different correlation of internet addiction risk with alcohol and smoking may be due to the different pharmacological effects of alcohol, a sedative, and nicotine, a stimulant. In addition, internet use and alcohol use may substitute for each other. As alcohol use generally takes place in groups, adolescents also play Massive multi-user on-line role playing game (MMORPG). MMORPG is known to be associated with high risk for internet addiction (Achab et al., 2011).

The prevalence rate of alcohol use (23.5%) in all students was almost twice than that of smoking (12.2%) in the current survey. In addition, a strong correlation between drug use and the risk for internet addiction was observed. The prevalence of drug use in all students is 1.9%. The drug use odds ratio for predicting a high risk for internet addiction is 2.59 compared to OR of 1.51 in smokers and an OR of 1.04 in drinkers. Moreover, multi-substance use strongly predicts a high risk for internet addiction (OR=9.26). Those results suggest that chemical and behavioral addictions may share similar biological characteristics. Recent genetic studies and reviews have suggested that the strong association among onset age of smoking, alcohol, and drug use provide evidence for biological vulnerability including heritability (Bierut et al., 1998; Kapoor et al., 2012).

The current study has several limitations. First, because the present study is a secondary analysis, we cannot interpret further the correlation between variables. Second, because the information was obtained by self-report, adolescents’ perceptions could have been inaccurate. Finally, the administration of a long questionnaire may have increased a tendency to provide inaccurate answers in some students.

5. Conclusions

As far as we know, this is the first large scale study of the correlation between substance abuse and internet addiction in teenagers. Because students with a high risk for internet addiction have vulnerability for addictive behaviors, co-morbid substance abuse should be evaluated and, if found, treated in adolescents with internet addiction.

Supplementary Material

supplemental

HIGHLIGHTS.

  • ►

    Of internet users, 3.0 percent were users with high risk for internet addiction.

  • ►

    The severity of internet use was correlated with the severity of substance use.

  • ►

    Earlier onset of substance use was associated with internet addiction.

Acknowledgments

We thank Korean Game Culture Foundation and Korea Center for Disease Control and Prevention (KCDCP) for presenting of data.

Role of funding sources: Funding for this study was the Korea Centers for Disease Control & Prevention (KCDCP), 2011. This work was also supported by Korean Game Culture Foundation. KCDCP and KGCF had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.addbeh.2012.12.024.

Footnotes

Contributors

Doug Hyun Han and Perry Renshaw designed the study and wrote the protocol. Young Sik Lee and Sun Mi Kim conducted literature searches and provided summaries of previous research studies. Doug Hyun Han conducted the statistical analysis. Doug Hyun Han and Perry Renshaw wrote the first draft of the manuscript and all authors contributed to and have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflicts of interest.

