TURKISH ARCHIVES of PEDIATRICS Original Article TÜRK PEDİATRİ ARŞİVİ Health behaviors in high school students in İzmir, Turkey Sabanur Çavdar1, Esin Çetinkaya Sümer1, Kayı Eliaçık2, Aslıhan Arslan2, Begüm Koyun2, Nurhan Korkmaz2, Ali Kanık2, Ethem Erginöz2, Oya Ercan3, Müjgan Alikaşifoğlu3 Department of Public Health, İstanbul University Cerrahpaşa School of Medicine, İstanbul, Turkey Clinic of Pediatrics, İzmir Tepecik Training and Research Hospital, İstanbul, Turkey 3 Department of Pediatrics, Division of Adolescent Health, İstanbul University Cerrahpaşa School of Medicine, İstanbul, Turkey 1 2 Abstract Aim: The aim of this cross-sectional study was to determine the prevalence of the self reported health behaviors and differences in these behaviors by gender and grades in high school adolescents in İzmir, Turkey. Material and Methods: A stratiied cluster sampling procedure was used for this cross-sectional study. The study sample included 2 296 students attending 22 high schools in Izmir. As a data collection instrument, some questions from the Health Behavior in School-aged Children Study 2009/2010 questionnnaire and questions which were developed by the researchers to understand behaviors of internet use in adolescents were used. Chi-square tests and Cramer’s V statistics were used for statistical analyses. Results: Among the high school students, 33.8% experimented smoking, 26.3% smoked cigarette during the 30 days before the survey, 14.9% smoked cigarette regularly during the 30 days before the survey, 54.1% experimented drinking alcohol, 38.4% drunk alcohol during the 30 days before the survey, 31.6% got drunk, 10.9% were adequately physically active, 59.9% watched TV for a long period of time, 72.8% used internet for a long period of time, 48.1% ate breakfast regularly, 36.2% ate adequate amount of fruit, 14.1% ate adequate amount of vegetable, 31.3% ate candies and chocolate very often, 18% drunk soft drink very often, 30.3% were bullied, 29.9% bullied others and 41% involved in a physical ight. Conclusions: These results showed that “Adolescent friendly health services” should be generalized all over Turkey, physicians should evaluate each adolescent for his/her health behaviors in each visit and implementation of prevention programs which adopt a health promotion perspective is necessary begining from the elementary school. (Turk Pediatri Ars 2016; 51: 22-34) Keywords: Alcohol, leisure time activities, adolescent, health behavior, violence, tobacco, eating behaviors Introduction Adolescent is a Latin term which means “maturation by development” and expresses the period of transition from childhood to adulthood. This period which is also called puberty is a stressful and luctuant period for the intividual and his/her connections during which many changes in terms of physical, cognitive, psychological and social aspects are experienced in association (1). Problems which may occur in one of the maturation processes may prevent healthy development in adolescents by affecting the other processes and cause to diseases which are also carried to adulthood (1, 2). The most important characteristic which is observed typically in adolescence is tendency to take risk. This tendency may cause to adoption of negative behaviors and attitudes related with health including use of tobacco, alcohol and substance which constitutes the ba- 22 sis for non-infectious diseases, unsafe sexual relation, consumption of nutrients with low nutritional value and high fat and calorie content and sedantary life-style or may render these behaviors and attitudes a life-style (2-5). These attitutes and behaviors may both affect the current health status of adolescents and cause to morbidity and mortality in adulthood (2, 3). For example, it is predicted that one ifth of adolescents aged 13-15 years smoke at the present time and approximately half of the individuals who smoke in adolescence will continue this behavior for at least 15 years more (2). The most important characteristic of the behaviors observed in adolescents which affect health adversely is that they are preventable to a large extent (5). Therefore, adolescence is considered an opportunity for preventing carriage of the behaviors which affect health adversely to adulthood (3). In addition, detailed information about health-related behaviors and diseases of adolescents is necessary in terms of accurate development of inter- Address for Correspondence: Oya Ercan E-mail: oyaercan@istanbul.edu.tr Received: 14.09.2015 Accepted: 13.11.2015 ©Copyright 2016 by Turkish Pediatric Association - Available online at www.turkpediatriarsivi.com DOI: 10.5152/TurkPediatriArs.2016.3389 Turk Pediatri Ars 2016; 51: 22-34 vention programs for both adolescence and childhood (3, 4). Another important point related with health-related behaviors in adolescents is that adolescents tend to exhibit multiple risky behaviors in association (6). The characteristic of the behaviors which may affect health adversely of clustering poses a risk of occurence of other behaviors in presence of one behavior, while it is an opportunity for interventions directed to one behavior to also prevent the others (6, 7). In our country in which the 15-19 age group constitutes 8.1% of the population, studies should be conducted in order obtain information about the health states of the young ones, their health-related behaviors and social and personal factors which lead to occurence of these behaviors, to evaluate the changes in these behaviors in time, to determine the priorities for programs which would be prepared to protect and develop the youngster’s health and to observe the effects of the interventions performed (8). In this cross-sectional study, it was aimed to determine the frequency of self-reported health-related behaviors of the young individuals who attended high school in Izmir, the distribution of these behaviors by gender and grade and the primary areas in adolescent health programs. Çavdar et al. Health behaviors in adolescents private area high schools were excluded from the study. The students in the high schools which were excluded constituted 69% of the population. Conclusively, the population was established as 193 370 students including 86 275 students studying at general high schools and 107 095 students studying at vocational high schools. The sample was calculated with 95% conidence interval, 80% power and 2% margin of error. Each stratum and each class was represented in the sample with its ratio inside the population and the sample size was determined to be 2 062 students with corrections including stratum, classroom and rounding. The design effect was calculated to be 1.28. It was thought that the high school students obtained from this sample would represent the young individuals attending high school in the province of İzmir. The classes for which the questionnaire form would be applied were determined by drawing lots at the level of district, school and class. It was taken care of that at least one general high school and one vocational high school were included in each district. The questionnaire form was applied to studets who were present on the day of application and who volunteered to participate in the study in the classes which were chosen by lot. The application was performed in a total of 101 classes in 22 schools. Material and Methods This study constituted a part of a cross-sectional survey based on evaluation of a questionnaire form read and responded by students. This questionnaire form was designed to determine self-reported health states, health-related behaviors and the factors related with these behaviors in adolescents attending high school in Izmir. Two thousand two hundred ninety six students who attended 22 high schools in Izmir were included in the study. Ethics