- Email: [email protected]
Psychopathological factors associated with problematic alcohol and problematic Internet use in a sample of adolescents in Germany
Psychiatry Research 240 (2016) 272–277
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Psychopathological factors associated with problematic alcohol and problematic Internet use in a sample of adolescents in Germany Lutz Wartberg a,n, Romuald Brunner b, Levente Kriston c, Tony Durkee d, Peter Parzer b, Gloria Fischer-Waldschmidt b, Franz Resch b, Marco Sarchiapone e, Camilla Wasserman e, Christina W. Hoven f, Vladimir Carli d, Danuta Wasserman d, Rainer Thomasius a, Michael Kaess b a
German Center for Addiction Research in Childhood and Adolescence (DZSKJ), University Medical Center Hamburg-Eppendorf, Germany Department of Child and Adolescent Psychiatry, University Hospital Heidelberg, Heidelberg, Germany Department of Medical Psychology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany d National Centre for Suicide Research and Prevention of Mental Ill-Health (NASP), Karolinska Institutet, Stockholm, Sweden e Department of Health Sciences, University of Molise, Campobasso, Italy f Department of Epidemiology, Mailman School of Public Health, Columbia University, NY, USA b c
art ic l e i nf o
a b s t r a c t
Article history: Received 27 September 2015 Received in revised form 17 January 2016 Accepted 17 April 2016 Available online 22 April 2016
In Germany, high prevalence rates for problematic alcohol use and problematic Internet use in adolescents were reported. The objective of the present study was to identify psychopathological factors associated with these two behavior patterns. To our knowledge, this is the first investigation assessing psychopathological factors for both problematic alcohol and problematic Internet use in the same sample of adolescents. We surveyed a sample of 1444 adolescents in Germany regarding problematic alcohol use, problematic Internet use, psychopathology and psychological well-being. We conducted binary logistic regression analyses. 5.6% of the sample showed problematic alcohol use, 4.8% problematic Internet use, and 0.8% both problematic alcohol and problematic Internet use. Problematic alcohol use was higher in adolescents with problematic Internet use compared to those without problematic Internet use. Conduct problems and depressive symptoms were statistically significant associated with both problematic alcohol and problematic Internet use. Prosocial behavior was related to problematic Internet use. Male gender and less peer problems were associated with problematic alcohol use. For the first time associations between adolescent problematic alcohol and problematic Internet use due to common psychopathological factors were identified. However, in addition to shared factors, we found also specific psychopathological correlates associated with these two behavior patterns. & 2016 Published by Elsevier Ireland Ltd.
Keywords: Alcohol Alcohol abuse Internet addiction Internet gaming disorder Psychopathology Adolescents
1. Introduction 1.1. Prevalence of problematic alcohol and problematic Internet use in adolescents in Germany During the last several years, the surge in reported cases of problematic Internet use in adolescents across the globe continues to be a growing problem (e.g. Wang et al., 2013). Following the definition of Shaw and Black (2008) problematic Internet use (also commonly referred as internet addiction or pathological Internet use) is “…characterized by excessive or poorly controlled preoccupations, urges or behaviors regarding computer use and n
Corresponding author. E-mail address: [email protected] (L. Wartberg).
http://dx.doi.org/10.1016/j.psychres.2016.04.057 0165-1781/& 2016 Published by Elsevier Ireland Ltd.
internet access that lead to impairment or distress”. Akin to problematic Internet use, problematic alcohol use is also relatively prevalent in adolescent populations. In contrast to problematic Internet use, problematic alcohol use has been rigorously investigated and considered to be an important public health concern for the past several year (e.g. see McCambridge et al., 2011). In Germany, high prevalence rates for problematic Internet use and problematic alcohol use in adolescents were reported. Four recently published studies (Durkee et al., 2012; Rumpf et al., 2014; Tsitsika et al., 2014; Wartberg et al., 2015) had investigated samples of German adolescents and estimated the prevalence of problematic Internet use in this age group between 0.9% (Tsitsika et al., 2014) and 4.8% (Durkee et al., 2012). Problematic alcohol use was assessed by the Federal Center for Health Education (BZgA, 2015) in a representative sample of adolescents (aged between 12 and 17 years). The measure for problematic alcohol use in the
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
BZgA-survey was “frequent binge drinking” (consumption of five or more standard drinks of alcohol at least on 4 out of the last 30 days) and in 2010 5.2% of the German adolescents reported this behavior pattern. 1.2. Psychopathological factors associated with problematic alcohol and problematic Internet use in adolescence Problematic alcohol use or problematic Internet use in adolescents could be described as a risk behavior or “problem-behavior” (Jessor, 1987). In general, such behavior patterns occur more frequently in adolescence compared to other stages of life (Scheithauer et al., 2008). According to the “Problem Behavior Theory” of Jessor (1998) different problematic behavior patterns can be understood as an expression of an underlying syndrome (Scheithauer et al., 2008) and could share a similar pattern of psychopathological factors. For reasons of consistency (with the Problem Behavior Theory) we use for the two behavior patterns the terms “problematic alcohol use” and “problematic Internet use”. Regarding psychopathological factors related to adolescent problematic Internet use, the best empirical evidence is available for conduct problems (e.g. Ko et al., 2009) and depressive symptoms (e.g. Wang et al., 2013). In a smaller number of studies, associations between problematic Internet use in adolescents and ADHD (e.g. Carli et al., 2013a) or social phobia/social anxiety (e.g. Yen et al., 2007) were revealed. Conduct problems (i.e. externalizing problems) are the most important psychopathological factors associated with problematic alcohol use in adolescents (e.g. Young et al., 2008). However, the findings regarding the associations between adolescent problematic alcohol use and internalizing problems (depressive symptoms, anxiety), as well as the results concerning ADHD (Marshall, 2014) are mixed (Chartier et al., 2010). In summary, conduct or externalizing problems seem to be shared psychopathological factors related to problematic alcohol and problematic Internet use in adolescence. This could be partly facilitated due to underlying personality characteristics, e.g. a recent study reported higher impulsivity in male adult patients with Internet gaming disorder and as well in patients with alcohol use disorder (Choi et al., 2014). The importance of internalizing problems (especially depressive symptoms) is empirically well-established for problematic Internet use, but for adolescent problematic alcohol use, the results are inconsistent. To our knowledge, our study is the first assessing psychopathological factors for both problematic alcohol and problematic Internet use in the same sample of adolescents. Whereas until now only results in different samples and between studies could be compared, this methodological approach (using the same measures in one sample) facilitates the direct comparison of psychopathological factors associated with these two problematic behavior patterns. 1.3. Aim and research questions of the study The objective of the present study was to identify psychopathological factors associated with problematic alcohol and problematic Internet use. Furthermore, we wanted to compare the patterns of these psychopathological factors. We explored the following research questions: How frequently do problematic alcohol use, problematic Internet use and a combined problematic use of the Internet and alcohol occur in adolescents? Which psychopathological factors are associated with problematic alcohol use and with problematic Internet use? Which similarities and differences in the patterns of psychopathological factors can be revealed?
