02-원저 hwp

Transcription

1 Original Article Korean J Health Promot 2017;17(4): pissn: eissn: 한국청소년의흡연과자살관련행동과의관계 : 제 12 차 (2016 년 ) 청소년건강행태온라인조사 김혜원, 김양근, 백재용, 홍창빈, 이가영, 박태진, 김진승 인제대학교의과대학부산백병원가정의학과 The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey Hye Won Kim, Yang Keun Kim, Jae Yong Paik, Chang Bin Hong, Kayoung Lee, Tae-Jin Park, Jinseung Kim Department of Family Medicine, Busan Paik Hospital, Inje University College of Medicine, Busan, Korea Background: Suicide is the first leading cause of death among teenagers in Korea, and smoking is reported to be one of the risk factors for suicide. The purpose of this study is to investigate the relationship between smoking and smoking amount and suicidal behavior using data from the Korea Youth Risk Behavior Web-based Survey (KYRBS). Methods: Using the 12th KYRBS 2016, we surveyed smoking, smoking amount, suicidal ideation, suicidal planning, and suicidal attempt of adolescents. And a total of 65,528 respondents were divided into current smoker, former smokers, and non-smoker, and 4,012 smokers were classified into 9 cigarettes/day group, cigarettes/day group, and 20 cigarettes/day group according to the amount of smokers. We analyzed the relationship between smoking and smoking amount and suicidal ideation, suicidal planning, and suicidal attempt using multivariate logistic regression analysis. Results: As a result of adjusting for all confounding variables, the current smoker has an odds ratio of (confidence interval ) with suicidal ideation compared to non-smoker and the odds ratio of suicidal planning and suicidal attempt is also statistically significant, respectively. And the odds ratio of all suicidal behaviors was significantly higher in 20 cigarettes/day group than in 9 cigarettes/day group, respectively. Conclusions: We found that smoking and smoking amount was related to suicidal ideation, suicidal planning, and suicidal attempts in adolescents. Therefore, suicidal prevention