 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Department of Social Medicine, Institute of Public Health, Faculty of Health Sciences
c Department of Pharmacology and Pharmacotherapy, Section of Social Pharmacy, Faculty of Pharmaceutical Sciences, University of Copenhagen, Copenhagen, Denmark
b Research Center for Quality in Medicine Use, Copenhagen, Denmark
d Department of Clinical Sciences, Faculty of Medicine, Lund University, Malmö, Sweden
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS. The study population included all students in grades 5, 7, and 9 (mean ages: 11.6, 13.6, and 15.6 years, respectively) in a random sample of schools in Denmark (participation rate: 88.5%; N = 5205). The students reported health problems, medicine use, bullying, and a range of psychosocial conditions in an anonymous standardized questionnaire. The outcome measure was self-reported medicine use for headache, stomachache, difficulties in getting to sleep, and nervousness. The determinant was frequency of exposure to bullying, measured with 1 item.
RESULTS. In multivariate models adjusted for age and social class, we found that adolescent victims of bullying used medicine for pains and psychological problems more often than did adolescents who were not bullied. The increased odds of using medicine were not explained by the higher prevalence of symptoms among the bullied children.
CONCLUSIONS. We found victimization from bullying to be associated with medicine use, even when we controlled for the higher prevalence of symptoms among bullied victims. The medications that adolescents use can have adverse effects, in addition to the potentially health-damaging effects of bullying. Policy makers, health care professionals, and school staff should be aware that the adolescent victims of bullying are prone to excess use of medicine, and preventive actions should be taken to decrease the level of bullying as well as the use of medicine among adolescents.
Key Words: bullying • medicine use • adolescents • population-based study
Abbreviations: HBSC—Health Behavior in School-aged Children • MOR—median odds ratio • OR—odds ratio • CI—confidence interval
Using medicine to alleviate common ailments is a frequent behavior among adolescents.1,2 The use of pain-relieving medicine and psychotropic drugs among adolescents is an important issue for health care professionals, because the substances are potentially toxic and may have harmful adverse effects.3
The patterns of medicine use according to gender and age are consistent across countries, despite considerable national differences in the prevalence of medication use. In general, girls seem to use medicines for headaches and stomachaches more often than do boys, and both prevalence and gender differences in use increase from 11 years to 15 years of age. The prevalence of medicine use for psychological problems is greater among younger boys, but the prevalence decreases with age for both genders. By the age of 15 years, the prevalence of the use of these medications is also higher among girls.1
One important stressor in adolescent life is frequent bullying,4,5 which, in turn, is clearly related to aches and psychological problems.5–18 Medicine use is a behavior that, over and above formal therapeutic indications, may reflect a general coping strategy to overcome daily stressors. Therefore, frequent bullying may increase the use of pain-relieving medicine and psychotropic drugs.
It is known that health and several health-related behaviors in adults seem to be rooted in childhood and adolescence. With a long-term perspective, early use of medicines to alleviate pain and psychological problems may prevent the learning of more appropriate ways of coping. In addition to the health-damaging effects of bullying, this practice adds the potential risk of lifelong medicine use. Despite the apparent relevance, however, we were unable to identify any previous studies dealing with the relationship between bullying and medicine use among adolescents. This study examined the association between victimization from bullying and medicine use for treatment of pain and psychological problems in a representative sample of adolescents.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The participation of the schools in the survey relied on the previous informed consent of the headmaster, the school board, and the board of pupils at each school. Fifty-five of the 68 randomly sampled schools agreed to participate in the study. In total, 99% of the pupils present on the day of data collection participated in the survey, corresponding to 88.5% of the pupils enrolled in the relevant classes (N = 5205). Because the study was conducted anonymously, we were unable to analyze the characteristics of nonparticipants.
Data Collection
Data were collected in the classroom by means of a carefully tested questionnaire, with standardized instructions from the teacher. The pupils returned the completed questionnaires to their teacher in sealed envelopes to protect anonymity, and no school staff members had access to the completed questionnaires.