273
2. Methods 2.1. Data collection The data have been collected within the project “Saving and Empowering Young Lives in Europe – Mental health promotion through prevention of risky and self-destructive behavior” (SEYLE), funded by the European Union (Wasserman et al., 2010). This project is a randomized controlled multicenter study, involving ten European countries and Israel. A study protocol for the SEYLE project was compiled by Wasserman et al. (2010) and additionally, Carli et al. (2013b) described further methodological aspects of the survey (e.g. including randomization strategy, translations and reliability values of the used instruments). In Germany, the randomization was based on schools in the Rhein-Neckar-District (100 schools were randomized, 41 of these schools were requested for participation and 26 schools accepted to participate in the SEYLE project). Due to the unique design of the German school system, schools were classified to one out of three categories (“Gymnasium”¼high educational level/“Realschule” ¼medium educational level/“Hauptschule” ¼low educational level). Subsequently, the schools were selected and randomized separately for each category (Carli et al., 2013b). Adolescents (8th and 9th graders) were recruited from the randomly selected 26 German schools (more details on the process of randomization can be found in the description of Carli et al., 2013b). The sample was representative for 8th and 9th graders in the Rhein-Neckar-District, which shows a sociodemographic composition that is comparable to overall Germany. All data were collected through structured questionnaires administered to adolescents in the 26 schools. Data collection was carried out in January 2010 on a sample of 1444 adolescents (response rate: 72% of the 2016 adolescents who were invited to participate) using a standardized self-report questionnaire. This investigation was conducted within the confines of the school classrooms. Data collection was carried out by trained members of the project team (they administered the questionnaires). Teachers were present during the survey; however, they were not involved in the assessment. Written informed consent of adolescents and at least one caregiver were obtained prior to the study. The study was approved by the local ethical committee of the Medical Faculty of the University Hospital Heidelberg. 2.2. Measures Problematic Internet use in adolescents was measured for the last six months using the Young Diagnostic Questionnaire (YDQ, Young, 1998). Because of its brevity (eight items, binary “yes” vs. “no” response format), the YDQ is very well suited for epidemiological studies and was used in other studies to measure “problematic Internet use” (e.g. Huang et al., 2009). As a screening instrument, the YDQ can be used for the determination of a dimensional score of problematic Internet use (but not to confirm a diagnosis, e.g. of “Internet gaming disorder”, American Psychiatric Association, 2013). “By summing up the values of all eight items of the questionnaire, a YDQ sum score was calculated with a higher sum indicating higher risk levels of problematic Internet use. According to Young (1998), adolescents who answered yes to five or more of the criteria were classified as having problematic Internet use. To assess problematic alcohol use, we combined two measures of alcohol consumption (frequency and quantity). The adolescents were asked how often they had consumed an alcoholic drink (1 ¼“never”, 2 ¼“once a month or less”, 3 ¼“2–4 times a month”, 4 ¼“2–3 times a week”, 5 ¼“4 or more times a week”, 6 ¼“every day”, 7 ¼“several times a day”). An alcoholic drink was specified as
274
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
0.33 l of beer, a glass of wine, or 0.04 l of spirituous liquor (what is equivalent to 1 standard drink of alcohol in each case). Subsequently, the adolescents were asked about the average number of alcoholic drinks they had consumed on a typical day when drinking (1 ¼ “I never drink alcohol”, 2 ¼“1 or 2″, 3 ¼“3 or 4″, 4 ¼“5 or 6″, 5 ¼“7 or more”). When an adolescent had stated to consume 5 or more standard alcoholic drinks on a typical occasion (“bingedrinking” ) during “2–3 times a week” or more often he or she was classified as having problematic alcohol use. We used the Strengths and Difficulties Questionnaire (SDQ, Goodman, 1997) as a dimensional measure of mental health over the last six months assessed by the adolescents. The SDQ consists of 25 items (statements describing own behavior) with a 3-level response format (0 ¼“not true” , 1¼ “somewhat true” , 2 ¼“certainly true). Five scales are computed, each consisting of five items: ”emotional symptoms” , “conduct problems” , “hyperactivity/inattention” , “peer (relationship) problems” , and “prosocial behavior”. The scores of the scales emotional symptoms, conduct problems, hyperactivity and peer problems can be summed to generate a total difficulties score (Goodman, 1997). High scores in these four scales indicate more psychopathological problems, whereas a high value regarding “prosocial behavior” is appraised as strength. Furthermore, we used the German version of the revised Beck Depression Inventory (BDI-II, Hautzinger et al., 2006) to measure the severity of depressive symptoms within the last two weeks. The questionnaire consists of 21 items (19 with a 4-level response format and two questions with a 7-level response format, respectively). Considering the age of the adolescents, the BDI-II item “loss of libido” was not used in our investigation, and this particular missing value was replaced by the mean of the other items. A higher BDI-II sum value indicates more depressive symptoms. Additionally, we used the WHO-5 Well-Being Index (WHO-5, Bech, 2004) to assess psychological well-being within the last two weeks. The WHO-5 consists of five questions (assessing the frequency of feelings) with a 6-level response format (0 ¼“at no time” , 1¼ “some of the time” , 2¼ “less than half of time” , 3 ¼“more than half of the time” , 4¼ “most of the time’, 5 ¼”all of the time” ). A higher sum value indicates a higher level of psychological wellbeing. In addition, socio-demographic data of the adolescents were collected (e.g. gender). 2.3. Sample The respective study included 1444 students attending the 8th and 9th grade in 26 German schools (all types of schools, only special needs school were excluded). The sample consisted of 752 girls (52%) and 692 boys (48%). The mean age of adolescents were 14.65 (SD ¼0.81) years (only 4.4% percent of the sample were younger than 14 years). In total, 33% of the adolescent sample attended “Gymnasium” (high educational level), 31% attended “Realschule” (medium educational level) and 36% attended “Hauptschule” (low educational level). 2.4. Statistical analyses After excluding 1 of the 1,444 cases (0.1% of the whole sample), who did not provide a valid response to any of the YDQ or alcohol items, statistical analyses were performed in 1443 cases. We used SPSS version 22.0 (IBM, 2013, New York, USA) to calculate prevalence rates for problematic alcohol and problematic Internet use, the corresponding 95%-confidence intervals (bootstrapping with 10,000 samples) as well as to perform binary logistic regression analyses to identify associations. In the first logistic regression analysis, the response variable was problematic alcohol use (yes/no), and in the second analysis it was problematic
Internet use (yes/no). For the models, we used as explanatory variables gender, all five SDQ scales, the BDI sum value, the WHO5 sum value and finally we adjusted for the other problem behavior (problematic alcohol use for problematic Internet use and vice versa). The goodness of fit for the logistic regression models was verified by the Hosmer-Lemeshow test.
3. Results 3.1. Prevalence rates We found a prevalence rate of 5.6% (95% confidence interval: 4.4–6.8%) for problematic alcohol use (80 cases, 25 girls, 55 boys) and of 4.8% (this value was reported before by Durkee et al., 2012; 95% confidence interval: 3.8–6.0%) for problematic Internet use (69 cases, 36 girls, 33 boys). A combined co-morbid problematic alcohol and problematic Internet use was reported by 0.8% of the sample (95% confidence interval: 0.3–1.3%, 11 adolescents, 5 girls, 6 boys). Chi-square tests demonstrated no differences between girls and boys in the frequency of problematic Internet use (χ2 ¼ 0.00, df¼ 1, p ¼0.993) and combined co-morbid use (χ2 ¼0.21, df¼1, p ¼0.646). More boys (8.0%) than girls (3.3%) reported a problematic alcohol use (χ2 ¼ 14.94, df ¼1, p ¼ o0.001). Adolescents reporting a problematic Internet use showed statistically significant more frequently a problematic alcohol use (χ2 ¼15.10, df¼1, p ¼ o0.001) than adolescents without problematic Internet use (15.9% vs. 5.0%). 3.2. Psychopathological factors associated with problematic alcohol use More conduct problems and depressive symptoms, male gender and a lower level of peer problems were statistically significantly associated with problematic alcohol use (see Table 1 for adjusted odds ratios). The other five independent variables (emotional problems, hyperactivity-inattention, prosocial behavior, well-being and problematic Internet use) did not have a significant effect. The Hosmer-Lemeshow test was not statistically Table 1 Associations of psychopathological factors with problematic alcohol and problematic Internet use identified by binary logistic regression analyses. Parameter (questionnaire)
Problematic alcohol use (n ¼1408) OR (95% CI)
Problematic Internet use (n¼ 1408) OR (95% CI)
Gender Emotional symptoms (SDQ) Conduct problems (SDQ) Hyperactivity-inattention (SDQ) Peer problems (SDQ) Prosocial behavior (SDQ) Depressive symptoms (BDI-II) Psychological well-being (WHO-5) Problematic alcohol use Problematic Internet use (YDQ) Nagelkerkes R2 Hosmer-Lemeshow test
0.27*** (0.14;0.50) 0.95 (0.81;1.11)
0.90 (0.49;1.65) 0.91 (0.78;1.06)
1.29** (1.10;1.50) 1.12 (0.98;1.28)
1.20* (1.03;1.40) 1.13 (0.99;1.30)
0.75** (0.63;0.89) 0.92 (0.80;1.07) 1.07*** (1.03;1.11)
0.96 (0.82;1.13) 0.81** (0.70;0.94) 1.07*** (1.03;1.11)
0.99 (0.97;1.00)
1.00 (0.98;1.01)
– 1.56 (0.71;3.42)
1.54 (0.71;3.35) –
0.19 p¼ 0.721
0.17 p ¼ 0.104
OR ¼Odds Ration, CI ¼Confidence Interval, SDQ ¼Strengths and Difficulties Questionnaire, BDI-II ¼ Beck Depression Inventory revised, WHO-5 ¼WHO-5 Well-Being Index. *
p o0.050. p o 0.010. *** p o0.001. **
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
significant for the computed model (indicating a goodness of fit). Nagelkerkes R2 showed a value of 0.19 for the final model (see Table 1), indicating that the four factors explained about a fifth of the variation between the two groups (adolescents with and without problematic alcohol use). 3.3. Psychopathological factors associated with problematic Internet use More conduct problems and depressive symptoms as well as a lower level of prosocial behavior were statistically significantly associated with problematic Internet use (see Table 1 for adjusted odds ratios). The other six explanatory variables (gender, emotional problems, hyperactivity-inattention, peer problems, wellbeing and problematic alcohol use) did not have a significant effect in the model. Again, the Hosmer-Lemeshow test was not statistically significant. Nagelkerkes R2 was 0.17 in the final model (see Table 1), showing that the three psychopathological factors explained approximately one sixth of the variation between the two groups (adolescents with and without problematic Internet use).