strategies considering this relationship should be established. Korean J Health Promot 2017;17(4): Keywords: Korea, Adolescent, Suicide, Behavior, Smoking 서 론 Received: July 3, 2017 Accepted: October 17, 2017 Corresponding author: Jinseung Kim, MD Department of Family Medicine, Busan Paik Hospital, Inje University College of Medicine, 75 Bokji-ro, Busanjin-gu, Busan 47392, Korea Tel: , Fax: jinseungkim@inje.ac.kr 2016년도에발표한통계청자료의 2015 년사망원인통계 에따르면자살은 10대부터 30대까지사망원인순위 1위이며, 경제협력개발기구 (Organization for Economic Cooperation and Development, OECD) 국가별자살률을비교해보면 OECD 표준인구 10만명당자살률은평균 12.0명에비해, 한국은 28.7명으로회원국중가장높은수준을나타내 Copyright c 2017 The Korean Society of Health Promotion and Disease Prevention This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

2 220 Korean J Health Promot Vol. 17, No. 4, 2017 고있다. 1) 자살은 2008년 10대인구의사망원인중 1위가된후그순위를지속적으로유지하고있으므로, 청소년의자살을예방하기위한대책이필요하다. 2) 청소년시기는신체적, 정신적급변과정을거쳐성인으로성장과발달을급격하게하는과도기이며, 이로인해외적스트레스가많고감정의기복이많은격동의시기인동시에부모의기대와사회적요구에처음으로직면하는시기이다. 이러한청소년시기는일시적인정서적불균형등으로인해자살의위험성이높으며, 3) 또한지속적인관심이필요한시기이다. 4) 청소년의자살은자살생각 (suicidal ideation), 자살계획 (suicidal planning), 자살시도 (suicidal attempt) 를포함하는일련의자살행동으로이해되고있다. 5) 자살을예방하기위해서는자살을자살생각, 자살계획, 자살시도에이어죽음으로연결되는하나의연속체로가정하여먼저자살위험성이높은청소년들을예측하는것이우선적이나실제적으로자살위험성을예측하기는매우어려운일이기때문에많은경우, 자살의위험성은자살생각을측정함으로써예측된다. 6) 자살생각이자살행위로바로이어지는것은아니지만, 자살생각을가지고있는사람이실제자살로사망할위험이높으며, 7,8) 지속된자살생각은충동적으로자살을시도하게하는요인으로보고되고있다. 9) 또한, 자살을시도한청소년은이전에자살생각을한적이있는경우가많으며, 10) 자살시도는자살생각에비해실제사망할가능성이보다더높다는위험성을가지고있다. 11) 자살을하기전청소년들은자살과관련된어떤행동을하게되는데, 그행동에영향을주는요인을파악하여부정적영향을중재함으로써자살을예방할수있을것으로여겨진다. 2) 청소년의자살생각에영향을미치는요인들을분석한최근의선행연구결과를보면우울, 무망감, 생활스트레스와같은심리적변인, 대중매체의자살보도, 인터넷중독과같은사회문화적변인, 가족관계스트레스와같은가정환경적변인, 학교생활스트레스, 학교폭력, 학업스트레스, 학업성취도와같은학교환경적변인, 경제적수준, 성별, 학년유형과같은개인특성별변인들이청소년자살생각관련위험변인으로보고되고있다. 12) 청소년을대상으로한또다른연구에서는부모, 또래와의부정적인관계, 스트레스, 우울, 비행이자살생각과유의한영향이있는것으로나타났으며, 13) 청소년의자살생각, 자살계획, 자살시도에영향을미치는요인에대한살펴본한국내연구결과에서는일반적특성및음주, 흡연, 주단위중증도운동등과같은건강행태관련특성및정서적특성이영향을미치는요인이라고보고하고있다. 