References

  1. Achab S, Nicolier M, Mauny F, Monnin J, Trojak B, Vandel P, et al. Massively multiplayer online role-playing games: Comparing characteristics of addict vs non-addict online recruited gamers in a French adult population. BMC Psychiatry. 2011;11:144. doi: 10.1186/1471-244X-11-144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ando B, Must A, Kurgyis E, Szkaliczki A, Drotos G, Rozsa S, et al. Personality traits and coping compensate for disadvantageous decision-making in long-term alcohol abstinence. Alcohol and Alcoholism. 2012;47:18–24. doi: 10.1093/alcalc/agr144. [DOI] [PubMed] [Google Scholar]
  3. Anghelescu I, Klawe C, Singer P, Fehr C, Hiemke C, Quante A, et al. Low novelty seeking and high self directedness scores in alcohol-dependent patients without comorbid psychiatric disorders homozygous for the A10 allele of the dopamine transporter gene. The World Journal of Biological Psychiatry. 2010;11:382–389. doi: 10.1080/15622970701775058. [DOI] [PubMed] [Google Scholar]
  4. Bae J, Joung H, Kim JY, Kwon KN, Kim YT, Park SW. Test–retest reliability of a questionnaire for the Korea Youth Risk Behavior Web-based Survey. Journal of Preventive Medicine and Public Health. 2010;43:403–410. doi: 10.3961/jpmph.2010.43.5.403. [DOI] [PubMed] [Google Scholar]
  5. Bakken IJ, Wenzel HG, Gotestam KG, Johansson A, Oren A. Internet addiction among Norwegian adults: A stratified probability sample study. Scandinavian Journal of Psychology. 2009;50:121–127. doi: 10.1111/j.1467-9450.2008.00685.x. [DOI] [PubMed] [Google Scholar]
  6. Bierut LJ, Dinwiddie SH, Begleiter H, Crowe RR, Hesselbrock V, Nurnberger JI, Jr, et al. Familial transmission of substance dependence: Alcohol, marijuana, cocaine, and habitual smoking: A report from the collaborative study on the genetics of alcoholism. Archives of General Psychiatry. 1998;55:982–988. doi: 10.1001/archpsyc.55.11.982. [DOI] [PubMed] [Google Scholar]
  7. Cho SC, Kim JW, Kim BN, Lee JH, Kim EH. Biogenetic temperament and character profiles and attention deficit hyperactivity disorder symptoms in Korean adolescents with problematic Internet use. Cyberpsychology, Behavior, and Social Network. 2008;11:735–737. doi: 10.1089/cpb.2007.0285. [DOI] [PubMed] [Google Scholar]
  8. Choi K, Son H, Park M, Han J, Kim K, Lee B, et al. Internet overuse and excessive daytime sleepiness in adolescents. Psychiatry and Clinical Neurosciences. 2009;63:455–462. doi: 10.1111/j.1440-1819.2009.01925.x. [DOI] [PubMed] [Google Scholar]
  9. Crockford DN, Goodyear B, Edwards J, Quickfall J, el-Guebaly N. Cue-induced brain activity in pathological gamblers. Biological Psychiatry. 2005;58:787–795. doi: 10.1016/j.biopsych.2005.04.037. [DOI] [PubMed] [Google Scholar]
  10. Han DH, Hwang JW, Renshaw PF. Bupropion sustained release treatment decreases craving for video games and cue-induced brain activity in patients with Internet video game addiction. Experimental and Clinical Psychopharmacology. 2010;18:297–304. doi: 10.1037/a0020023. [DOI] [PubMed] [Google Scholar]
  11. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. (NIH Pub No. 05-5727 1 secondary school students).Monitoring the future, national survey results on drug use, 1975–2004. 2004 [Google Scholar]
  12. Kapoor M, Wang JC, Bertelsen S, Bucholz K, Budde JP, Hinrichs A, et al. Variants located upstream of CHRNB4 on chromosome 15q25.1 are associated with age at onset of daily smoking and habitual smoking. PLoS One. 2012;7:e33513. doi: 10.1371/journal.pone.0033513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. KCDCP. The statistics of 5th Korea youth risk behavior web-based survey (KYRBWS) in 2010. 2011. Korea Centers for Disease Control and Prevention. In. [Google Scholar]
  14. Kim DI, Jeng YJ, Lee EA, Kim DM, Joe YM. Developmental of internet addiction proneness scale-short form (KS-scale) Korean Journal of Counseling. 2008;9:1703–1722. [Google Scholar]
  15. Ko CH, Liu GC, Hsiao S, Yen JY, Yang MJ, Lin WC, et al. Brain activities associated with gaming urge of online gaming addiction. Journal of Psychiatric Research. 2009;43:739–747. doi: 10.1016/j.jpsychires.2008.09.012. [DOI] [PubMed] [Google Scholar]
  16. Lee YS, Han DH, Yang KC, Daniels MA, Na C, Kee BS, et al. Depression like characteristics of 5HTTLPR polymorphism and temperament in excessive internet users. Journal of Affective Disorders. 2008;109:165–169. doi: 10.1016/j.jad.2007.10.020. [DOI] [PubMed] [Google Scholar]
  17. Padilla-Walker LM, Nelson LJ, Carroll JS, Jensen AC. More than a just a game: Video game and internet use during emerging adulthood. Journal of Youth and Adolescence. 2010;39:103–113. doi: 10.1007/s10964-008-9390-8. [DOI] [PubMed] [Google Scholar]
  18. Spear LP. The adolescent brain and age-related behavioral manifestations. Neuroscience and Biobehavioral Reviews. 2000;24:417–463. doi: 10.1016/s0149-7634(00)00014-2. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplemental
NIHMS736752-supplement-supplemental.doc (102.5KB, doc)

RESOURCES