4. Discussion In the present study, we estimated the prevalence rates of problematic alcohol and problematic Internet use in adolescence. To our knowledge for the first time, psychopathological factors associated with these two problematic behavior patterns were investigated in one sample of adolescents. In our sample, 5.6% of the adolescents in Germany showed a problematic use of alcohol and 4.8% a problematic use of the Internet (prevalence value reported before by Durkee et al., 2012). A combined co-morbid problematic alcohol and problematic Internet use showed 0.8% of the adolescents in the present study. German adolescents with a problematic Internet use showed more frequently a problematic alcohol use compared to adolescents without problematic Internet use. This result is in accordance with the finding reported by Ko et al. (2008) for adolescents in Taiwan. Until now, results on the comorbidity of substance-related and unrelated addictions in adolescence are extremely limited. Our results suggest some, but rather weak, associations between problematic alcohol use and problematic Internet use in German adolescents. According to Bibbey et al. (2015), both excessive Internet and alcohol use appear to be associated with a host of maladaptive outcomes. Regarding the pattern of psychopathological factors, we found in our sample of adolescents both similarities and differences between problematic alcohol use and problematic Internet use. Both problematic behavior patterns were associated with more conduct problems and depressive symptoms. For conduct problems, this finding is in line with many published results for problematic alcohol use (e.g. Stolle et al., 2009) and for problematic Internet use (e.g. Ko et al., 2009) in adolescence. To sum up, conduct problems is an established factor associated with both adolescent behavior patterns. In contrast to our findings regarding externalizing behavior, for the other common factor in our sample (depressive symptoms) there was only clear empirical evidence regarding problematic Internet use (e.g. Wang et al., 2013). For problematic alcohol use, previous findings were mixed. Following King et al. (2004), alcohol consumption in adolescence is stronger associated with depressive symptoms than with other internalizing problems (e.g. anxiety). However, for problematic alcohol use, Patrick and Schulenburg (2014) stated that depressive affect is not as powerful a predictor as are conduct problems. According to the results of Maslowsky and Schulenberg (2013), the combined presence of conduct
275
problems and depressive symptoms is especially for alcohol use of younger adolescents of high importance. But in general, the effects or interactions of combined (co-morbid) internalizing and externalizing problems for problematic alcohol and problematic Internet use in adolescence are hardly been investigated so far. In addition to conduct problems and depressive symptoms, gender and peer problems were also statistically significant associated with problematic alcohol use. Compared to the girls, a higher percentage of boys reported problematic alcohol use. Analogously to this finding, in 2010 statistically significantly more male (7.6%) than female (2.6%) adolescents in Germany showed “frequent binge drinking” (BZgA, 2015). Extreme patterns of alcohol consumption were more common in German boys than in girls in 2010. Taking this result into account, the relation between “male gender” and problematic alcohol use appears comprehensible. Surprisingly, in our sample peer problems was found to be inversely associated with problematic alcohol use (odds ratio is below 1). Possibly, the circumstances of adolescent alcohol consumption could provide an explanation for this result. Most often adolescents consume alcohol together with peers. Drinking is potentially associated with common activities, shared experiences, recognition, social acceptance, and maybe also with peer pressure (Han et al., 2012). Therefore, adolescents (drinking usually with peers) could view their peer relationships as positive and enriching, even when a part of their common activities (e.g. binge drinking) is regarded as harmful from a clinical or public health perspective. This finding (peer problems as a specific psychopathological factor for problematic alcohol use) needs a replication, because we have a minor overlap in the two confidence intervals (see Table 1). Prosocial behavior was statistically significant related to problematic Internet use in the present study. In contrast to conduct problems and depressive symptoms, prosocial behavior is inversely associated with problematic Internet use. Again a replication for this result (prosocial behavior as a specific psychopathological factor for problematic Internet use) is needed, because we have a substantial overlap in the confidence intervals (see Table 1). Until now, the number of studies investigating the role of prosocial behavior for problematic Internet use is very limited. Cao and Su (2006) reported a result similar to the finding in our study: Chinese adolescents diagnosed as problematic Internet users had lower scores on the SDQ subscale “prosocial behavior” compared to a control group with average Internet use. As limitations of our study, first we have to reflect on the constructs “problematic Internet use” and “problematic alcohol use”. Problematic Internet use is not included in the most important classification systems (ICD-10 or DSM-5), although a closely related construct (“Internet gaming disorder” ) was recently incorporated in the appendix of the DSM-5 (American Psychiatric Association, 2013) and recognized as a condition that requires further study (Schimmenti et al., 2014). Strittmatter et al. (2015) compared the items of the YDQ and the DSM-5 criteria for Internet gaming disorder and stated a large similarity with regard to both the criteria and the cut-off value. Therefore, the results of the present study could also be valuable for research regarding the new DSM-5 diagnosis Internet gaming disorder. Nonetheless, Billieux et al. (2015) emphasized that factors associated with problematic online gaming are not necessarily related to other types of problematic Internet use (like excessive use of social network or cybersex sites). The used screening instrument for problematic internet use, the Young Diagnostic Questionnaire (YDQ, Young, 1998), may represent a further limitation. The proposed cut-off value for the YDQ (Z5) by Young (1998) was never empirically validated and no information on sensitivity and specificity of the questionnaire is available to date. Maraz et al., (2015) described the methodological
276
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
problems in estimating a prevalence with a screening instrument (high proportion of false positive cases) if a disorder is relatively rare and the specificity of the instrument is not extremely high. Other limitations concern the selection of measured psychological constructs in the present survey. In a few studies the importance of motivation for the incidence of problematic Internet use was emphasized (e.g. Kardefelt-Winther, 2014), but such data were not collected in this study. Surprisingly, until now an internationally accepted definition of “problematic alcohol use” (or “risky drinking”, etc.) in adolescence is still lacking (another limitation of our study). In the UK, the Institute of Alcohol Studies (IAS) has published low risk drinking guidelines for adolescents, whereupon they should be at least 15 years old before start consuming alcohol, never drink more than once a week, and never exceed the recommended adult daily limits (3–4 units of alcohol for men and 2–3 units for women). The German Federal Center for Health Education (BZgA, 2015) reported in their regular surveys “frequent binge drinking” (consumption of five or more standard drinks of alcohol at least on 4 out of the last 30 days) as the measure for problematic alcohol use. In our study, the adolescents assigned to the group with problematic alcohol use had consumed at least two times a week five or six standard drinks of alcohol. According to this threshold, these adolescents definitely exceeded both the criteria of the IAS for low risk drinking and the BZgA criteria for frequent binge drinking (indicating a problematic alcohol use for this age group). Alternatively, we could have assessed problematic alcohol use by using a cut off value of a questionnaire (e.g. Alcohol Use Disorders Identification Test, Babor et al., 2001). A proceeding that would have been more in line with the approach we used for problematic Internet use. In general, problematic alcohol use in adolescence has a high importance for the development of this behavior in the course of life and is associated with alcohol related problems in adulthood (including alcohol dependence, e.g. Viner and Taylor, 2007). For the rather new phenomenon of problematic Internet use, there are only few longitudinal surveys published. Currently it is hard to predict, if a problematic use of the Internet is a stable behavior pattern and how it develops in adulthood. In the present study, we found first indications of associations between problematic alcohol use and problematic Internet use by identifying common psychopathological factors in a cross sectional design. However, in addition to shared factors, we also found specific psychopathological correlates associated with problematic Internet use and problematic alcohol use. The reported findings are important for prevention and clinical interventions. They underline the equal importance of psychiatric co-morbidities in both behavior patterns, and indicate that treatment of both externalizing (e.g conduct problems) and internalizing problems (e.g. depressive symptoms) should be taken into consideration when treating adolescents with problematic alcohol or problematic Internet use. However, some specific recommendations could also be made from our data: a training of social skills (e.g. to develop prosocial behavior) should be integrated in interventions for adolescents showing problematic Internet use as this is a specific deficit in this group. In adolescents reporting a problematic alcohol use, the interdependencies of social contacts and consumption of alcohol should be explored and (if necessary) should be targeted in prevention and treatment. Our results partially confirm the idea of a common syndrome for different adolescent problem behaviors as postulated in the Problem Behavior Theory of Jessor (1998), but replications in other studies are needed. For a deeper understanding of shared and specific psychopathological factors related to problematic alcohol use and problematic Internet use, further (prospective longitudinal) studies investigating combined substance-related and behavioral addictions are necessary.
Financial support The data for the present article has been gathered in the SEYLE project. The SEYLE project was supported by the European Union [Seventh Framework Program, FP7; Coordination Theme 1 (Health), grant agreement number HEALTH-F2-2009-223091].
Contributors DW was the project leader and coordinator of the SEYLE project. VC, CWH, CW, and MS were members of the executive committee. RB was the leader of the German study site. GFW and RB were involved in the data collection for the study in Germany. PP participated in the design of the study. LW and LK performed the statistical analysis. LW wrote the first draft of the manuscript. MK supervised the writing of the manuscript including critical revision of the manuscript. TD, FR, RT, VC, GFW, DW and LK also revised the article critically. All authors reviewed the manuscript for intellectual content. All authors read and approved the final version of the manuscript.
Conflict of interest None.