Assessment of Variables
The dependent variable "medicine use" was measured with the following item: "During the past month, did you take any tablets or medicine for (1) headache, (2) stomachache, (3) difficulties in getting to sleep, or (4) nervousness?" The responses were categorized as (1) no or (2) yes, once or yes, several times. The items have been tested to show high external validity, and qualitative analyses have shown good face and content validity of the items.19
The independent variable "bullying" was measured with the following item: "How often have you been bullied in school this term?" The responses were categorized into 3 levels, namely, (1) not at all, (2) once or twice/a few times, and (3) once per week/several times per week. Previous studies have demonstrated a high degree of validity in the applied measurement of bullying.20,21
Family social class was derived from 2 items on father's and mother's occupations. Children's reports of their parents’ occupations have been shown to be valid.22 In accordance with the standards of the Danish National Institute of Social Research,23 pupils’ responses were coded from I (high) to V (low), with VI indicating parents receiving social benefits. Approximately 12% of the pupils did not provide sufficient information to allow coding of social class, and we included lack of information on family social class as a specific category in the analyses. Each pupil was coded in accordance with the parent with the highest rank. The pupils were categorized into 4 groups, namely, (1) I/II, (2) III/IV, (3) V/VI, and (4) missing data on social class.
Symptom prevalence was measured with 4 items from the validated HBSC Symptom Check List,24 as follows: "During the past 6 months, how often have you experienced (1) headache, (2) stomachache, (3) difficulties in getting to sleep, or (4) nervousness?" The responses were recoded into 3 levels, namely, (1) almost daily/more than once per week, (2) about once per week/about once per month, or (3) less often or never. Category 3 was used as reference in the comparisons.
Statistical Procedures
We used SAS 8.2 software (SAS Institute, Cary, NC) for all analyses. We performed preliminary separate logistic regression analyses for each kind of medicine use, with bullying as the independent variable, and supplemented each of these analyses with multilevel logistic regression analyses (Tables 1 and 2). Analyses including both genders were performed to test the interaction of gender with the association between bullying and medicine use. The interaction term was not significant for medicine use to alleviate headache, stomachache, or nervousness (P = .111, .396, and .205, respectively), whereas the interaction was significant (P = .046) for medicine for sleeping difficulties. Therefore, all analyses were conducted separately for boys and girls, and we present the results according to gender specifically (Tables 1 and 2).
|
|
We used the median OR (MOR) to express school and school class variance in medicine use.27–29 The MOR quantifies differences (ie, variance 
2) between schools and between school classes by comparing 2 children with the same covariates but from 2 randomly chosen, different schools or school classes. This procedure yields a distribution of ORs, with 1 OR for each comparison pair. The MOR is the median of this distribution of pairwise ORs. That is, the MOR expresses how much (in median) the individual probability of using a certain medicine would increase if a child moved to another school or school class with higher use of medicine. The MOR measure is always 
1. If the MOR is 1, then there are no differences in medicine use between schools (no second-level variation). If there are considerable school differences, then the MOR is large. The measure is directly comparable to fixed-effects ORs, which makes quantification of school variance easier to appreciate in terms of the familiar ORs.
In our investigation, we used the school class-level variance to compute the school class MOR. The interpretation of this measure would be how much (in median) the individual probability of using a certain medicine would increase if an individual within a school moved to another school class with higher use of medicine. An equivalent interpretation is valid for MOR at the school level. Obviously, when a child changes schools, he or she also changes school classes. Therefore, we could calculate the school-level MOR by using the sum of the school and school class variances. However, using only the school-level variance allows us to obtain information about the relative importance of the school versus the school class environment for understanding medication use. In the present investigation, we were mostly interested in measuring the association between bullying and medication use, rather than analyzing the variance of medicine use between schools and between school classes. Therefore, we used multilevel regression analyses only to account for the hierarchical structure of the data, and we did not attempt to explain statistically the school- and school class-level variances.30
Sensitivity analyses were conducted (data not shown), and our results were robust with respect to changes in the definition of the outcome medicine use. Patterns of findings were similar whether they were based on comparisons of the extreme categories of medicine use (yes, several times, versus no) or the categories of yes, once/yes, several times, versus no.
Participants who failed to provide full information on exposure to bullying, prevalence of symptoms, or medicine use were excluded from the analyses (156 boys and 165 girls). There were very few missing data regarding exposure to bullying (0.7%), medicine use (between 1.7% for use of medicines for headache and 4.5% for use of medicines for nervousness and sleep difficulties), and symptom prevalence (between 1.9% for headache and 4.0% for nervousness), and the final analyses included 2425 boys and 2459 girls.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
For boys, the highest monthly symptom prevalence was found for psychological symptoms (52.3% for nervousness and 50.8% for difficulties in getting to sleep), compared with 30.9% of boys having stomachache within the past month. For girls, the prevalence of symptoms was approximately the same for all 4 symptoms (ranging from 55.9% for difficulties in getting to sleep to 60.0% for headache). The prevalence of symptoms was higher among girls than among boys, but gender differences were small for the 2 psychological symptoms (nervousness and difficulties in getting to sleep).