committee approval was obtained from the Tepecik Education and Research Hospital Ethics Committee (meeting dated 09.30.2014,decision number: 10) and the necessary approval was obtained from the Izmir provincial directorate for national education (article dated 12.10.2014 with number 6297744). Sample selection Stratiied cluster sampling method was used for selection of the sample. The sampling unit was accepted to be “school”. The high schools were considered as two strata including general high schools and vocational high schools according to the type of education. The 2013/2014 academic year of İzmir Provincial Directorate for National Education data were used to frame the study population. Private high schools and Data collection tools used in the study A questionnaire form composed of two parts was used as data collection tool in this study. The irst part consisted of 37 questions. These questions were obtained from the international questionnaire form belonging to an international survey named “Health Behaviour in School-aged Children Study-2009/2010 (HBSC 2009/2010)” (9). The questions related with gender, class, age, eating habits, dieting, physical activity, use of tobacco and alcohol, state of well-being, body weight, height, health complaints, body image, bullying, socioeconomical status, school life and family structure were selected from this questionnaire form. Since the conducters of the Turkey group of the “Health Behaviour in School-aged Children Study” were included in this study, a seperate approval from the international coordinatorship of the HBSC study was not necessary for use of the questions. The second part consisted of a questionnaire form which investigated the type of internet access of the students, the period of usage of internet, which activities in the internet they participated in and with what frequency they participated in these activities, which was established as a result of literature screening per23 Turk Pediatri Ars 2016; 51: 22-34 Çavdar et al. Health behaviors in adolescents formed by the investigators and face to face interviews with the young individuals and which took its inal shape as a result of a pilot study. In this study, only the time of usage of internet was included in the evaluation among these questions. last 30 days: having consumed an alcoholic drink in the last 30 days albeit rarely. Having got drunk: having got drunk at least for one time for a life time. Having got drunk in the last 30 days: having got drunk in the last 30 days for at least one time. Application Pilot application: The pilot study for the questions which measured internet use was conducted with 30 adolescents aged between 15 and 18 years who presented to Cerrahpaşa School of Medicine, Department of Pediatrics, Adolescent Outpatient Clinic because of different causes. These adolescents were primarily asked to ill in the questionnaire form alone. Afterwards, the questions were discussed with each of them using a semi-structured interview technique. Changes in the espressions were made in order to clarify some questions after the pilot study. Spare time activities: Suficient physical activity: performing moderate or severe physical activity for at least 1 hour each day of the week. Watching TV for long periods of time: watching TV for at least 2 hours a day. Using the internet for long periods of time: using the internet for at least 2 hours a day. General application: The questionnaire form was applied in a class period of 45-60 minutes in all classes. Two investigators and the class teacher were present during the application in the classsroom. Before the application, the students were informed about the content of the study and that the participation was voluntary by reading a standard instruction. None of the students refused to participate in the study. The process of application of the questionnaire forms was completed in December 2014-January 2015. Data input and conirmation Data input was performed in January-February 2015 according to the data input guideline. Conirmation of the day input was completed by 2 members of the study team who worked in Cerrahpaşa Medical Faculty, Department of Public Health in accordance with the study protocol by reentering one of each 5 questionnaire forms. Data conirmation was completed in April 2915. Deinition of behaviors In this study, the answers were primarily transformed into dichotomous variables for all behaviors investigated for statistical evaluation. Consumption of tobacco behaviors: Experimenting smoking: having smoked more than one breath in a life time. Having smoked in the last 30 days: having smoked at least one time in the last 30 days. Regular smoking in the last 30 days: having smoked 20 or more cigarettes in the last 30 days. Experimenting nargileh: having smoked nargileh at least one time for a life time. Consumption of alcoholic drink: Experimenting alcohol: having consumed alcohol more than a small amount for a life time. Having consumed alcohol in the 24 Eating behaviors: Having regular breakfast: having breakfast for 7 days a week. Having regular breakfast on weekdays: having breakfast on each weekday. Consuming adequate amounts of fruit: eating fruit each day at least for once. Consuming adequate amounts of vegetable: eating vegetables each day at least for once. Consuming candy or chocolate frequently: consuming candy or chocolate at least for once every day. Consuming soft drinks frequently: consuming cola or other soft drinks for at least once every day. Dieting: Being on a diet at that moment in order to lose weight. Behaviors related with violence: Having been bullied: having been bullied for at least once in the last 1-2 months at school. Having been bullied frequently: having been bullied for at least twice in the last 1-2 months at school. Having bullied others: having bullied other students for at least once in the last 1-2 months at school. Having bullied others frequently: having bullied other students for at least twice in the last 1-2 months at school. Involving in a physical ight: having involved in a physical ight for at least once in the last 12 months. Having been involved in a physical ight frequently: having involved in a physical ight for at least 3 times in the last 12 months. Statistical analysis The frequencies of the behaviors in the same grade by gender were compared using chi-square test. It was evaluated if the frequencies of behaviors of students with the same gender in different grades varied by testing correlation using Cramer’s V coeficient method. Results Two thousand two hundred ninety six students attending 22 high schools in the province of Izmir were included in the study. While the number of registered students was 2 735 in the grades included in the study, 439 students were absent on the day of application (the rate of absenteesim was 16.5%). Therefore, a total of 2 296 students could be reached. The questionnaire Turk Pediatri Ars 2016; 51: 22-34 Çavdar et al. Health behaviors in adolescents Table 1. Distribution of the study group by gender and grades 9th grade Total Number of students 10th grade Female Male Female 11th grade Male Female n n %a n %a n %a n %a 1986 304 49.2 314 50.8 238 50.4 234 49.6 Male %a n 12th grade 248 52.8 Female n %a 222 47.2 Male %a n %a n 249 58.5 177 41.5 a Percentages of distribution by gender in each grade Table 2. Mean ages of the study group by gender and grades 9th grade Total n Mean age Female 10th grade Male Female 11th grade Male Female 12th grade Male Female Male Mean ±SD Mean ±SD Mean ±SD Mean ±SD Mean ±SD Mean ±SD Mean ±SD Mean ±SD 14.8 ±0.55 14.9 ±0.62 15.8 ±0.59 15.9 ±0.67 16.7 ±0.53 16.9 ±0.64 17.8 ±0.55 17.8 ±0.64 1984 a a Percentages of distribution by gender in each grade Table 3. Behaviors of consumption of tobacco and alcoholic