References American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, fifth ed. (DSM-5). American Psychiatric Association, Arlington. Babor, T.F., Higgins-Biddle, J.C., Saunders, J.B., Monteiro, M.G., 2001. The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Care, second ed. World Health Organization, Geneva. Bech, P., 2004. Measuring the dimensions of psychological general well-being by the WHO-5. QoL Newsl. 32, 15–16. Bibbey, A., Phillips, A.C., Ginty, A.T., Carroll, D., 2015. Problematic Internet use, excessive alcohol consumption, their comorbidity and cardiovascular and cortisol reactions to acute psychological stress in a student population. J. Behav. Addict. 4, 44–52. Billieux, J., Schimmenti, A., Khazaal, Y., Maurage, P., Heeren, A., 2015. Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research. J. Behav. Addict. 4, 119–123. Bundeszentrale für gesundheitliche Aufklärung (BZgA), 2015. Der Alkoholkonsum Jugendlicher und junger Erwachsener in Deutschland 2014 – Ergebnisse einer aktuellen Repräsentativbefragung und Trends [Alcohol consumption of adolescents and young adults in Germany 2014 – results of a representative survey and trends]. Bundeszentrale für gesundheitliche Aufklärung, Cologne. Cao, F., Su, L., 2006. Internet addiction among Chinese adolescents: prevalence and psychological features. Child Care Health Dev. 33, 275–281. Carli, V., Durkee, T., Wasserman, D., Hadlaczky, G., Despalins, R., Kramarz, E., Wasserman, C., Sarchiapone, M., Hoven, C.W., Brunner, R., Kaess, M., 2013a. The association between pathological internet use and comorbid psychopathology: a systematic review. Psychopathology 46, 1–13. Carli, V., Wasserman, C., Wasserman, D., Sarchiapone, M., Apter, A., Balazs, J., Bobes, J., Brunner, R., Corcoran, P., Cosman, D., Guillemin, F., Haring, C., Kaess, M., Kahn, J.P., Keeley, H., Keresztény, A., Iosue, M., Mars, U., Musa, G., Nemes, B., Postuvan, V., Reiter-Theil, S., Saiz, P., Varnik, P., Varnik, A., Hoven, C.W., 2013b. The saving and empowering young lives in Europe (SEYLE) randomized controlled trial (RCT): methodological issues and participant characteristics. BMC Public Health 13, 479. Chartier, K.G., Hesselbrock, M.N., Hesselbrock, V.M., 2010. Development and vulnerability factors in adolescent alcohol use. Child Adolesc. Psychiatr. Clin. N. Am. 19, 493–504. Choi, S.W., Kim, H.S., Kim, G.Y., Jeon, Y., Park, S.M., Lee, J.Y., Jung, H.Y., Sohn, B.K., Choi, J.S., Kim, D.J., 2014. Similarities and differences among Internet gaming disorder, gambling disorder and alcohol use disorder: a focus on impulsivity and compulsivity. J. Behav. Addict. 3, 246–253. Durkee, T., Kaess, M., Carli, V., Parzer, P., Wasserman, C., Floderus, B., Apter, A., Balazs, J., Barzilay, S., Bobes, J., Brunner, R., Corcoran, P., Cosman, D., Cotter, P., Despalins, R., Graber, N., Guillemin, F., Haring, C., Kahn, J.P., Mandelli, L., Marusic, D., Mészáros, G., Musa, G.J., Postuvan, V., Resch, F., Saiz, P.A., Sisask, M., Varnik, A., Sarchiapone, M., Hoven, C.W., Wasserman, D., 2012. Prevalence of pathological internet use among adolescents in Europe: demographic and social factors. Addiction 107, 2210–2222. Goodman, R., 1997. The strengths and difficulties questionnaire: a research note. J.
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
Child Psychol. Psychiatry 38, 581–586. Han, Y., Grogan-Kaylor, A., Delva, J., Castillo, M., 2012. The role of peers and parents in predicting alcohol consumption among chilean youth. Int. J. Child Adolesc. Health 5, 53–64. Hautzinger, M., Keller, F., Kühner, C., 2006. Beck Depressions-Inventar (BDI-II). Revision. [Beck-Depression-Inventory (BDI-II). Revised version.] Harcourt Test Services, Frankfurt. Huang, R.L., Lu, Z., Liu, J.J., You, Y.M., Pan, Z.Q., Wei, Z., He, Q., Wang, Z.Z., 2009. Features and predictors of problematic internet use in Chinese college students. Behav. Inf. Technol. 28, 485–490. Jessor, R., 1987. Problem-behavior theory, psychosocial development, and adolescent problem drinking. Br. J. Addict. 82, 331–342. Jessor, R., 1998. New Perspectives on Adolescent Risk Behavior. Cambridge University Press, Cambridge. Kardefelt-Winther, D., 2014. Problematizing excessive online gaming and its psychological predictors. Comput. Hum. Behav. 31, 118–122. King, S.M., Iacono, W.G., McGue, M., 2004. Childhood externalizing and internalizing psychopathology in the prediction of early substance use. Addiction 99, 1548–1559. Ko, C.H., Yen, J.Y., Yen, C.F., Chen, C.S., Weng, C.C., Chen, C.C., 2008. The association between internet addiction and problematic alcohol use in adolescents: the problem behavior model. Cyberpsychol. Behav. 11, 571–576. Ko, C.H., Yen, J.Y., Chen, C.S., Yeh, Y.C., Yen, C.F., 2009. Predictive values of psychiatric symptoms for Internet addiction in adolescents: a 2-year prospective study. Arch. Pediatr. Adolesc. Med. 163, 937–943. Marshall, E.J., 2014. Adolescent alcohol use: risks and consequences. Alcohol Alcohol. 49, 160–164. Maraz, A., Király, O., Demetrovics, Z., 2015. Commentary on: are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research (The diagnostic pitfalls of surveys: if you score positive on a test of addiction, you still have a good chance not to be addicted). J. Behav. Addict. 4, 151–154. Maslowsky, J., Schulenberg, J.E., 2013. Interaction matters: quantifying conduct problem x depressive symptoms interaction and its association with adolescent alcohol, cigarette, and marijuana use in a national sample. Dev. Psychopathol. 25, 1029–1043. McCambridge, J., McAlaney, J., Rowe, R., 2011. Adult consequences of late adolescent alcohol consumption: a systematic review of cohort studies. PLoS Med. 8, e1000413. Patrick, M.E., Schulenburg, J.E., 2014. Prevalence and predictors of adolescent alcohol useand binge drinking in the United States. Alcohol Res. 35, 193–200. Rumpf, H.J., Vermulst, A.A., Bischof, A., Kastirke, N., Gürtler, D., Bischof, G., Meerkerk, G.J., John, U., Meyer, C., 2014. Occurence of internet addiction in a general population sample: a latent class analysis. Eur. Addict. Res. 20, 159–166.
277
Scheithauer, H., Hayer, T., Niebank, K., 2008. Problemverhaltensweisen und risikoverhalten im jugendalter – ein überblick [problem behavior and risk behavior in adolescence – an overview]. In: Scheithauer, H., Hayer, T., Niebank, K. (Eds.), Problemverhalten und Gewalt im Jugendalter. Erscheinungsformen, Entstehungsbedingungen, Prävention und Intervention. Kohlhammer, Stuttgart, pp. 11–33. Schimmenti, A., Caretti, V., La Barbera, D., 2014. Internet gaming disorder or internet addiction? A plea for conceptual clarity. Clin. Neuropsychiatry 11, 145–146. Stolle, M., Sack, P.M., Thomasius, R., 2009. Binge drinking in childhood and adolescence: epidemiology, consequences, and interventions. Dtsch. Ärzteblatt Int. 106, 323–328. Strittmatter, E., Kaess, M., Parzer, P., Fischer, G., Carli, V., Hoven, C.W., Wasserman, C., Sarchiapone, M., Durkee, T., Apter, A., Bobes, J., Brunner, R., Cosman, D., Sisask, M., Värnik, P., Wasserman, D., 2015. Pathological Internet use among adolescents: comparing gamers and non-gamers. Psychiatry Res. 228, 128–135. Tsitsika, A., Janikian, M., Schoenmakers, T.M., Tzavela, E.C., Olafsson, K., Wójcik, S., Macarie, G.F., Tzavara, C., Richardson, C., 2014. Internet addictive behavior in adolescence: a cross-sectional study in seven European countries. Cyberpsychol. Behav. Soc. Netw. 17, 528–535. Viner, R.M., Taylor, B., 2007. Adult outcomes of binge drinking in adolescence: findings from a UK national birth cohort. J. Epidemiol. Community Health 61, 902–907. Wang, L., Luo, J., Bai, Y., Kong, J., Gao, W., Sun, X., 2013. Internet addiction of adolescents in China: prevalence, predictors, and association with well-being. Addict. Res. Theory 21, 62–69. Wartberg, L., Kriston, L., Kammerl, R., Petersen, K.U., Thomasius, R., 2015. Prevalence of pathological Iuse in a representative german sample of adolescents: results of a latent profile analysis. Psychopathology 48, 25–30. Wasserman, D., Carli, V., Wasserman, C., Apter, A., Balazs, J., Bobes, J., Bracale, R., Brunner, R., Bursztein-Lipsicas, C., Corcoran, P., Cosman, D., Durkee, T., Feldman, D., Gadoros, J., Guillemin, F., Haring, C., Kahn, J.P., Kaess, M., Keeley, H., Marusic, D., Nemes, B., Postuvan, V., Reiter-Theil, S., Resch, F., Sáiz, P., Sarchiapone, M., Sisask, M., Varnik, A., Hoven, C.W., 2010. Saving and empowering young lives in Europe (SEYLE): a randomized controlled trial. BMC Public Health 10, 192. Yen, J.Y., Ko, C.H., Yen, C.F., Wu, H.Y., Yang, M.J., 2007. The comorbid psychiatric symptoms of Internet addiction: attention deficit and hyperactivity disorder (ADHD), depression, social phobia, and hostility. J. Adolesc. Health 41, 93–98. Young, K.S., 1998. Internet addiction: the Emergence of a new clinical disorder. Cyberpsychol. Behav. 1, 237–244. Young, R., Sweeting, H., West, P., 2008. A longitudinal study of alcohol use and antisocial behaviour in young people. Alcohol Alcohol. 43, 204–214.