Medicine use for all 4 symptoms was more than twice as high among frequently bullied boys, compared with nonbullied boys (ORs between 2.26 [95% CI: 1.7–3.0] for medicine for headache and 4.43 [95% CI: 3.0–6.5] for medicine for nervousness) (Table 1, model 1). Some of this increased risk was accounted for when analyses were adjusted for age group and social class (model 2). However, even with control for these variables and for the higher symptom prevalence among bullied boys, medicine use was higher among boys who were bullied frequently than among nonbullied boys (model 3: ORs between 1.45 [95% CI: 0.9–2.4] for medicine for stomachache and 2.68 [95% CI: 1.7–4.2] for medicine for sleeping difficulties). Also among girls, we found exposure to bullying to be associated with medicine use, except for medicine use for stomachache, but the estimates for the association were lower than those for boys (ORs between 1.11 [95% CI: 0.8–1.6] for medicine for stomachache and 4.10 [95% CI: 2.7–6.4] for medicine for nervousness) (Table 2, model 1).
The results were attenuated for medicine for sleep difficulties and nervousness when we adjusted for age group and social class (model 2). Even when we took the higher prevalence of symptoms among frequently bullied girls into account (model 3), there was greater frequency of medicine use among bullied girls (ORs between 1.13 [95% CI: 0.8–1.7] for medicine for stomachache and 2.40 [95% CI: 1.5–3.8] for medicine for nervousness).
The MOR measure indicated considerable variance between schools and between school classes, especially for use of medicine for sleep difficulties and medicine for nervousness. In general, the variance was not affected greatly when the individual characteristics of the children were considered. The school class environment seemed to play a more relevant role (ie, higher MOR) than the school environment.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The most prevalent type of medicine use in this age group was use of medicine for headache, and our results showed 40% to 70% increased odds of use of this type of medicine among bullied children. The ORs for the effect of bullying on medicine use were higher for the less-prevalent types of medicine use, such as use of medicine for difficulties in getting to sleep, for which bullied children had ORs of 2.40, compared with nonbullied children. However, whereas only 3% to 4% of the children used medicine for difficulties in getting to sleep, we are most concerned with the somewhat low odds of use of medicine for headache among the bullied children (ORs of 1.73 for boys and 1.43 for girls), because this increased risk affects the greatest number of children. More than one third of the boys and almost one half of the girls used medicine for headaches, and our analyses showed that a large number of those children were bullied sometimes or often.
Substantial proportions of adolescents use medicines for common health problems. Internationally, our own studies in 28 countries found prevalence rates for use of medicine for headache ranging between 21.1% (Slovak Republic) and 49.9% (Scotland) among boys and between 28.3% (Greenland) and 65.9% (United States) among girls.1 We found only one other study that addressed the issue of bullying and medicine use. Van Cleave and Davis31 studied bullying and victimization among children with special health care needs, among them a group of children using prescribed medicines. Those authors found the odds of being bullied among those children to be 1.09, compared with children without special health care needs.
It is important to identify factors that influence medicine use in this age group and to understand why bullying is associated with medicine intake. Our study is cross-sectional and does not reveal how bullying may contribute to medicine use, but there are several possible explanations for the association.
First, bully victimization has severe consequences for the health and well-being of the children involved. Studies have shown associations between bullying and a plethora of health-related factors, including physical symptoms, overweight, psychological symptoms, bedwetting, poor thriving, low self-esteem, social marginalization, and even suicidal ideation.5–18 Although most of those studies were cross-sectional, the causal direction for a number of those associations has been confirmed through longitudinal studies. For instance, longitudinal studies have confirmed that bullying leads to increased risk of physical symptoms, whereas the direction of causality is less clear for some psychological factors. Some studies have shown that vulnerable children are at increased risk of being bullied,13,14,31–33 whereas other studies have found that victimization leads to the onset of emotional problems and that emotional problems do not increase the risk of being bullied.12 Part of the explanation for these differences may be the difference in the ages of the studied populations. Studies of young children (5–9 years of age) found vulnerable children to be at increased risk of bullying, whereas, in their study of adolescents (13–14 years of age), Bond et al12 found bullying to be the cause and emotional problems to be the effect. The differences in results may be a consequence of different mechanisms operating at different ages of the children or may reflect loops of effects between bullying and psychological distress. Our cross-sectional data leave the same question of directionality. We think that the increased pressure on children who are bullied may increase their risk of using any means of coping with the distress and that use of medicine may be one way of trying to relieve the strain.