2) 성인의자살생각에영향을미치는위험요인에대한연구에서도결혼, 교육수준과같은인구사회학적변인뿐만아 니라흡연, 음주, 수면, 주관적건강상태, 주관적체형인식등의건강행위와건강상태가성인의자살생각에영향을미치는위험요인으로작용하고있다고보고하고있다. 14,15) 청소년의자살에영향을미치는여러위험요인중청소년의흡연에대한연구를살펴보면, 청소년의흡연은자아통제력이부족한시기에흡연을일찍시작할수록흡연을중단하기어렵고규칙적인흡연자가되기쉬워서, 흡연이장기간지속되고흡연의양도많아지게된다. 16) 또한어린나이에흡연을시작하면니코틴의존성이높아져정신건강에부정적인영향을미친다고알려져있다. 17) 청소년의자살에영향을미치는요인을분석한외국연구의결과에서도청소년흡연은단독요인으로청소년의자살과자살생각에유의한영향을미치는중요한인자이며, 흡연이자살에영향을미칠뿐만아니라흡연의양이자살과유의한관련이있었다. 16,18) 이전의연구들을살펴보면, 청소년자살연구들은자살생각과의관련요인들을연구한것이주를이루고있으며, 각각의요인과자살관련행동과의연관성을파악한연구는부족한실정이다. 흡연과자살관련행동과의관계를살펴본외국연구에따르면흡연과자살관련행동의교차비는유의미하게나왔음을알수있으며, 19) 간호사를대상으로흡연과자살과의관계를살펴본한외국의연구에서도흡연은자살에영향을미치는요인으로나타나있다. 20) 이렇듯자살에흡연은상당한영향을미치는요인으로알려져오고있어, 본연구는청소년건강행태온라인조사자료를이용하여한국청소년의흡연여부및흡연량과자살생각, 자살계획, 자살시도 ( 자살관련행동 ) 와의연관성을살펴보고자하였다. 또한, 이를토대로청소년금연교육및흡연예방교육과자살예방교육프로그램을위한기초자료를제공하고자한다. 1. 연구대상 방 법 본연구는질병관리본부에서주관한 2016년제12차청소년건강행태온라인조사원시자료를이용하여한국청소년의흡연여부및흡연량과자살관련행동과의연관성을알아보았다. 이자료는온라인조사홈페이지에서사용승인을받아연구의목적에맞게분석하였다. 21) 이조사는 2016년 4월기준의전국중학교 1학년부터고등학교 3학년학생을대상으로하고있으며, 중학교 400개교, 고등학교 400개교, 총 800개교의 67,983 명을대상으로조사하였고, 798개교, 65,528명이조사에참여하여 96.4% 참여율을보였다. 본연구의분석대상자는제12차청소년

3 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 221 건강행태온라인조사참여자로대상자수는남학생 33,803명, 여학생 31,725명, 총 65,528명이다. 2. 자료수집제12차청소년건강행태온라인조사는인터넷이가능한학교컴퓨터실에서표본학급학생들에게 1인 1대컴퓨터를배정하고, 무작위로자리를배치하여익명성자기기입식온라인조사방법으로수행하였다. 3. 변수정의 (Figure 1) 1) 종속변수본연구에서종속변수는자살생각, 자살계획, 자살시도이다. 자살생각은 최근 12개월동안, 심각하게자살을생각한적이있었습니까? 라는문항에 최근 12개월동안없다, 최근 12개월동안있다 로응답하도록구성되어있으며, 없다, 있다 로재구성하였다. 자살계획은 최근 12개월동안, 자살하기위해구체적인계획을세운적이있었습니까? 라는문항에 없다, 있다, 자살시도는 최근 12개월동안, 자살을시도한적이있었습니까? 라는문항에 없 다, 있다 의응답으로재구성하였다. 2) 독립변수본연구에서는자살관련행동과의연관성을알아보기위한독립변수로흡연여부와흡연량을이용하였다. 흡연관련문항에서 지금까지담배를한두모금이라도피워본적이있습니까? 라는문항에 있다 라고응답한경우를 흡연경험자 로정의하였고, 없다 라고응답한경우를 비흡연자 로정의하였다. 또한 최근 30일동안, 담배를한대 ( 한개비 ) 라도피운날은며칠입니까? 라는문항에 최근 30일동안없다 라고응답한경우 과거흡연자, 월 1-2일 에서 매일 까지응답한경우를 현재흡연자 로정의하였고, 두문항의응답을바탕으로분석대상자를 비흡연자, 과거흡연자, 현재흡연자 로분류하여분석하였다. 흡연량은 최근 30일동안담배를하루에평균몇개비피웠습니까? 라는문항의응답을통해 9개비이하, 개비, 20개비이상 으로재분류하여분석하였다. 3) 혼란변수본연구에서혼란변수는종전의연구에서살펴본자살생각에영향을미치는요인들을바탕으로 2,12-15) 크게인구사 Figure 1. Framework.