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Psychopathological factors associated with problematic alcohol and problematic Internet use in a sample of adolescents in Germany Lutz Wartberg a,n, Romuald Brunner b, Levente Kriston c, Tony Durkee d, Peter Parzer b, Gloria Fischer-Waldschmidt b, Franz Resch b, Marco Sarchiapone e, Camilla Wasserman e, Christina W. Hoven f, Vladimir Carli d, Danuta Wasserman d, Rainer Thomasius a, Michael Kaess b a
German Center for Addiction Research in Childhood and Adolescence (DZSKJ), University Medical Center Hamburg-Eppendorf, Germany Department of Child and Adolescent Psychiatry, University Hospital Heidelberg, Heidelberg, Germany Department of Medical Psychology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany d National Centre for Suicide Research and Prevention of Mental Ill-Health (NASP), Karolinska Institutet, Stockholm, Sweden e Department of Health Sciences, University of Molise, Campobasso, Italy f Department of Epidemiology, Mailman School of Public Health, Columbia University, NY, USA b c
art ic l e i nf o
a b s t r a c t
Article history: Received 27 September 2015 Received in revised form 17 January 2016 Accepted 17 April 2016 Available online 22 April 2016
In Germany, high prevalence rates for problematic alcohol use and problematic Internet use in adolescents were reported. The objective of the present study was to identify psychopathological factors associated with these two behavior patterns. To our knowledge, this is the first investigation assessing psychopathological factors for both problematic alcohol and problematic Internet use in the same sample of adolescents. We surveyed a sample of 1444 adolescents in Germany regarding problematic alcohol use, problematic Internet use, psychopathology and psychological well-being. We conducted binary logistic regression analyses. 5.6% of the sample showed problematic alcohol use, 4.8% problematic Internet use, and 0.8% both problematic alcohol and problematic Internet use. Problematic alcohol use was higher in adolescents with problematic Internet use compared to those without problematic Internet use. Conduct problems and depressive symptoms were statistically significant associated with both problematic alcohol and problematic Internet use. Prosocial behavior was related to problematic Internet use. Male gender and less peer problems were associated with problematic alcohol use. For the first time associations between adolescent problematic alcohol and problematic Internet use due to common psychopathological factors were identified. However, in addition to shared factors, we found also specific psychopathological correlates associated with these two behavior patterns. & 2016 Published by Elsevier Ireland Ltd.
Keywords: Alcohol Alcohol abuse Internet addiction Internet gaming disorder Psychopathology Adolescents
1. Introduction 1.1. Prevalence of problematic alcohol and problematic Internet use in adolescents in Germany During the last several years, the surge in reported cases of problematic Internet use in adolescents across the globe continues to be a growing problem (e.g. Wang et al., 2013). Following the definition of Shaw and Black (2008) problematic Internet use (also commonly referred as internet addiction or pathological Internet use) is “…characterized by excessive or poorly controlled preoccupations, urges or behaviors regarding computer use and n
Corresponding author. E-mail address: [email protected] (L. Wartberg).
http://dx.doi.org/10.1016/j.psychres.2016.04.057 0165-1781/& 2016 Published by Elsevier Ireland Ltd.
internet access that lead to impairment or distress”. Akin to problematic Internet use, problematic alcohol use is also relatively prevalent in adolescent populations. In contrast to problematic Internet use, problematic alcohol use has been rigorously investigated and considered to be an important public health concern for the past several year (e.g. see McCambridge et al., 2011). In Germany, high prevalence rates for problematic Internet use and problematic alcohol use in adolescents were reported. Four recently published studies (Durkee et al., 2012; Rumpf et al., 2014; Tsitsika et al., 2014; Wartberg et al., 2015) had investigated samples of German adolescents and estimated the prevalence of problematic Internet use in this age group between 0.9% (Tsitsika et al., 2014) and 4.8% (Durkee et al., 2012). Problematic alcohol use was assessed by the Federal Center for Health Education (BZgA, 2015) in a representative sample of adolescents (aged between 12 and 17 years). The measure for problematic alcohol use in the
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
BZgA-survey was “frequent binge drinking” (consumption of five or more standard drinks of alcohol at least on 4 out of the last 30 days) and in 2010 5.2% of the German adolescents reported this behavior pattern. 1.2. Psychopathological factors associated with problematic alcohol and problematic Internet use in adolescence Problematic alcohol use or problematic Internet use in adolescents could be described as a risk behavior or “problem-behavior” (Jessor, 1987). In general, such behavior patterns occur more frequently in adolescence compared to other stages of life (Scheithauer et al., 2008). According to the “Problem Behavior Theory” of Jessor (1998) different problematic behavior patterns can be understood as an expression of an underlying syndrome (Scheithauer et al., 2008) and could share a similar pattern of psychopathological factors. For reasons of consistency (with the Problem Behavior Theory) we use for the two behavior patterns the terms “problematic alcohol use” and “problematic Internet use”. Regarding psychopathological factors related to adolescent problematic Internet use, the best empirical evidence is available for conduct problems (e.g. Ko et al., 2009) and depressive symptoms (e.g. Wang et al., 2013). In a smaller number of studies, associations between problematic Internet use in adolescents and ADHD (e.g. Carli et al., 2013a) or social phobia/social anxiety (e.g. Yen et al., 2007) were revealed. Conduct problems (i.e. externalizing problems) are the most important psychopathological factors associated with problematic alcohol use in adolescents (e.g. Young et al., 2008). However, the findings regarding the associations between adolescent problematic alcohol use and internalizing problems (depressive symptoms, anxiety), as well as the results concerning ADHD (Marshall, 2014) are mixed (Chartier et al., 2010). In summary, conduct or externalizing problems seem to be shared psychopathological factors related to problematic alcohol and problematic Internet use in adolescence. This could be partly facilitated due to underlying personality characteristics, e.g. a recent study reported higher impulsivity in male adult patients with Internet gaming disorder and as well in patients with alcohol use disorder (Choi et al., 2014). The importance of internalizing problems (especially depressive symptoms) is empirically well-established for problematic Internet use, but for adolescent problematic alcohol use, the results are inconsistent. To our knowledge, our study is the first assessing psychopathological factors for both problematic alcohol and problematic Internet use in the same sample of adolescents. Whereas until now only results in different samples and between studies could be compared, this methodological approach (using the same measures in one sample) facilitates the direct comparison of psychopathological factors associated with these two problematic behavior patterns. 1.3. Aim and research questions of the study The objective of the present study was to identify psychopathological factors associated with problematic alcohol and problematic Internet use. Furthermore, we wanted to compare the patterns of these psychopathological factors. We explored the following research questions: How frequently do problematic alcohol use, problematic Internet use and a combined problematic use of the Internet and alcohol occur in adolescents? Which psychopathological factors are associated with problematic alcohol use and with problematic Internet use? Which similarities and differences in the patterns of psychopathological factors can be revealed?