beverages of the students by grades and gender 9th grade Total Female Behaviors a n n b % Male d p n % n % p n Male c % n %c pd 33.2 93 42.3 0.044 100 40.7 94 53.4 0.010 Smoking in the 1 967 237 last 30 days 29.9 <0.001 51 17.1 58 18.8 0.582 52 21.9 62 27.1 0.197 54 22 Smoking regularly 1 967 111 10.8 183 in the last 30 days 19.6 31 22 46 20.1 0.001 28 11.4 53 23.9 <0.001 46 18.5 53 30.1 0.005 Experimenting 1 956 317 30.7 388 nargileh 41.8 <0.001 53 17.7 76 31.9 102 44.3 0.006 88 35.5 100 45.9 0.023 100 41.3 109 62.6 <0.001 Experimenting 1 966 502 48.8 562 alcool 60 <0.001 110 36.9 152 Having consumed 1 968 326 31.7 430 alcohol in the last 30 days 45.7 <0.001 69 23.2 104 33.5 Having got drunk 1 956 261 25.4 357 38.5 <0.001 41 13.7 73 Having got 1 964 131 12.8 218 drunk in the last one month 23.3 <0.001 22 7.4 49 % d 79 34.8 0.343 81 9.3 n Female c 73 30.7 77 25.1 0.027 % Male c 72 23.3 0.897 0.019 n d 71 23.7 10 p Female c 0.031 5 % Male c 36.3 15 n d Experimenting 1 959 325 31.6 338 smoking <0.001 % Female c 12th grade p 280 n Male c 11th grade % 23 n Female b 10th grade 80 36 0.001 80 32.1 80 45.5 0.005 0.003 116 48.7 142 61.2 0.007 133 54.1 137 62.3 0.073 143 57.9 131 74.9 <0.001 0.005 68 28.7 114 49.4 <0.001 94 38.2 109 49.1 0.018 95 38.3 103 58.2 <0.001 23.9 0.001 61 25.6 92 39.8 0.001 73 29.8 90 41.1 0.011 86 35.1 102 59 <0.001 43 13.9 0.009 26 11 57 24.7 <0.001 38 15.4 52 23.5 0.027 45 18.2 66 37.5 <0.001 n=Number of students who answered the relevant question Percentage of having performed the behavior by gender c Percentage of having performed the behavior by gender in each grade a b d Pearson’s chi-square test forms of 25 disabled students, 284 students who gave inconsistent answers to questions related with each other and 1 student who did not state gender were excluded from the evaluation. None of the students refused to participate in the study. Analysis of this study was performed with the data of 1 986 students. The rate of data loss was calculated to be 13.5%. Nine hundred fourty five of the student who participated in this study were male (47.7%) and 1 039 were female (52.3%). The distribution of the students by grades and gender is shown in Table 1. The mean age of the study group was 16.18±1.26. The mean ages by grades and gender are shown in Table 2. 25 Turk Pediatri Ars 2016; 51: 22-34 Çavdar et al. Health behaviors in adolescents Table 4. The state of having performed the behaviors of having experimented smoking. alcohol and nargileh or having got drunk for the irst time at the age of 11 years or younger by age groups Age groups Total 15 years and below a b 16 years 17 years b 18 years and above b 11 years or younger n % n % n % n % n %b pc Experimeting smoking 44 6.6 21 8.8 8 4.7 12 6.3 3 3.3 0.217 Experimenting nargileh 30 4.3 14 5.9 9 4.8 7 3.3 0 0 0.087 Experimenting alcohol 133 12.5 60 14.5 31 11.7 28 9.7 14 12.2 0.294 Having got drunk 33 5.3 12 5.9 8 4.8 9 4.7 4 4.5 0.931 a The percentage among the students who reported that they performed the behavior b The percentage of having performed the behavior at the age of 11 years or younger in that age group among the students who performed the behavior c Pearson’s chi-square test (comparison of the rates of having experimented the behavior before the age of 11 years according to age groups) Table 5. The states of having smoked. having consumed alcohol and having got drunk in the last 30 days by the age at which this behavior was performed fort he irst time in the students who have experimented smoking. alcohol and nargileh and who have got drunk Behaviors Having smoked in the last 30 days Having consumed alcohol in the last 30 days Having got drunk in the last 30 days n %a OR %95 CI n %a OR %95 CI n %a OR %95 CI Experimenting <=11 years 32 72.7 0.808 0.406-1.609 26 59.1 0.729 0.391-1.360 16 36.4 1.045 0.553-1.973 smoking >11 years 485 76.7 420 66.5 222 35.4 Experimenting <=11 years 18 60.0 23 79.3 15 50.0 1.804 0.867-3.753 nargileh >11 years 382 55.4 456 66.1 Experimenting <=11 years 67 51.1 113 85.0 2.143 1.482-3.099 alcohol >11 years 383 41.0 643 68.6 Experimenting <=11 years 19 57.6 32 97.0 3.220 1.376-7.532 alcohol >11 years 52.8 480 79.5 317 1.209 0.574-2.549 1.509 1.046-2.177 1.212 0.596-2.461 1.967 0.790-4.898 2.592 1.580-4.253 8.267 1.119-61.090 245 35.7 64 48.5 285 30.5 26 78.8 323 53.6 The percentage of having smoked. having consumed alcohol and ahving got drunk in the last 30 days a Behaviors of consuming tobacco The distribution of the students by grades and gender is shown in Table 3. Experimenting smoking: six hundred sixthy three (33.8%) of the students reported that they experimented smoking. This rate was 31.6% in the girls (n=325) and 36.3% in the boys (n=338). The difference was signiicant (p=0.031). This behavior was observed more frequently as the grade increased both in the girls and boys (Cramer’s V=0.133, p<0.001; Cramer’s V=0.228, p<0.001, respectively). Six point six percent of the students (n=44) who reported that they experimented smoking experimented smoking at the age of 11 years or before. When the students were examined after dividing them into age groups, the rate of experimenting smoking at the age of 11 years and before did not change (Table 4). Having experimented smoking at the age of 11 years and younger did not increase the risk of having smoked in the last 30 days, having 26 consumed alcohol in the last 30 days and having got drunk in the last 30 days (Table 5). Having smoked in the last 30 days: 517 of the students (26.3%) reported that they smoked at least once in the last 30 days. This rate was 23.0% in the girls (n=237) and 29.9% in the boys (n=280). The difference was statistically signiicant (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.132, p<0.001; Cramer’s V=0.215, p<0.001, respectively). Having smoked regularly in the last 30 days: 294 of the students (14.9%) reported that they smoked regularly in the last 30 days. This rate was 10.8% (n=111) in the girls and 19.6% (n=183) in the boys. The difference was statistically signiicant (p<0.001). This behavior was observed more frequentky as the grade got older both in the girls and boys (Cramer’s V=0.160, p<0.001; Cramer’s V=0.188, p<0.001, respectively). Experimenting nargileh: 705 (36%) of the students reported that they experimented smoking nargileh. This rate was 30.9% in the girls (n=317) and 41.8% in the boys Turk Pediatri Ars 2016; 51: 22-34 Table 6. Çavdar et al. Health behaviors in adolescents Spare time activities and eating behaviors of the students by grades and gender 9th grade Total Female Male n %b n Suficient 1 970 72 physical activity 7 142 Female Male Female Male %c n 15.1 <0.001 25 8.4 56 18 0.001 19 8 32 13.8 0.043 Watchişng 1 960 576 56.1 599 television for long periods of time 64.2 <0.001 202 67.6 208 68 0.913 137 58.5 155 67.4 0.049 138 55.6 141 64.1 0.063 99 40.2 95 53.7 0.006 Using the 1 963 735 71.4 695 internet for long periods of time 74.4 0.138 216 71.8 217 70.2 0.676 181 77.4 182 78.8 0.708 189 76.5 165 75.3 0.767 149 60.3 131 74.9 0.002 Having breakfast 1 938 446 43.8 487 regularly 53 <0.001 131 44.3 184 60.5 <0.001 97 41.3 119 53.4 0.010 112 46.3 104 47.9 0.724 106 43.1 80 45.7 0.593 Having breakfast 1 964 495 48.1 543 regularly on weekdays 58.1 <0.001 141 Consuming 1 973 406 39.3 308 suficient amounts of fruit 32.7 <0.001 133 44.2 118 37.8 0.109 102 43 81 35.1 0.077 82 33.3 69 31.2 0.626 89 35.9 40 22.6 0.003 Consuming suficient amounts of vegetable 111 11.9 15 28 12.3 0.412 36 14.6 25 11.4 0.305 41 16.6 20 11.3 0.125 Consuming candy 1 954 396 38.6 216 or chocholate frequently 23.3 22.2 <0.001 103 43.6 55 24.3 <0.001 83 33.9 56 25.6 0.051 89 Consuming carbonated beverages frequently 1 955 144 22.4 <0.001 49 16.5 60 19.5 0.331 35 14.8 64 28.2 <0.001 34 13.9 51 23.2 0.010 26 10.5 33 18.8 0.016 Dieting 1 980 109 10.5 48 5.1 16 26 10.9 11 11 14 208 <0.001 121 40.5 <0.001 34 11.3 n %c n pd 9 3.6 34 15.5 <0.001 19 n %c Male n 52 17.4 %c Female pd 0.01 pd Male %c 47 %c Female n 16 pd 12th grade %c 1 954 164 %b 11th grade n Behaviors na 10th grade n %c pd 7.6 20 11.3 0.196 198 63.9 <0.001 103 43.3 132 58.1 0.001 119 48.8 117 53.2 0.343 132 53.2 96 54.2 0.837 38 12.4 0.083 68 5.1 0.005 35 4.7 0.012 27 7 3.2 0.001 22 36 37 21 0.001 8.9 14 7.9 0.726 n=Number of students who answered the relevant question Percentage of having performed the behavior by gender c Percentage of having performed the behavior by gender in each grade d Pearson’s chi-square test a b (n=388) which was signiicantly higher compared to the girls (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.195, p<0.001; Cramer’s V=0.271, p<0.001, respectively). 