4 222 Korean J Health Promot Vol. 17, No. 4, 2017 회학적요인, 건강상태요인, 정신건강요인부분으로나누어분석하였다. 인구사회학적요인은성별, 학교형태, 경제상태, 거주형태로구성하였고, 건강상태요인은신체활동, 평생음주경험, 약물경험, 주관적건강인지로구성하였다. 정신건강요인은스트레스인지, 학업성적, 주관적행복, 슬픔 / 절망감경험을포함하여구성하였다. 학교형태는중학교, 고등학교로구분하였고, 경제상태는상, 중, 하로분류하였다. 거주형태 가족과함께살고있다 와 그렇지않다 로분류하였고, 주관적건강인지는 평상시자신의건강상태가어떻다고생각합니까? 라는문항의응답에서 매우건강한편이다 와 건강한편이다 를 건강한편이다 로 보통이다 를 보통이다 로 건강하지못한편이다 와 매우건강하지못한편이다 를 건강하지못한편이다 로재분류하였다. 신체활동에관한변수는미국스포츠의학대학 (American College of Sports Medicine) 에서청소년의건강을위해주 3일이상, 20분의격렬한신체활동이나, 주 5일이상의 30분의중등도신체활동및적어도주 2일의고강도훈련을권장하고있어 22) 이를바탕으로 최근 7일동안심장박동이평상시보다증가하거나, 숨이찬정도의신체활동을하루에총합이 60분이상한날은며칠입니까? 라는문항의응답에서 주 1일미만, 주 1-4일, 주 5일이상 으로재분류하였다. 평생음주경험은 지금까지 1잔이상술을마셔본적이있습니까? 에 있다 로응답한경우를 음주경험자, ' 없다 ' 로응답한경우를 음주비경험자 로분류하였다. 약물경험에대해서는 지금까지습관적으로또는일부러약물을먹거나부탄가스, 본드등을마신적이있습니까? 라는문항에 있다 로응답한경우 약물경험자, 없다 로응답한경우 약물비경험자 로분류하였다. 스트레스인지는 평상시스트레스를얼마나느끼고있습니까? 라는문항에 대단히많이느낀다, 많이느낀다, 조금느낀다 로응답한경우를 인지하고있다 로 별로느끼지않는다, 전혀느끼지않는다 로응답한경우를 인지하고있지않다 로재분류하였다. 학업성적에대해서는 최근 12개월동안, 학업성적은어떻습니까? 라는문항의응답을통해상, 중, 하로분류하였고, 주관적행복은 행복한편이다, 보통이다, 불행한편이다 로재분류하여분석하였다. 슬픔 / 절망감경험은 최근 12개월동안, 2주내내일상생활을중단할정도로슬프거나절망감을느낀적이있었습니까? 라는문항에 없다, 있다 응답으로구분하였다. 4. 통계분석청소년건강행태온라인조사는복합표본설계를이용한자료이므로복합표본분석통계방법을이용하여분석하였다. 대상자의일반적특성은빈도분석을통해빈도와백분율을분석하였고, 일반적특성에따른자살관련행동의차이는교차분석을이용하여분석하였다. 흡연여부및흡연량과자살관련행동과의연관성을알아보기위해다변량로지스틱회귀분석을이용하였으며, 자살관련행동에영향을줄수있는혼란변수를통제하여분석하였다. 분석에는 IBM SPSS Statistics version 23.0 (IBM Corp., New York, NY, USA) 프로그램을이용하였으며, 유의수준은 P<0.05 로하였다. 결과 1. 연구대상자의일반적특성 (Table 1) 연구대상자는총 65,528명으로남학생 33,803명 (51.6%), 여학생 31,725명 (48.4%) 이었고, 학교형태는중학교가 32,219명 (45.4%), 고등학교가 33,309명 (54.6%) 이었다. 거주형태는 가족과함께살고있다 가전체 62,263명 (95.5%) 이었고, 경제상태는 중 이 31,056명 (47.3%) 으로가장많았다. 주관적건강은 건강한편이다 로응답한학생이 71.8% 로가장많았고, 신체활동은 주 1-4일 로응답한학생이전체학생의 50.1%, 남학생은 53.1%, 여학생은 주 1일미만, 주 1-4일 이각각 46.2%, 46.8% 였다. 평생음주경험은남학생이 14,294명 (43.5%), 여학생이 10,510명 (33.6%) 으로남학생이더많았고, 약물경험도남학생이 347명 (1.1%), 여학생이 206명 (0.7%) 을나타내었다. 스트레스인지는전체학생의 80.3% 가 인지하고있다 로응답하였으나, 주관적행복은전체학생의 66.6% 가 행복한편이다 로응답하였다. 흡연에관한문항에서평생흡연경험은전체 9,511명 (14.8%) 이 있다 에응답하였고, 이중과거흡연자가 5,499명 (8.5%), 현재흡연자가 4,012명 (6.3%) 으로나타났으며, 흡연량은 9개비이하 라고응답한학생이 3,171명 (78.4%) 으로가장많았다. 자살생각은 12.1%, 자살계획은 4.0%, 자살시도는 2.4% 의학생들이 있다 로응답하였으며, 특히여학생은자살생각을한적이있는학생이 14.9%, 자살계획을세운적이있는학생이 4.3%, 자살시도를한적이있는학생이 2.7% 로남학생보다비율이높았다. 2. 일반적특성에따른자살관련행동의차이 1) 일반적특성에따른자살생각의차이 (Table 2) 거주형태에따른자살생각은가족과함께사는학생중자살생각을한학생은 11.9%, 그렇지않은학생중자살생각을한학생은 17.4% (P<0.001) 였으며, 경제상태가 상 인학생중자살생각을한학생은 11.1%, 하 인학생은 19.3%