273
2. Methods 2.1. Data collection The data have been collected within the project “Saving and Empowering Young Lives in Europe – Mental health promotion through prevention of risky and self-destructive behavior” (SEYLE), funded by the European Union (Wasserman et al., 2010). This project is a randomized controlled multicenter study, involving ten European countries and Israel. A study protocol for the SEYLE project was compiled by Wasserman et al. (2010) and additionally, Carli et al. (2013b) described further methodological aspects of the survey (e.g. including randomization strategy, translations and reliability values of the used instruments). In Germany, the randomization was based on schools in the Rhein-Neckar-District (100 schools were randomized, 41 of these schools were requested for participation and 26 schools accepted to participate in the SEYLE project). Due to the unique design of the German school system, schools were classified to one out of three categories (“Gymnasium”¼high educational level/“Realschule” ¼medium educational level/“Hauptschule” ¼low educational level). Subsequently, the schools were selected and randomized separately for each category (Carli et al., 2013b). Adolescents (8th and 9th graders) were recruited from the randomly selected 26 German schools (more details on the process of randomization can be found in the description of Carli et al., 2013b). The sample was representative for 8th and 9th graders in the Rhein-Neckar-District, which shows a sociodemographic composition that is comparable to overall Germany. All data were collected through structured questionnaires administered to adolescents in the 26 schools. Data collection was carried out in January 2010 on a sample of 1444 adolescents (response rate: 72% of the 2016 adolescents who were invited to participate) using a standardized self-report questionnaire. This investigation was conducted within the confines of the school classrooms. Data collection was carried out by trained members of the project team (they administered the questionnaires). Teachers were present during the survey; however, they were not involved in the assessment. Written informed consent of adolescents and at least one caregiver were obtained prior to the study. The study was approved by the local ethical committee of the Medical Faculty of the University Hospital Heidelberg. 2.2. Measures Problematic Internet use in adolescents was measured for the last six months using the Young Diagnostic Questionnaire (YDQ, Young, 1998). Because of its brevity (eight items, binary “yes” vs. “no” response format), the YDQ is very well suited for epidemiological studies and was used in other studies to measure “problematic Internet use” (e.g. Huang et al., 2009). As a screening instrument, the YDQ can be used for the determination of a dimensional score of problematic Internet use (but not to confirm a diagnosis, e.g. of “Internet gaming disorder”, American Psychiatric Association, 2013). “By summing up the values of all eight items of the questionnaire, a YDQ sum score was calculated with a higher sum indicating higher risk levels of problematic Internet use. According to Young (1998), adolescents who answered yes to five or more of the criteria were classified as having problematic Internet use. To assess problematic alcohol use, we combined two measures of alcohol consumption (frequency and quantity). The adolescents were asked how often they had consumed an alcoholic drink (1 ¼“never”, 2 ¼“once a month or less”, 3 ¼“2–4 times a month”, 4 ¼“2–3 times a week”, 5 ¼“4 or more times a week”, 6 ¼“every day”, 7 ¼“several times a day”). An alcoholic drink was specified as
274
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
0.33 l of beer, a glass of wine, or 0.04 l of spirituous liquor (what is equivalent to 1 standard drink of alcohol in each case). Subsequently, the adolescents were asked about the average number of alcoholic drinks they had consumed on a typical day when drinking (1 ¼ “I never drink alcohol”, 2 ¼“1 or 2″, 3 ¼“3 or 4″, 4 ¼“5 or 6″, 5 ¼“7 or more”). When an adolescent had stated to consume 5 or more standard alcoholic drinks on a typical occasion (“bingedrinking” ) during “2–3 times a week” or more often he or she was classified as having problematic alcohol use. We used the Strengths and Difficulties Questionnaire (SDQ, Goodman, 1997) as a dimensional measure of mental health over the last six months assessed by the adolescents. The SDQ consists of 25 items (statements describing own behavior) with a 3-level response format (0 ¼“not true” , 1¼ “somewhat true” , 2 ¼“certainly true). Five scales are computed, each consisting of five items: ”emotional symptoms” , “conduct problems” , “hyperactivity/inattention” , “peer (relationship) problems” , and “prosocial behavior”. The scores of the scales emotional symptoms, conduct problems, hyperactivity and peer problems can be summed to generate a total difficulties score (Goodman, 1997). High scores in these four scales indicate more psychopathological problems, whereas a high value regarding “prosocial behavior” is appraised as strength. Furthermore, we used the German version of the revised Beck Depression Inventory (BDI-II, Hautzinger et al., 2006) to measure the severity of depressive symptoms within the last two weeks. The questionnaire consists of 21 items (19 with a 4-level response format and two questions with a 7-level response format, respectively). Considering the age of the adolescents, the BDI-II item “loss of libido” was not used in our investigation, and this particular missing value was replaced by the mean of the other items. A higher BDI-II sum value indicates more depressive symptoms. Additionally, we used the WHO-5 Well-Being Index (WHO-5, Bech, 2004) to assess psychological well-being within the last two weeks. The WHO-5 consists of five questions (assessing the frequency of feelings) with a 6-level response format (0 ¼“at no time” , 1¼ “some of the time” , 2¼ “less than half of time” , 3 ¼“more than half of the time” , 4¼ “most of the time’, 5 ¼”all of the time” ). A higher sum value indicates a higher level of psychological wellbeing. In addition, socio-demographic data of the adolescents were collected (e.g. gender). 2.3. Sample The respective study included 1444 students attending the 8th and 9th grade in 26 German schools (all types of schools, only special needs school were excluded). The sample consisted of 752 girls (52%) and 692 boys (48%). The mean age of adolescents were 14.65 (SD ¼0.81) years (only 4.4% percent of the sample were younger than 14 years). In total, 33% of the adolescent sample attended “Gymnasium” (high educational level), 31% attended “Realschule” (medium educational level) and 36% attended “Hauptschule” (low educational level). 2.4. Statistical analyses After excluding 1 of the 1,444 cases (0.1% of the whole sample), who did not provide a valid response to any of the YDQ or alcohol items, statistical analyses were performed in 1443 cases. We used SPSS version 22.0 (IBM, 2013, New York, USA) to calculate prevalence rates for problematic alcohol and problematic Internet use, the corresponding 95%-confidence intervals (bootstrapping with 10,000 samples) as well as to perform binary logistic regression analyses to identify associations. In the first logistic regression analysis, the response variable was problematic alcohol use (yes/no), and in the second analysis it was problematic
Internet use (yes/no). For the models, we used as explanatory variables gender, all five SDQ scales, the BDI sum value, the WHO5 sum value and finally we adjusted for the other problem behavior (problematic alcohol use for problematic Internet use and vice versa). The goodness of fit for the logistic regression models was verified by the Hosmer-Lemeshow test.
3. Results 3.1. Prevalence rates We found a prevalence rate of 5.6% (95% confidence interval: 4.4–6.8%) for problematic alcohol use (80 cases, 25 girls, 55 boys) and of 4.8% (this value was reported before by Durkee et al., 2012; 95% confidence interval: 3.8–6.0%) for problematic Internet use (69 cases, 36 girls, 33 boys). A combined co-morbid problematic alcohol and problematic Internet use was reported by 0.8% of the sample (95% confidence interval: 0.3–1.3%, 11 adolescents, 5 girls, 6 boys). Chi-square tests demonstrated no differences between girls and boys in the frequency of problematic Internet use (χ2 ¼ 0.00, df¼ 1, p ¼0.993) and combined co-morbid use (χ2 ¼0.21, df¼1, p ¼0.646). More boys (8.0%) than girls (3.3%) reported a problematic alcohol use (χ2 ¼ 14.94, df ¼1, p ¼ o0.001). Adolescents reporting a problematic Internet use showed statistically significant more frequently a problematic alcohol use (χ2 ¼15.10, df¼1, p ¼ o0.001) than adolescents without problematic Internet use (15.9% vs. 5.0%). 3.2. Psychopathological factors associated with problematic alcohol use More conduct problems and depressive symptoms, male gender and a lower level of peer problems were statistically significantly associated with problematic alcohol use (see Table 1 for adjusted odds ratios). The other five independent variables (emotional problems, hyperactivity-inattention, prosocial behavior, well-being and problematic Internet use) did not have a significant effect. The Hosmer-Lemeshow test was not statistically Table 1 Associations of psychopathological factors with problematic alcohol and problematic Internet use identified by binary logistic regression analyses. Parameter (questionnaire)
Problematic alcohol use (n ¼1408) OR (95% CI)
Problematic Internet use (n¼ 1408) OR (95% CI)
Gender Emotional symptoms (SDQ) Conduct problems (SDQ) Hyperactivity-inattention (SDQ) Peer problems (SDQ) Prosocial behavior (SDQ) Depressive symptoms (BDI-II) Psychological well-being (WHO-5) Problematic alcohol use Problematic Internet use (YDQ) Nagelkerkes R2 Hosmer-Lemeshow test
0.27*** (0.14;0.50) 0.95 (0.81;1.11)
0.90 (0.49;1.65) 0.91 (0.78;1.06)
1.29** (1.10;1.50) 1.12 (0.98;1.28)
1.20* (1.03;1.40) 1.13 (0.99;1.30)
0.75** (0.63;0.89) 0.92 (0.80;1.07) 1.07*** (1.03;1.11)
0.96 (0.82;1.13) 0.81** (0.70;0.94) 1.07*** (1.03;1.11)
0.99 (0.97;1.00)
1.00 (0.98;1.01)
– 1.56 (0.71;3.42)
1.54 (0.71;3.35) –
0.19 p¼ 0.721
0.17 p ¼ 0.104
OR ¼Odds Ration, CI ¼Confidence Interval, SDQ ¼Strengths and Difficulties Questionnaire, BDI-II ¼ Beck Depression Inventory revised, WHO-5 ¼WHO-5 Well-Being Index. *
p o0.050. p o 0.010. *** p o0.001. **
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
significant for the computed model (indicating a goodness of fit). Nagelkerkes R2 showed a value of 0.19 for the final model (see Table 1), indicating that the four factors explained about a fifth of the variation between the two groups (adolescents with and without problematic alcohol use). 3.3. Psychopathological factors associated with problematic Internet use More conduct problems and depressive symptoms as well as a lower level of prosocial behavior were statistically significantly associated with problematic Internet use (see Table 1 for adjusted odds ratios). The other six explanatory variables (gender, emotional problems, hyperactivity-inattention, peer problems, wellbeing and problematic alcohol use) did not have a significant effect in the model. Again, the Hosmer-Lemeshow test was not statistically significant. Nagelkerkes R2 was 0.17 in the final model (see Table 1), showing that the three psychopathological factors explained approximately one sixth of the variation between the two groups (adolescents with and without problematic Internet use).