4.3% of the students (n=30) who reported that they experimented smoking nargileh experimented nargileh for the irst time at the age of 11 years or younger. When the students were examined by age groups, the rate of experimenting nargileh at the age of 11 years and younger did not change (Table 4). Having experimented smoking nargileh at the age of 11 years and younger did not increase the risk of having smoked in the last 30 days, having consumed alcohol in the last 30 days and having got drunk in the last 30 days (Table 5). Behaviors of consuming alcoholic drinks The distribution of the behaviors of drinking alcoholic drinks by grades and gender is shown in Table 3. Experimenting alcohol: 1 064 of the students (54.1%) reported that they experimented alcohol. This rate was 48.8% in the girls (n=502) and 60.0% in the boys (n=562). The difference was statistically signiicant (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.164, p<0.001; Cramer’s V=0.185, p<0.001, respectively). 12.5% of the students who stated that they experimented alcohol (n=133) drunk their irst alcoholic drink at the age of 11 years or younger. When the students were examined by age groups, the rate of 27 Çavdar et al. Health behaviors in adolescents having experimented alcohol at the age of 11 years and younger did not change (Table 4). Having experimented alcohol at the age of 11 years and younger increased the rate of having smoked in the last 30 days by 1.51fold, the rate of having consumed alcohol in the last 30 days by 2.59-fold and the rate of having got drunk in the last 30 days by 2.14-fold (Table 5). Having consumed alcohol in the last 30 days: 756 of the students (38.4%) reported that they consumed alcohol in the last 30 days, albeit rarely. This rate was 31.7% in the girls (n=326) and 45.7% in the boys (n=430). The difference was statistically signiicant (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.142, p<0.001; Cramer’s V=0.184, p<0.001, respectively). Having got drunk: 618 of the students (31.6%) reported that they had gotten drunk at least for once. This rate was 25.4% in the girls (n=261) and 38.5% in the boys (n=357). The difference was statistically signiicant (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.188, p<0.001; Cramer’s V=0.252, p<0.001, respectively). 5.3% of the students (n=33) who reported that they got drunk had gotten drunk at the age of 11 years or younger. When the students were examined by age groups, the rate of having got drunk for the irst time at the age of 11 years and younger did not change (Table 4). Having got drunk for the irst time at the age of 11 years and younger did not increase the risk of smoking in the last 30 days, whereas it increased the rate of having consumed alcohol in the last 30 days by 8.27-fold and the rate of having got drunk in the last 30 days by 3.22-fold (Table 5). Having got drunk in the last 30 days: 349 of the students (17.8%) reported that they got drunk in the last 30 days at least for once. This rate was found to be 12.8% in the girls (n=131) and 23.3% in the boys (n=218). The difference was statistically signiicant (p<0.001). This behavior was observed more frequently as the grade got older both in the girls and boys (Cramer’s V=0.127, p=0.001; Cramer’s V=0.194, p<0.001, respectively). Spare time activities The distribution of spare time activities of the studenst by grades and gender is shown in Table 6. Suficient physical activity: 214 of the students (10.9%) reported that they performed suficient physical activity. This rate was 7% (n=71) in the girls and 15.1% (n=142) in the boys and the difference was statistically signiicant (p<0.001). This behavior did not change as the grade got older both in the girls and in the boys (Cramer’s V=0.075, p=0.125; Cramer’s V=0.068, p=0.224, respectively). Watching TV for a long periods of time: 1 175 (59.9%) of the students reported that they watched TV for long periods of time. This rate was 56.1% in the 28 Turk Pediatri Ars 2016; 51: 22-34 girls (n=576) and 64.2% in the boys (n=599). The difference was statistically signiicant (p<0.001). This behavior was observed less frequently as the grade got older both in the girls and in the boys (Cramer’s V=0.201, p<0.001; Cramer’s V=0.111, p=0.01, repsectively). Using the internet for long periods of time: 1 430 of the students (72.8%) reported that they used the internet for at least 2 hours a day. This rate was 71.4% in the girls (n=735) and 74.4% in the boys (n=695). No signiicant difference was observed between the boys and girls (p=0.138). This behavior was observed less frequently as the grade got older in the girls (Cramer’s V=0.147, p<0.001) and did not change in the boys (Cramer’s V=0.075, p=0.152). Eating behaviors The distribution of eating behaviors of the students by grades and gender are shown in Table 6. Having regular breakfast: 933 of the students (48.1%) reported that they were having regular breakfast. This rate was 43.8% in the girls (n=446) and 53% in the boys (n=487). The difference was statistically significant (p<0.001). This behavior was observed less frequently as the grade got older in the boys (Cramer’s V=0.118, p=0.005) and did not change in the girls (Cramer’s V=0.036, p=0.732). Having a regular breakfast on weekdays: 1038 of the students (52.9%) reported that they were having a regular breakfast on weekdays. This rate was 48.1% in the girls (n=495) and 58.1% in the boys (n=543). The difference was statistically signiicant (p<0.001). This behavior did not show any change as the grade got older both in the girls and in the boys (Cramer’s V=0.070, p=0.171; Cramer’s V=0.09, p=0.057, respectively). Consuming adequate amonts of fruit: 714 of the students (36.2%) reported that they were consuming adequate amounts of fruit. This rate was 39.3% in the girls (n=406) and 32.7% in the boys (n=308). The difference was statistically signiicant (p<0.001). This behavior was observed less frequently as the grade got older both in the girls and in the boys (Cramer’s V=0.095, p=0.026; Cramer’s V=0.116, p=0.005, repsectively). Consuming adequate amounts of vegetables: 275 of the students (14.1%) reported that they were consuming adequate amounts of vegetables. This rate was 16% in the girls (n=164) and 11.9% in the boys (n=111). The difference was statistically signiicant (p=0.01). This behavior did not show any change as the grade got older neither in the girls not in the boys (Cramer’s V=0.032, p=0.786; Cramer’s V=0.016, p=0.973, repsectively). Consuming candy or chocolate frequently: 612 of the students (31.3%) reported that they were consuming candy or chocolate frequently. This rate was 38,6% in the girls (n=396) and 23.3% in the boys (n=216). The difference was statistically signiicant (p<0.001). This behavior did Turk Pediatri Ars 2016; 51: 22-34 Table 7. Çavdar et al. Health behaviors in adolescents Violence-related behaviors of the students by grades and gender 9th grade Total Female Behaviors na n %b Male n %b Female pd n %c 10th grade Male n %c Female pd n %c 11th grade Male n %c Female pd n %c 12th grade Male n %c Female pd n %c Male n %c pd Being bullied 1 962 293 28.4 302 32.4 0.053 104 34.4 121 39.2 0.226 69 29.4 74 32.7 0.433 61 24.7 56 25.5 0.850 59 23.9 51 29 0.239 Being bullied frequently 1 962 99 9.6 136 14.6 0.001 39 12.9 20 8.5 33 14.6 0.040 22 8.9 Bullying others 1 962 254 24.7 333 35.7 0.001 83 27.6 111 35.8 0.029 64 27.2 74 32.7 0.197 65 26.3 79 35.7 0.027 42 17.1 69 39.2 <0.001 Bullying others 1 962 100 9.7 160 frequently 17.1 <0.001 31 10.3 30 12.8 41 18.1 0.110 24 9.7 Involving in a physical ight 1 977 271 26.2 539 57.3 <0.001 108 35.6 195 65.2 <0.001 68 28.8 140 60.6 <0.001 55 22.2 119 53.8 <0.001 40 16.1 85 48 <0.001 Involving in a physical ight frequently 1 977 156 15.1 395 42 13.3 86 38.9 <0.001 17 55 17.8 0.094 51 16.5 0.026 <0.001 64 21.1 138 44.2 <0.001 42 17.7 112 48.5 <0.001 33 26 11.8 0.301 18 35 15.8 0.046 15 7.3 22 12.5 0.071 6.1 33 18.8 <0.001 6.9 59 33.3 <0.001 n=Number of students who answered the relevant question Percentage of having performed the behavior by gender c Percentage of having performed the behavior by gender in each grade d Pearson’s chi-square test a b not show any change as the grade got older neither in the girls not in the boys (Cramer’s V=0.076, p=0.113; Cramer’s V=0.040, p=0.688, respectively). Consuming soft drinks frequently: 352 of the students (18%) reported that they were consuming cola or other sof drinks frequently. This rate was 14% in the girls (n=144) and 22.4% in the boys (n=208). The difference was statistically signiicant (p<0.001). This behavior did not show any change as the grade got older neither in the girls not in the boys (Cramer’s V=0.063, p=0.249; Cramer’s V=0.088, p=0.064, respectively). Being on a diet: 157 of the students (7.9%) reported that they were on a diet. This rate was 10.5% in the girls (n=109) and 5.1% in the boys (n=48). The difference was statistically signiicant (p<0.001). This behavior did not show any change as the grade got older neither in the girls not in the boys (Cramer’s V=0.031, p=0.805; Cramer’s V=0.070, p=0.196, respectively). Violence-related behaviors The distribution of violence-related behaviors of the students by grades and gender is shown in Table 7. Being bullied: 595 of the students (30.3%) reported that they had been bullied at least for once. This rate was 28.4% in the girls (n=293) and 32.4% in the boys (n=302). There was no signiicant difference between genders in terms of the rate of being bullied (p=0.053). Being exposed to this behavior was observed less frequently as the grade got older both in the girls and in the boys (Cramer’s V=0.097, p=0.022; Cramer’s V=0.115, p=0.007, respectively). Being bullied frequently: 235 of the students (12%) reported that they had been bullied at least for 2 times. This rate was 9.6% in the girls (n=99) and 14.6% in the boys (n=136). The difference was statistically signiicant (p=0.001). Being exposed to this behavior did not show any change as the grade got older neither in the girls nor in the boys (Cramer’s V=0.075, p=0.122; Cramer’s V=0.070, p=0.210, respectively). Bullying: 587 of the students (29.9%) reported that they bullied other students at least for once. This rate was 24.7% in the girls (n=254) and 35.7% in the boys (n=333). The difference was statistically signiicant (p=0.001). This behavior was observed less frequently as the grade got older in the girls (Cramer’s V=0.100, p=0.017) and did not change in the boys (Cramer’s V=0.044, p=0.614). Bullying others frequently: 260 of the students (13.3%) reported that they bullied other students at least for 2 times. This rate was 9.7% in the girls (n=100) and 17.1% in the boys (n=160). The difference was statistically signiicant (p<0.001). Being exposed to this behavior did not change as the grade got older neither in the girls nor in the boys (Cramer’s V=0.078, p=0.099; Cramer’s V=0.030, p=0.838, respectively). Being involved in a physical ight: 810 of the students (41%) reported that they were involved in a physical ight at least for once. This rate was 26.2% in the girls (n=271) and 57.3% in the boys (n=539). The difference was statistically signiicant (p<0.001). This behavior was observed less frequently as the grade got older both in the girls and in the boys (Cramer’s V=0.170, p<0.001; Cramer’s V=0.112, p<0.008, respectively). Being involved in a physical ight frequently: 551 of the students (27.9%) reported that were involved in a physical ight for at least 3 times in the last 12 months. This rate was 15.1% in the girls (n=156) and 42% in the boys (n=395). The difference was statistically 29 Çavdar et al. Health behaviors in adolescents signiicant (p<0.001). This behavior was observed less frequently as the grade got older both in the girls and in the boys (Cramer’s V=0.151, p<0.001; Cramer’s V=0.108, p=0.012, respectively). conducted with adolescents who were smokers, it was shown that the age of starting smoking was remembered incorrectly as the age got older and a tendency to state an older age occured (24). Discussion The frequency of experimeting nargileh which has been deined as a new public health problem by the World Health organization (WHO) was found to be 35% in high school students in İzmir in this study (25). This rate is higher compared to the rate reported by the students who participated in our study and markedly lower compared to the frequency of having axperimented nargileh (45,8%) found in the study conducted with students of the 10th grade in 2012 in İstanbul (12). Tobacco consumption behaviors In our study, it was found that 33.8% of high-school students in İzmir experimented smoking, 26.3% smoked cigarette in the last 30 days, 14.9% smoked cigarettes regularly in the last 30 days, this behavior was observed with a signiicantly higher rate in the boys compared to the girls and the frequency of smoking increased signiicantly as the grade got older in both genders. In studies conducted in different provinces with adolescents at different ages in Turkey, the rate of having experimented smoking has been reported to range between 24.4% and 50% (10-14), the rate of smoking at the present time has been reported to range between 6.9% and 24.27% (10-21) and this behaviors have been reported with higher rates in boys (10-17, 20). Although the frequency of having experimented smoking was markedly lower in this study compared to the study conducted in 2013 in the United States of America (USA) with high-school students (41.1%), it was notable that the frequencies of having smoked in the last 30 days and having smoked regularly in the last 30 days were higher (15.7% and 5.6%, respectively) compared to the frequencies found in USA (22). The inding that the frequencies of these behaviors increased signiicantly as the grade got older was similar to the results of the study conduced in USA (22). In the HBSC 2009/2010 study in which many countries from Europe and North America participated, the rate of smoking for at least once in a week in students aged 15 years was found to be 1% in girls in Armenia and 61% in girls in Greenland (the mean frequency among countries 18%) and in contrast to our study, this behavior was observed to have a similar frequency in both genders in 23 of 38 countries and to be higher in girls in four countries (9). It is known that starting smoking at an earlier age increases the possibility of smoking for long periods of time, being a heavier smoker and nicotine addiction (23). In our study, the rate of starting smoking at the age of 11 years and younger was found to be 6.6%. In the Global Youth Tobacco survey (2003), the rate of starting smoking before the age of 10 years in adolescents aged between 13 and 15 years was found to be 30.7% throughout Turkey (10). The fact that the result