5 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 223 Table 1. Characteristics of study population Charateristics Total P Boys (n=33,803) P Girls (n=31,725) P a Gender Boys 33,803 (51.6) Girls 31,725 (48.4) School <0.001 <0.001 <0.001 Middle school 32,219 (45.4) 16,742 (45.4) 15,477 (45.4) High school 33,309 (54.6) 17,061 (54.6) 16,248 (54.6) Residence type <0.001 <0.001 <0.001 With family 62,263 (95.5) 32,044 (95.2) 30,219 (95.8) None 3,265 (4.5) 1,506 (4.8) 1,506 (4.2) Economic status <0.001 <0.001 <0.001 High 24,244 (37.2) 13,324 (39.5) 10,920 (34.6) Middle 31,056 (47.3) 15,145 (44.7) 15,911 (50.1) Low 10,228 (15.6) 5,334 (15.9) 4,894 (15.3) Perceived health <0.001 <0.001 <0.001 Healthy 47,233 (71.8) 26,008 (76.6) 21,225 (66.5) Average health 14,223 (21.9) 6,106 (18.3) 8,117 (25.9) Unhealthy 4,072 (6.3) 1,689 (5.1) 2,383 (7.7) Physical activity for a week b, days <0.001 <0.001 <0.001 <1 23,817 (36.8) 9,441 (28.1) 14,376 (46.2) ,844 (50.1) 17,805 (53.1) 15,039 (46.8) 5 8,867 (13.1) 6,557 (18.8) 2,310 (7.0) Alcohol drinking <0.001 <0.001 <0.001 No 40,724 (61.2) 19,509 (56.5) 21,215 (66.4) Yes 24,804 (38.8) 14,294 (43.5) 10,510 (33.6) Drug experience <0.001 <0.001 <0.001 No 64,975 (99.1) 33,456 (98.9) 31,519 (99.3) Yes 553 (0.9) 347 (1.1) 206 (0.7) Perceived stress <0.001 <0.001 <0.001 No 13,161 (19.7) 8,635 (25.0) 4,526 (14.0) Yes 52,367 (80.3) 25,168 (75.0) 27,199 (86.0) Academic performance <0.001 <0.001 <0.001 High 25,239 (38.1) 13,244 (38.9) 11,995 (37.2) Middle 18,626 (28.6) 9,301 (27.8) 9,325 (29.6) Low 21,663 (33.3) 11,258 (33.4) 10,405 (33.2) Percevied happiness <0.001 <0.001 <0.001 Happiness 43,956 (66.6) 23,787 (69.8) 20,169 (63.1) Average happiness 16,743 (25.8) 7,732 (23.2) 9,011 (28.6) Unhappiness 4,829 (7.6) 2,284 (6.9) 2,545 (8.3) Sadness during the past 12 months <0.001 <0.001 <0.001 No 48,993 (74.5) 26,885 (79.1) 22,108 (69.5) Yes 16,535 (25.5) 6,918 (20.9) 9,617 (30.5) Smoking <0.001 <0.001 <0.001 Non-smoker 56,017 (85.2) 26,535 (78.1) 29,482 (92.9) Former smoker 5,499 (8.5) 4,110 (12.3) 1,389 (4.4) Current smoker 4,012 (6.3) 3,158 (9.6) 854 (2.7) Smoking