4. Discussion In the present study, we estimated the prevalence rates of problematic alcohol and problematic Internet use in adolescence. To our knowledge for the first time, psychopathological factors associated with these two problematic behavior patterns were investigated in one sample of adolescents. In our sample, 5.6% of the adolescents in Germany showed a problematic use of alcohol and 4.8% a problematic use of the Internet (prevalence value reported before by Durkee et al., 2012). A combined co-morbid problematic alcohol and problematic Internet use showed 0.8% of the adolescents in the present study. German adolescents with a problematic Internet use showed more frequently a problematic alcohol use compared to adolescents without problematic Internet use. This result is in accordance with the finding reported by Ko et al. (2008) for adolescents in Taiwan. Until now, results on the comorbidity of substance-related and unrelated addictions in adolescence are extremely limited. Our results suggest some, but rather weak, associations between problematic alcohol use and problematic Internet use in German adolescents. According to Bibbey et al. (2015), both excessive Internet and alcohol use appear to be associated with a host of maladaptive outcomes. Regarding the pattern of psychopathological factors, we found in our sample of adolescents both similarities and differences between problematic alcohol use and problematic Internet use. Both problematic behavior patterns were associated with more conduct problems and depressive symptoms. For conduct problems, this finding is in line with many published results for problematic alcohol use (e.g. Stolle et al., 2009) and for problematic Internet use (e.g. Ko et al., 2009) in adolescence. To sum up, conduct problems is an established factor associated with both adolescent behavior patterns. In contrast to our findings regarding externalizing behavior, for the other common factor in our sample (depressive symptoms) there was only clear empirical evidence regarding problematic Internet use (e.g. Wang et al., 2013). For problematic alcohol use, previous findings were mixed. Following King et al. (2004), alcohol consumption in adolescence is stronger associated with depressive symptoms than with other internalizing problems (e.g. anxiety). However, for problematic alcohol use, Patrick and Schulenburg (2014) stated that depressive affect is not as powerful a predictor as are conduct problems. According to the results of Maslowsky and Schulenberg (2013), the combined presence of conduct
275
problems and depressive symptoms is especially for alcohol use of younger adolescents of high importance. But in general, the effects or interactions of combined (co-morbid) internalizing and externalizing problems for problematic alcohol and problematic Internet use in adolescence are hardly been investigated so far. In addition to conduct problems and depressive symptoms, gender and peer problems were also statistically significant associated with problematic alcohol use. Compared to the girls, a higher percentage of boys reported problematic alcohol use. Analogously to this finding, in 2010 statistically significantly more male (7.6%) than female (2.6%) adolescents in Germany showed “frequent binge drinking” (BZgA, 2015). Extreme patterns of alcohol consumption were more common in German boys than in girls in 2010. Taking this result into account, the relation between “male gender” and problematic alcohol use appears comprehensible. Surprisingly, in our sample peer problems was found to be inversely associated with problematic alcohol use (odds ratio is below 1). Possibly, the circumstances of adolescent alcohol consumption could provide an explanation for this result. Most often adolescents consume alcohol together with peers. Drinking is potentially associated with common activities, shared experiences, recognition, social acceptance, and maybe also with peer pressure (Han et al., 2012). Therefore, adolescents (drinking usually with peers) could view their peer relationships as positive and enriching, even when a part of their common activities (e.g. binge drinking) is regarded as harmful from a clinical or public health perspective. This finding (peer problems as a specific psychopathological factor for problematic alcohol use) needs a replication, because we have a minor overlap in the two confidence intervals (see Table 1). Prosocial behavior was statistically significant related to problematic Internet use in the present study. In contrast to conduct problems and depressive symptoms, prosocial behavior is inversely associated with problematic Internet use. Again a replication for this result (prosocial behavior as a specific psychopathological factor for problematic Internet use) is needed, because we have a substantial overlap in the confidence intervals (see Table 1). Until now, the number of studies investigating the role of prosocial behavior for problematic Internet use is very limited. Cao and Su (2006) reported a result similar to the finding in our study: Chinese adolescents diagnosed as problematic Internet users had lower scores on the SDQ subscale “prosocial behavior” compared to a control group with average Internet use. As limitations of our study, first we have to reflect on the constructs “problematic Internet use” and “problematic alcohol use”. Problematic Internet use is not included in the most important classification systems (ICD-10 or DSM-5), although a closely related construct (“Internet gaming disorder” ) was recently incorporated in the appendix of the DSM-5 (American Psychiatric Association, 2013) and recognized as a condition that requires further study (Schimmenti et al., 2014). Strittmatter et al. (2015) compared the items of the YDQ and the DSM-5 criteria for Internet gaming disorder and stated a large similarity with regard to both the criteria and the cut-off value. Therefore, the results of the present study could also be valuable for research regarding the new DSM-5 diagnosis Internet gaming disorder. Nonetheless, Billieux et al. (2015) emphasized that factors associated with problematic online gaming are not necessarily related to other types of problematic Internet use (like excessive use of social network or cybersex sites). The used screening instrument for problematic internet use, the Young Diagnostic Questionnaire (YDQ, Young, 1998), may represent a further limitation. The proposed cut-off value for the YDQ (Z5) by Young (1998) was never empirically validated and no information on sensitivity and specificity of the questionnaire is available to date. Maraz et al., (2015) described the methodological
276
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
problems in estimating a prevalence with a screening instrument (high proportion of false positive cases) if a disorder is relatively rare and the specificity of the instrument is not extremely high. Other limitations concern the selection of measured psychological constructs in the present survey. In a few studies the importance of motivation for the incidence of problematic Internet use was emphasized (e.g. Kardefelt-Winther, 2014), but such data were not collected in this study. Surprisingly, until now an internationally accepted definition of “problematic alcohol use” (or “risky drinking”, etc.) in adolescence is still lacking (another limitation of our study). In the UK, the Institute of Alcohol Studies (IAS) has published low risk drinking guidelines for adolescents, whereupon they should be at least 15 years old before start consuming alcohol, never drink more than once a week, and never exceed the recommended adult daily limits (3–4 units of alcohol for men and 2–3 units for women). The German Federal Center for Health Education (BZgA, 2015) reported in their regular surveys “frequent binge drinking” (consumption of five or more standard drinks of alcohol at least on 4 out of the last 30 days) as the measure for problematic alcohol use. In our study, the adolescents assigned to the group with problematic alcohol use had consumed at least two times a week five or six standard drinks of alcohol. According to this threshold, these adolescents definitely exceeded both the criteria of the IAS for low risk drinking and the BZgA criteria for frequent binge drinking (indicating a problematic alcohol use for this age group). Alternatively, we could have assessed problematic alcohol use by using a cut off value of a questionnaire (e.g. Alcohol Use Disorders Identification Test, Babor et al., 2001). A proceeding that would have been more in line with the approach we used for problematic Internet use. In general, problematic alcohol use in adolescence has a high importance for the development of this behavior in the course of life and is associated with alcohol related problems in adulthood (including alcohol dependence, e.g. Viner and Taylor, 2007). For the rather new phenomenon of problematic Internet use, there are only few longitudinal surveys published. Currently it is hard to predict, if a problematic use of the Internet is a stable behavior pattern and how it develops in adulthood. In the present study, we found first indications of associations between problematic alcohol use and problematic Internet use by identifying common psychopathological factors in a cross sectional design. However, in addition to shared factors, we also found specific psychopathological correlates associated with problematic Internet use and problematic alcohol use. The reported findings are important for prevention and clinical interventions. They underline the equal importance of psychiatric co-morbidities in both behavior patterns, and indicate that treatment of both externalizing (e.g conduct problems) and internalizing problems (e.g. depressive symptoms) should be taken into consideration when treating adolescents with problematic alcohol or problematic Internet use. However, some specific recommendations could also be made from our data: a training of social skills (e.g. to develop prosocial behavior) should be integrated in interventions for adolescents showing problematic Internet use as this is a specific deficit in this group. In adolescents reporting a problematic alcohol use, the interdependencies of social contacts and consumption of alcohol should be explored and (if necessary) should be targeted in prevention and treatment. Our results partially confirm the idea of a common syndrome for different adolescent problem behaviors as postulated in the Problem Behavior Theory of Jessor (1998), but replications in other studies are needed. For a deeper understanding of shared and specific psychopathological factors related to problematic alcohol use and problematic Internet use, further (prospective longitudinal) studies investigating combined substance-related and behavioral addictions are necessary.
Financial support The data for the present article has been gathered in the SEYLE project. The SEYLE project was supported by the European Union [Seventh Framework Program, FP7; Coordination Theme 1 (Health), grant agreement number HEALTH-F2-2009-223091].
Contributors DW was the project leader and coordinator of the SEYLE project. VC, CWH, CW, and MS were members of the executive committee. RB was the leader of the German study site. GFW and RB were involved in the data collection for the study in Germany. PP participated in the design of the study. LW and LK performed the statistical analysis. LW wrote the first draft of the manuscript. MK supervised the writing of the manuscript including critical revision of the manuscript. TD, FR, RT, VC, GFW, DW and LK also revised the article critically. All authors reviewed the manuscript for intellectual content. All authors read and approved the final version of the manuscript.
Conflict of interest None.