of our study was much lower than this rate may be considered favourable, but one should keep in mind the age difference between the study groups. In a longitudinal study 30 Turk Pediatri Ars 2016; 51: 22-34 Behaviors of consuming alcoholic beverages In our study, the rates of having experimented an alcoholic beverage and having drunk an alcoholic beverage in the last 30 days were found to be 54.1% and 38.4%, respectively, in high school students in Izmir. It was observed that the rate of having experimented an alcoholic beverage was lower compared to the rate found in high school students in Istanbul (61%) and higher compared to the rate found in high school students in Samsun in 2008 (26.07%) (11, 14). In a research study reviewing the studies which investigated the frequencies of consumption of alcohol in adolescents below the age of 18 years in Turkey, it was reported that the frequency of lifelong consumption of alcohol ranged between 4.4% and 65%, whereas the frequencies of consuming alcohol currently were reported to range between 6.5% and 42% in different studies conducted in different provinces in Turkey with adolescents at different ages (11, 14, 19, 21, 26, 27). In this study, the rate of having experimented an alcoholic beverage was found to be lower compared to the rate found in high school students in USA in 2013 (66.2), but the fact that the rate of having drunk an alcoholic beverage at least for once in the last one month was higher compared to the rate found in USA (34.9%) draws attention to this problem (22). It is known that early onset of regular alcohol consumption increases the risk of being alcoholic in the future (28). The results of this study showed that 12.5% of the students had drunk their irst alcoholic beverage at the age of 11 years and younger. The fact that the rates of smoking, consuming alcohol and getting drunk at the time of data collection were higher in these students compared to the ones who experimented an alcoholic beverage after the age of 11 years is an important inding. These results suggest that regulations related with alcohol alone are insuficient and education programs for children and families should be conducted in order to delay the age of alcohol inititation in children and adolescents. In this study, it was observed that the rate of being drunk (31.6%) was higher compared to Turk Pediatri Ars 2016; 51: 22-34 the rate found in high school students in Istanbul in 2000 (19) and the rate found in high school students in Samsun in 2008 (10.41%) (11, 14). In both studies, the rate of the students who got drunk increased as the grade got older similar to the results of our study (11, 14). In the “Health Behaviour in School-aged Children Study” 2009/2010, the rate of having been drunk for at least two times in the 15-year age group was found to range between 8% (Macedonia, girls) and 57% (Lithuania, boys) (mean: 32%) (9). The fact that the frequency of being drunk was higher compared to the previous studies conducted in our country and close to the mean value of the countries included in the HBSC study and the rate of having drunk an alcoholic beverage for at least once in the last one month was higher compared to the rate found in USA suggests that the prejudice that our country is not risky in terms of consumption of alcoholic beverages because of its cultural and religious properties is incorrect (at least regionally). Nationwide studies should be conducted to regionally and nationally evaluate this risky situation related with behaviors of consuming alcoholic beverages which was noted in Izmir. In this study, it was additionally found that the frequencies of consuming alcohol and being drunk in the last 30 days in the students who stated that they got drunk for the irst time before the age of 11 years (5.3%) were 8-fold and 3-fold higher compared to the students who got drunk for the irst time after the age of 11 years. These indings show that intervention programs should be initiated in the primary school years. On the other hand, our results showed that the behaviors of drinking alcoholic beverages and getting drunk were more common in boys compared to girls similar to the results of the Istanbul 2000 and HBSC 2009/2010 surveys (9, 14). In contrast to these surveys, the behavior of having experimented an alcoholic beverage was found with a higher rate in girls compared to boys in USA 2013 high school students and the rates of having experimented an alcoholic beverage and consuming alcohol currently increased as the grade got older similar to the results of our study (22). Spare time activities According to the results of this study, the rate of performing suficient physical activity was found to be 10.9%, the rate of watching television for long periods of time was found to be 59.9% and the rate of using the internet for long periods of time was found to be 72.8%. While the boys performed physical exercise and watched TV with a higher rate, the rate of using the internet for long periods of time was found to be similar in both genders. The frequency of performing suficient physical exercise did not change as the grade got older in both genders, whereas the frequency of Çavdar et al. Health behaviors in adolescents watching television for long periods of time increased. The frequency of using the internet for long periods of time decreased in the girls and did not change in the boys as the grade got older. The frequency of performing suficient physical activity was found to be lower in both genders compared to the frequency in the 15-year age group in the HBSC 2006 Turkey survey (girls 12.1%, boys 15.7%), the frequency found in the 15-year age group in the HBSC 2009/2010 survey Turkey data (girls 9%, boys 18%) and the frequency found in the 15-18-year age group in the 2010 Turkey Nutrition and Health Survey (girls 9.9%, boys 14.6%) (9, 29, 30). According to the results of the “Health Behaviour in School-aged Children Study” 2009/2010, the rate of the 15-year group students who performed suficient physical activity was 15% (USA, boys 33% and Italy, girls 5%). This rate was 27.1% in high school student in USA (9, 22). In our study, it was noted that the frequency of using the internet for long periods of time was markedly higher compared to the frequency found in adolescents in 2012 in Eskişehir (35.1%) and the frequency of watching television for long periods of time was higher compared to the frequency found in 2008 in Ankara (35,8%) (27.31). In the “Health Behaviour in School-aged Children Study” 2009/2010, the mean rate of using computer for more than 2 hours a day on the weekdays in order to do homework or use the internet was found to be 55% in the 15-year age group (26% in Greenland and 73% in Iceland). This rate was reported to be 37% in Turkey (9). The causes which lead to reduced physical activity in adolescents have been reported to be tests, preperation courses, watching television and excess time spent for comupter games (32). These results show that high school students in Izmir have a more sedantary life and spend more time watching television compared to their peers. Eating behaviors In this study, the frequencies of having breakfast regularly and having breakfast regularly on weekdays were found to be 48.1% and 52.9%, respectively. The frequency of the behavior of having breakfast regularly was found to be higher compared to the frequency found in high school students in 2013 in USA (38.1%), whereas the frequency of of having breakfast regularly on weekdays was found to be lower compared to the mean frequency found in the countries which participated in the HBSC 2009/2010 survey (55%) (Armenia, girls 34% and Netherlands, boys 79%) (9,22). The frequency of having breakfast regularly in both genders was found to be similar to the frequency