Second, we know from other studies that children with chronic diseases are at greater risk of being bullied.17,31 It is likely that vulnerable children (for example, children with chronic diseases) use more medicine than other children and that the increased level of bullying among medicine users is partly explained by this mechanism. We were unable to control for chronic diseases in our study.
Third, we observed high MORs across schools and school classes in our multilevel analyses, which indicated that medicine use was sensitive to characteristics of the school and the school class. Behavioral patterns of medicine use may be communicated from one child to another. This mechanism is well known for adolescent health behaviors such as smoking and physical activity, and the same mechanism may explain part of the school and class effects that we found for medicine use for headache, sleeping difficulties, and nervousness. Interestingly, we found almost no class effect in medicine use for stomachache, which may indicate that the individual-level factors (ie, the strong age and symptom effects, especially among girls) account for the school and class variations for this type of medicine use behavior. Rules at the school level, such as the practices of the school staff regarding whether to distribute medicines to children on request, may influence the level of medicine intake at the school. The large differences in mean levels of bullying among schools and classes may represent another possible explanation for the school- and class-level variances in medicine use,11 because children from schools with a high prevalence of bullying may use more medicine to cope with the social strain of the school environment, compared with children from schools with a more pleasant environment.
Fourth, Andersen et al34 found medicine use to cluster with smoking and drunkenness and suggested that medicine use is part of a behavioral pattern of misuse. In our study, medicine use for headache showed a positive association with age, and it is likely that this type of medicine use was related to a pattern of risk behavior clustering among older children; however, the negative associations between other types of medicine use and age make it unlikely that they are involved in this suggested mechanism. To prevent excess medication use in adolescence, it is important to know when adolescents use medications for psychological distress and when they take them for other purposes, such as recreational use.
Longitudinal research is needed to study the causal mechanisms underlying adolescents’ use of medication. Research in this area is challenged by the developmental changes that occur during the adolescent life period and should address not only the reasons for medication use but also any changes in causes over time.
The study was based on a large, representative study population with a high response rate. Validation studies showed the included variables to be valid for age-equivalent populations.19–22,24 Furthermore, although the response rate was high, the nonparticipants (11.5%) might have higher prevalence rates of both exposure to bullying and medicine use. If this was the case, then we likely underestimated the associations between exposure to bullying and medicine use.
Medicine use among adolescents is a global public health issue that needs to be addressed through research and public health policies. Bush et al35 made an important start by developing guidelines for teaching children and adolescents about medicine. These guidelines have been adapted by the International Pharmaceutical Federation,36 and their implementation should be brought into focus.
The results of our study indicate that policymakers, health care professionals, and school staff members should be aware that the adolescent victims of bullying are prone to excess use of medicine. Several intervention studies have demonstrated that pedagogical intervention programs help reduce bullying at school.37–39 Our study gives a second reason to take steps to prevent bullying, that is, to avoid harmful patterns of excess medicine use among adolescents.
|
ACKNOWLEDGMENTS |
|---|
We thank Mogens Trab Damsgaard, Institute of Public Health, University of Copenhagen, for his support in discussions on analytical strategy. We also thank Charlotte Ørsted Hougaard for her help with the multilevel analyses presented in Tables 1 and 2. We thank Birgit Pallesen and Karen Steenhard for their help with language and the reviewers for their thorough reviews and helpful comments.
|
FOOTNOTES |
|---|
Address correspondence to Pernille Due, MD, Department of Social Medicine, Institute of Public Health, Faculty of Health Sciences, University of Copenhagen, Øster Farimagsgade 5, PO Box 2099, DK-1014 Copenhagen, Denmark. E-mail: p.due{at}pubhealth.ku.dk
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. L. Hansen, E. H. Hansen, and B. E. Holstein Using Analgesics as Tools: Young Women's Treatment for Headache Qual Health Res, February 1, 2008; 18(2): 234 - 243. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Bejerot and M. Humble Relevance of Motor Skill Problems in Victims of Bullying Pediatrics, November 1, 2007; 120(5): 1226 - 1227. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