References American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, fifth ed. (DSM-5). American Psychiatric Association, Arlington. Babor, T.F., Higgins-Biddle, J.C., Saunders, J.B., Monteiro, M.G., 2001. The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Care, second ed. World Health Organization, Geneva. Bech, P., 2004. Measuring the dimensions of psychological general well-being by the WHO-5. QoL Newsl. 32, 15–16. Bibbey, A., Phillips, A.C., Ginty, A.T., Carroll, D., 2015. Problematic Internet use, excessive alcohol consumption, their comorbidity and cardiovascular and cortisol reactions to acute psychological stress in a student population. J. Behav. Addict. 4, 44–52. Billieux, J., Schimmenti, A., Khazaal, Y., Maurage, P., Heeren, A., 2015. Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research. J. Behav. Addict. 4, 119–123. Bundeszentrale für gesundheitliche Aufklärung (BZgA), 2015. Der Alkoholkonsum Jugendlicher und junger Erwachsener in Deutschland 2014 – Ergebnisse einer aktuellen Repräsentativbefragung und Trends [Alcohol consumption of adolescents and young adults in Germany 2014 – results of a representative survey and trends]. Bundeszentrale für gesundheitliche Aufklärung, Cologne. Cao, F., Su, L., 2006. Internet addiction among Chinese adolescents: prevalence and psychological features. Child Care Health Dev. 33, 275–281. Carli, V., Durkee, T., Wasserman, D., Hadlaczky, G., Despalins, R., Kramarz, E., Wasserman, C., Sarchiapone, M., Hoven, C.W., Brunner, R., Kaess, M., 2013a. The association between pathological internet use and comorbid psychopathology: a systematic review. Psychopathology 46, 1–13. Carli, V., Wasserman, C., Wasserman, D., Sarchiapone, M., Apter, A., Balazs, J., Bobes, J., Brunner, R., Corcoran, P., Cosman, D., Guillemin, F., Haring, C., Kaess, M., Kahn, J.P., Keeley, H., Keresztény, A., Iosue, M., Mars, U., Musa, G., Nemes, B., Postuvan, V., Reiter-Theil, S., Saiz, P., Varnik, P., Varnik, A., Hoven, C.W., 2013b. The saving and empowering young lives in Europe (SEYLE) randomized controlled trial (RCT): methodological issues and participant characteristics. BMC Public Health 13, 479. Chartier, K.G., Hesselbrock, M.N., Hesselbrock, V.M., 2010. Development and vulnerability factors in adolescent alcohol use. Child Adolesc. Psychiatr. Clin. N. Am. 19, 493–504. Choi, S.W., Kim, H.S., Kim, G.Y., Jeon, Y., Park, S.M., Lee, J.Y., Jung, H.Y., Sohn, B.K., Choi, J.S., Kim, D.J., 2014. Similarities and differences among Internet gaming disorder, gambling disorder and alcohol use disorder: a focus on impulsivity and compulsivity. J. Behav. Addict. 3, 246–253. Durkee, T., Kaess, M., Carli, V., Parzer, P., Wasserman, C., Floderus, B., Apter, A., Balazs, J., Barzilay, S., Bobes, J., Brunner, R., Corcoran, P., Cosman, D., Cotter, P., Despalins, R., Graber, N., Guillemin, F., Haring, C., Kahn, J.P., Mandelli, L., Marusic, D., Mészáros, G., Musa, G.J., Postuvan, V., Resch, F., Saiz, P.A., Sisask, M., Varnik, A., Sarchiapone, M., Hoven, C.W., Wasserman, D., 2012. Prevalence of pathological internet use among adolescents in Europe: demographic and social factors. Addiction 107, 2210–2222. Goodman, R., 1997. The strengths and difficulties questionnaire: a research note. J.
L. Wartberg et al. / Psychiatry Research 240 (2016) 272–277
Child Psychol. Psychiatry 38, 581–586. Han, Y., Grogan-Kaylor, A., Delva, J., Castillo, M., 2012. The role of peers and parents in predicting alcohol consumption among chilean youth. Int. J. Child Adolesc. Health 5, 53–64. Hautzinger, M., Keller, F., Kühner, C., 2006. Beck Depressions-Inventar (BDI-II). Revision. [Beck-Depression-Inventory (BDI-II). Revised version.] Harcourt Test Services, Frankfurt. Huang, R.L., Lu, Z., Liu, J.J., You, Y.M., Pan, Z.Q., Wei, Z., He, Q., Wang, Z.Z., 2009. Features and predictors of problematic internet use in Chinese college students. Behav. Inf. Technol. 28, 485–490. Jessor, R., 1987. Problem-behavior theory, psychosocial development, and adolescent problem drinking. Br. J. Addict. 82, 331–342. Jessor, R., 1998. New Perspectives on Adolescent Risk Behavior. Cambridge University Press, Cambridge. Kardefelt-Winther, D., 2014. Problematizing excessive online gaming and its psychological predictors. Comput. Hum. Behav. 31, 118–122. King, S.M., Iacono, W.G., McGue, M., 2004. Childhood externalizing and internalizing psychopathology in the prediction of early substance use. Addiction 99, 1548–1559. Ko, C.H., Yen, J.Y., Yen, C.F., Chen, C.S., Weng, C.C., Chen, C.C., 2008. The association between internet addiction and problematic alcohol use in adolescents: the problem behavior model. Cyberpsychol. Behav. 11, 571–576. Ko, C.H., Yen, J.Y., Chen, C.S., Yeh, Y.C., Yen, C.F., 2009. Predictive values of psychiatric symptoms for Internet addiction in adolescents: a 2-year prospective study. Arch. Pediatr. Adolesc. Med. 163, 937–943. Marshall, E.J., 2014. Adolescent alcohol use: risks and consequences. Alcohol Alcohol. 49, 160–164. Maraz, A., Király, O., Demetrovics, Z., 2015. Commentary on: are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research (The diagnostic pitfalls of surveys: if you score positive on a test of addiction, you still have a good chance not to be addicted). J. Behav. Addict. 4, 151–154. Maslowsky, J., Schulenberg, J.E., 2013. Interaction matters: quantifying conduct problem x depressive symptoms interaction and its association with adolescent alcohol, cigarette, and marijuana use in a national sample. Dev. Psychopathol. 25, 1029–1043. McCambridge, J., McAlaney, J., Rowe, R., 2011. Adult consequences of late adolescent alcohol consumption: a systematic review of cohort studies. PLoS Med. 8, e1000413. Patrick, M.E., Schulenburg, J.E., 2014. Prevalence and predictors of adolescent alcohol useand binge drinking in the United States. Alcohol Res. 35, 193–200. Rumpf, H.J., Vermulst, A.A., Bischof, A., Kastirke, N., Gürtler, D., Bischof, G., Meerkerk, G.J., John, U., Meyer, C., 2014. Occurence of internet addiction in a general population sample: a latent class analysis. Eur. Addict. Res. 20, 159–166.
277
Scheithauer, H., Hayer, T., Niebank, K., 2008. Problemverhaltensweisen und risikoverhalten im jugendalter – ein überblick [problem behavior and risk behavior in adolescence – an overview]. In: Scheithauer, H., Hayer, T., Niebank, K. (Eds.), Problemverhalten und Gewalt im Jugendalter. Erscheinungsformen, Entstehungsbedingungen, Prävention und Intervention. Kohlhammer, Stuttgart, pp. 11–33. Schimmenti, A., Caretti, V., La Barbera, D., 2014. Internet gaming disorder or internet addiction? A plea for conceptual clarity. Clin. Neuropsychiatry 11, 145–146. Stolle, M., Sack, P.M., Thomasius, R., 2009. Binge drinking in childhood and adolescence: epidemiology, consequences, and interventions. Dtsch. Ärzteblatt Int. 106, 323–328. Strittmatter, E., Kaess, M., Parzer, P., Fischer, G., Carli, V., Hoven, C.W., Wasserman, C., Sarchiapone, M., Durkee, T., Apter, A., Bobes, J., Brunner, R., Cosman, D., Sisask, M., Värnik, P., Wasserman, D., 2015. Pathological Internet use among adolescents: comparing gamers and non-gamers. Psychiatry Res. 228, 128–135. Tsitsika, A., Janikian, M., Schoenmakers, T.M., Tzavela, E.C., Olafsson, K., Wójcik, S., Macarie, G.F., Tzavara, C., Richardson, C., 2014. Internet addictive behavior in adolescence: a cross-sectional study in seven European countries. Cyberpsychol. Behav. Soc. Netw. 17, 528–535. Viner, R.M., Taylor, B., 2007. Adult outcomes of binge drinking in adolescence: findings from a UK national birth cohort. J. Epidemiol. Community Health 61, 902–907. Wang, L., Luo, J., Bai, Y., Kong, J., Gao, W., Sun, X., 2013. Internet addiction of adolescents in China: prevalence, predictors, and association with well-being. Addict. Res. Theory 21, 62–69. Wartberg, L., Kriston, L., Kammerl, R., Petersen, K.U., Thomasius, R., 2015. Prevalence of pathological Iuse in a representative german sample of adolescents: results of a latent profile analysis. Psychopathology 48, 25–30. Wasserman, D., Carli, V., Wasserman, C., Apter, A., Balazs, J., Bobes, J., Bracale, R., Brunner, R., Bursztein-Lipsicas, C., Corcoran, P., Cosman, D., Durkee, T., Feldman, D., Gadoros, J., Guillemin, F., Haring, C., Kahn, J.P., Kaess, M., Keeley, H., Marusic, D., Nemes, B., Postuvan, V., Reiter-Theil, S., Resch, F., Sáiz, P., Sarchiapone, M., Sisask, M., Varnik, A., Hoven, C.W., 2010. Saving and empowering young lives in Europe (SEYLE): a randomized controlled trial. BMC Public Health 10, 192. Yen, J.Y., Ko, C.H., Yen, C.F., Wu, H.Y., Yang, M.J., 2007. The comorbid psychiatric symptoms of Internet addiction: attention deficit and hyperactivity disorder (ADHD), depression, social phobia, and hostility. J. Adolesc. Health 41, 93–98. Young, K.S., 1998. Internet addiction: the Emergence of a new clinical disorder. Cyberpsychol. Behav. 1, 237–244. Young, R., Sweeting, H., West, P., 2008. A longitudinal study of alcohol use and antisocial behaviour in young people. Alcohol Alcohol. 43, 204–214.