found in the 15-year age group in the HBSC 2006 Turkey study (girls 49.4%, boys 60.7%) and the frequency found in the 15year age group in the HBSC 2009/2010 study (girls 48%, boys 59%) (9, 33). In these surveys, the frequency of the 31 Çavdar et al. Health behaviors in adolescents behavior of having breakfast regularly was found to be higher in the boys similar to our study (9, 22, 29, 33). Since a correlation between absence of having breakfast regularly and obesity has been shown, we think that interventions directed to develop the behavior of having breakfast regularly should be included in the programs which ight against obesity (34). When eating behaviors in high school students in İzmir were evaluated, the rate of consuming suficient amounts of fruit was found to be 36.2%, the rate of consuming suficient amounts of vegetables was found to be 14.1%, the rate of consuming candy or chocolate frequently was found to be 31.3% and the rate of consuming soft drinks was found to be 18%. It was observed that the rate of the behavior of consuming suficient amounts of fruit decreased as the grade got older both in the girls and in the boys. Similar to the results of the “Health Behaviour in School-aged Children Study” 2006 Turkey (15-year age group), HBSC 2005/2006 (15 year age group), HBSC 2009/2010 (15 year age group), USA 2013 (high school students) surveys, the behavior of consuming suficient amounts of fruit was observed more frequently in the girls (girls 42.1%, boys 25.5%; girls 34%, boys 25%; girls 35%, boys 27%; girls 60%, boys 65.3%, respectively) and the behavior of consuming soft drinks frequently was observed more frequently in the boys (girls 21.1%, boys 26.4%; girls 25%, boys 32%; girls 22%, boys 28%; girls 24.1%, boys 29.9%, respectively) (9, 22, 29, 33). In this study, the behaviors of consuming suficient amounts of vegetable and consuming candy/chocolate frequently were observed more frequently in the girls compared to the boys (girls 24.6%, boys 19.3%; girls 36.7%, boys 22.3%, respectively) similar to the results of the HBSC 2006 Turkey survey (15-year age grouo) (33). It has been reported that the behavior of dieting is observed more frequently in girls and the behavior of performing physical exercise is observed more frequently in boys due to esthetic concerns (32). In this study, the rate of dieting was also found to be higher in the girls compared to the boys. In the HBSC 2009/2010 survey, the rate of dieting in girls in the 15-year age group was found to be 8% which was lower compared to all the other countries (9). The fact that this rate was found to be 10.5% in our study suggests that the dimension of the problem is not very huge, because it is known that a very high rate (62.6%) has been reported in USA (22). Violence-related behaviors In this study, the frequency of having been bullied was found to be 30.3%, the frequency of having been bullied frequently was found to be 12%, the frequency of having bullied others was found to be 29.9% and 32 Turk Pediatri Ars 2016; 51: 22-34 the frequency of having bullied other frequently was found to be 13.3%. It is known that the psychological health is affected negatively and the frequency of the behaviors which affect health negatively including consumption of tobacco, alcohol and substance increase in students who are bullied (35). In this study, the frequency of having been bullied was found to be lower compared to the frequency found in the 15-year age group in the HBSC 2006 Turkey survey (65.2%) and markedly higher compared to the data of the USA 2013 high school students (19.6%) (22, 33). It was notable that the frequencies of having bullied others in both genders (17.1% in the boys and 9.7% in the girls) were higher compared to the same study (13.2% in the boys, 7.3% in the girls) (33). According to our results, violence-related behaviors were found with a higher rate in the boys compared to the girls excluding the behavior of being bullied similar to different studies conducted in previous years in our country and the international results of the HBSC 2009/2010 survey (9, 25, 33, 36-38). No difference was observed between genders in terms of the frequency of the behavior of being bullied. In the Istanbul 2000 survey, violence-related behaviors did not change as the grade got older. In contrast, it was found that the frequency of the behavior of being bullied decreased in both genders as the grade got older and the behavior of bullying others decreased in the girls and did not change in the boys as the grade got older in our study (14). In our country, the frequency of having involved in a physical ight has been found to range between 30.2% and 41% in high school students in different studies conducted in different years (14, 21, 39). The frequency of having involved in a physical ight in the last 12 months which was found to be 41% in this study was notably higher compared to the mean frequency of the countries included in the HBSC 2009/2010 survey (the lowest 18% in Germany, the highest 50% in Greece) and compared to the frequency found in 2013 in USA (24.7%) (9, 22). In addition, the rate of getting frequently involved in a physical ight (27.9%) was found to be markedly higher compared to the mean frequency found in the HBSC 2009/2010 survey (10%) (2% Greenland, girls and 54% Armenia, boys) (9). All these results show that violence-related behaviors in high school students maintain their importance. The fact that this study was a cross-sectional study based on the reports of adolescents and did not included direct measurements of the behaviors was a limitation of this study. It may be questioned if the adolescents reported correctly. Since the survey was anonymous, we think that the rate of correct reporting was high. Turk Pediatri Ars 2016; 51: 22-34 In conlusion, the results of this study show that the behaviors which may affect health negatively are common in adolescents attending high school in Izmir similar to the other studies conducted in our country and in other countries. The most notable inding was that the students were not suficiently active physically and the behaviors of watching television and using the internet for long periods of time which are considered a part of sedentary life style were very common. The second notable inding was that the rates of consuming suficient amounts of vegetable were very low. When this behavior was associated with not having breakfast regularly and not consuming suficient amounts of fruit, it was observed that unhealthy eating behaviors were very common. In addition, the frequency and severity of the behaviors related with consuming tobacco and alcohol were notable. Violence-related behaviors were substantially common especially in the boys and their frequency was higher compared to many countries. All these results suggest that “Adolescent-friend Healthcare Services” should be generalized in our country and physicians should evaluate adolescents in terms of health-related behaviors at each visit. In addition, we think that intervention programs in which a health developing approach is adopted are needed starting from the primary school years. This and similar studies should be repeated regularly to monitor the eficiency of the programs applied and the changes which occur in behavior tendencies in time. Ethics Committee Approval: Ethics committee approval was received for this study. Informed Consent: Verbal informed consent was obtained from parents who participated in this study. Çavdar et al. Health behaviors in adolescents 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Peer-review: Externally peer-reviewed. Author Contributions: Concept - K.E., M.A., O.E.; Design S.Ç., E.Ç.S., E.E., M.A., O.E.; Supervision - E.E., M.A., O.E.; Data Collection and/or Processing - K.A., A.A., B.K., N.K., A.K., S.Ç., E.Ç.S.; Analysis and/or Interpretation - S.Ç., E.Ç.S., M.A., E.E.; Literature Review - S.Ç., E.Ç.S., M.A.; Writing - S.Ç., E.Ç.S., M.A., O.E., E.E.; Critical Review - O.E., M.A., E.E. Conlict of Interest: No conlict of interest was declared by the authors. Financial Disclosure: The authors declared that this study has received no inancial support. 12. 13. 14. 15. References 1. Pekcan H. Adolesan (Delikanlı) Sağlığı. İçinde: Güler Ç, Akın L, editör. Halk Sağlığı Temel Bilgiler. Ankara: Hacettepe Üniversitesi Yayınları; 2012.p.486-535. 16. Anthony D. The state of the world’s children 2011-adolescence: an age of opportunity. NewYork: United Nations Children’s Fund (UNICEF); 2011. World Health Organization. Health for the World’s adolescents: a second chance in the second decade. Geneva, Switzerland: WHO Document Production Services; 2014. Blum RW, Bastos FI, Kabiru CW, Le LC. Adolescent health in the 21st century. Lancet 2012; 379: 1567-8. [CrossRef ] Michaud PA, Ambresin AE, Catalano RF, Diers J, Patton GC. Adolescent Health. In: Detels R, Gulliford M, Karim QA, Tan CC editors. Oxford Textbook of Global Public Health. 6th ed. NewYork: Oxford University Press; 2015.p.1359-70. [CrossRef ] DuRant RH, Smith JA, Kreiter SR, Krowchuk DP. The relationship between early age of onset of initial substance use and engaging in multiple health risk behaviours among young adolescents. Arch Pediatr Adolesc Med 1999; 153: 286-91. [CrossRef ] Kipping RR, Campbell RM, MacArthur GJ, Gunnell DJ, Hickman M. Multiple risk behaviour in adolescence. J Public Health 2012; 34: i1-i2. [CrossRef ] Hacettepe Üniversitesi Nüfus Etütleri Enstitüsü. 2013 Türkiye Nüfus ve Sağlık Araştırması. Ankara: Elma Teknik Basım Matbaacılık; 2014.p.18. Currie C, Zanotti C, Morgan A. Social determinants of health and well-being among young people. Health Behaviour in School-aged Children (HBSC) study: international report from the 2009/2010 survey. Copenhagen: WHO Regional Ofice for Europe, (Health Policy for Children and Adolescents, No. 6); 2012. Erguder T, Çakır B, Aslan D, Warren C W, Jones NR, Asma S. Evaluation of the use of Global Youth Tobacco Survey (GYTS) data for developing evidence-based tobacco control policies in Turkey. BMC Public Health, 2008; 8: 4. [CrossRef ] Arslan HN, Terzi Ö, Dabak Ş, Pekşen Y. Substance, cigarette and alcohol use among high school students in the provincial center of Samsun, Turkey. Erciyes Medical J 2012; 34: 79-84. [CrossRef ] Evren C, Evren B, Bozkurt M. Tobacco use among 10th grade students in Istanbul and related variables. Asian J Psychiatr 2014; 8: 69-75. [CrossRef ] Akdur R. Gençlerde sigara salgını. Sigara veya Sağlık Sempozyumu- 2009, Ankara, 2009. http://www. ssuk.org.tr/content.php?haber_id=602 (Erişim tarihi: 07.09.2015). Ercan O, Alikaşifoğlu M, Erginöz E, et al. İstanbul lise gençlerinde riskli davranışların sıklığı ve cinsiyete göre dağılımı Cerrahpaşa Gençlik Sağlığı Araştırması 2000. Türk Pediatr Arş 2001; 36: 199-211. Çelik P, Esen A, Yorgancıoğlu A, Şen FS, Topçu F. Manisa ilinde lise öğrencilerinin sigaraya karşı tutumları. Toraks Dergisi 2000; 1: 61-6. Güler N, Güler G, Ulusoy H, Bekar M. Lise öğrencileri arasında sigara, alkol kullanımı ve intihar düşüncesi sıklığı. Cumhuriyet Tıp Dergisi 2009; 31: 340-5. 33 Turk Pediatri Ars 2016; 51: 22-34 Çavdar et al. Health behaviors in adolescents 17. Aras S, Semin S, Gunay T, Orcin E, Ozan S. Sexual Attitudes and Risk Taking Behaviors of High School Students in Turkey. J Sch Health 2007; 77: 359-66. [CrossRef ] 18. Doğan DG, Ulukol B. Ergenlerin sigara içmesini etkileyen faktörler ve sigara karşıtı iki eğitim modelinin etkinliği. İnönü Üniversitesi Tıp Fakültesi Dergisi 2010; 17: 179-85. 19. Özbay H, Göka E, Mavili Aktaş A, et al. Ergenlerde sigara, alkol ve madde kullanımının benlik imajı, depresyon ve anksiyete ölçümleri ile ilişkileri. Düşünen Adam 1991; 4: 53-8. 20. Palanci Y, Saka G, Tanrikulu AC, Acemoğlu H. Cigarette use frequency and the affecting factors in primary and high school students in Diyarbakir. Tuberk Toraks 2009; 57: 136-44. 21. Kaya F, Bilgin H, Singer MI. Contributing factors to aggressive behaviors in high school students in Turkey. J Sch Nurs 2012; 28: 56-69. [CrossRef ] 22. Kann L, Kinchen S, Shanklin SL, et al. Youth risk behavior surveillance-United States, 2013. MMWR Surveill Summ 2014; 63: 1-168. 23. Breslau N, Peterson EL. Smoking cessation in young adults: age at initiation of cigarette smoking and other suspected inluences. Am J Public Health 1996; 86: 21420. [CrossRef ] 24. Kaestle CE. Age of smoking milestones: Longitudinal inconsistencies and recanting. J Adol Health 2015; 56: 3828. [CrossRef ] 25. WHO Study Group on Tobacco Product Regulation (TobReg). Advisory note: waterpipe tobacco smoking: health effects, research needs and recommended actions by regulators - 2nd ed. Geneva: World Health Organization; 2015. 26. Ülger Z, Acar C, Torun P. Türkiye’de Okullarda Alkol Kullanımı Konusunda Yapılan Araştırmalar Yeterli mi? Addicta: Turkish J Addictions 2015; 2: 85-112. [CrossRef ] 27. Ortabag T, Ozdemir S, Bakir, B, Tosun N. Health promotion and risk behaviors among adolescents in Turkey. J Sch Nurs 2011; 27: 304-15. [CrossRef ] 28. Kandel DB, Logan JA. Patterns of drug use from adolescence to young adulthood: I. Periods of risk for initiation, continued use, and discontinuation. Am J Public Health 1984; 74: 660-6. [CrossRef ] 29. Currie C, gabhainn SN, Godeau E. Inequalities in young people’s health. Health Behaviour in School-aged 34 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. Children (HBSC) study: international report from the 2005/2006 survey. Copenhagen: WHO Regional Ofice for Europe, (Health Policy for Children and Adolescents, No 5); 2008. Sağlık Bakanlığı Sağlık Araştırmaları Genel Müdürlüğü, Hacettepe Üniversitesi Sağlık Bilimleri Fakültesi Beslenme ve Diyetetik Bölümü, Ankara Numune Eğitim ve Araştırma Hastanesi. Türkiye Beslenme ve Sağlık Araştırması 2010: Beslenme Durumu ve Alışkanlıklarının Değerlendirilmesi Sonuç Raporu. Ankara: Sağlık Bakanlığı Yayın No: 931; 2014. Koyuncu T, Unsal A, Arslantas D. Assessment of internet addiction and loneliness in secondary and high school students. J Pak Med Assoc 2014; 64: 998-1002. 32-Baltacı G, Ersoy G, Karaağaoğlu N ve ark. Ergenlerde Sağlıklı Beslenme, Hareketli Yaşam. Ankara: Sinem Matbaacılık; 2006. Ercan O, Alikaşifoğlu M, Erginöz E ve ark. Okul çağındaki çocukların sağlık davranışları araştırması (HBSC) Türkiye 2006 raporu. İstanbul: Galenos yayınevi; 2009. Haug E, Rasmussen M, Samdal Oet al. Overweight in school-aged children and its relationship with demographic and lifestyle factors: results from the WHOCollaborative Health Behaviour in School-aged Children (HBSC) study. International Journal of Public Health 2009; 54: 167-79. [CrossRef ] Fleming LC, Jacobsen KH. Bullying among middle-school students in low and middle income countries. Health Promot Int 2010; 25: 73-84. [CrossRef ] Alikasifoglu M, Erginoz E, Ercan O, Uysal O, AlbayrakKaymak D. Bullying behaviours and psychosocial health: results from a cross-sectional survey among high school students in Istanbul, Turkey. Eur J Pediatr 2007; 166: 1253-60. [CrossRef ] Alikasifoglu M, Erginoz E, Ercan O, Uysal O, Kaymak DA, Ilter O. Violent behaviour among Turkish high school students and correlates of physical ighting. Eur J Public Health 2004; 14: 173-7. [CrossRef ] Erginoz E, Alikasifoglu M, Ercan O, et al. The Role of Parental, School, and Peer Factors in Adolescent Bullying Involvement Results From the Turkish HBSC 2005/2006 Study. Asia Pac J Public Health 2015; 27: NP1591-603. [CrossRef ] Ozcan S, Ergin A, Saatci E, Bozdemir N, Kurdak H. The prevalence of risky behaviors related to violence in high school students in a southern city, Turkey. Coll Antropol 2008; 32: 1053-8.