amount, cigarettes/day <0.001 <0.001 < ,171 (78.4) 2,489 (78.4) 682 (78.7) (13.4) 444 (14.1) 95 (10.8) (8.1) 225 (7.5) 77 (10.5) Suicidal ideation <0.001 <0.001 <0.001 No 57,683 (87.9) 30,636 (90.5) 27,047 (85.1) Yes 7,845 (12.1) 3,167 (9.5) 4,678 (14.9) Suicidal planning <0.001 <0.001 <0.001 No 62,894 (96.0) 32,529 (96.2) 30,365 (95.7) Yes 2,634 (4.0) 1,274 (3.8) 1,360 (4.3) Suicidal attempt <0.001 <0.001 <0.001 No 63,998 (97.6) 33,137 (98.0) 30,861 (97.3) Yes 1,530 (2.4) 666 (2.0) 864 (2.7) Values are presented as number (%) using frequency analysis in complex sample design. a P values are obtained by F-test. b Physical activity for a week is defined as a day in which a total of 60 minutes or more of a day physical activity of the heartbeat is increased or breathed during the past 7 days.

6 224 Korean J Health Promot Vol. 17, No. 4, 2017 Table 2. Differences in suicidal ideation according to general characteristics Total Boys Girls Variables Suicidal ideation P a Suicidal ideation P a Suicidal ideation No Yes No Yes No Yes P a School Middle school 28,378 (88.0) 3,841 (12.0) 15,200 (90.6) 1,542 (9.4) 13,178 (85.2) 2,299 (14.8) High school 29,305 (87.8) 4,004 (12.2) 15,436 (90.4) 1,625 (9.6) 13,869 (85.0) 2,379 (15.0) Residence type <0.001 < With family 54,963 (88.1) 7,300 (11.9) 29,146 (90.8) 2,898 (9.2) 25,817 (85.2) 4,402 (14.8) None 2,720 (82.6) 545 (17.4) 1,490 (83.7) 269 (16.3) 1,230 (81.4) 276 (18.6) Economic status <0.001 <0.001 <0.001 High 21,627 (89.0) 2,617 (11.0) 12,165 (91.1) 1,159 (8.9) 9,462 (86.5) 1,458 (13.5) Middle 22,782 (89.4) 3,274 (10.6) 13,926 (91.8) 1,219 (8.2) 13,856 (87.0) 2,055 (13.0) Low 8,274 (80.7) 1,954 (19.3) 4,545 (85.1) 789 (13.8) 3,729 (75.6) 1,165 (24.4) Perceived health <0.001 <0.001 <0.001 Healthy 42,951 (90.9) 4,282 (9.1) 24,083 (92.5) 1,925 (7.5) 18,868 (88.8) 2,357 (11.2) Average health 11,891 (83.5) 2,332 (16.5) 5,297 (86.7) 809 (13.3) 6,594 (81.1) 1,523 (18.9) Unhealthy 2,841 (69.1) 1,231 (30.9) 1,256 (73.0) 433 (27.0) 1,585 (66.3) 798 (33.7) Physical activity for a week, days <0.001 <1 21,015 (88.1) 2,802 (11.9) 8,578 (90.7) 863 (9.3) 12,437 (86.4) 1,939 (13.5) ,830 (87.7) 4,014 (12.3) 16,122 (90.5) 1,683 (9.5) 12,708 (84.2) 2,331 (13.6) 5 7,838 (88.1) 1,029 (11.9) 5,936 (90.2) 621 (9.8) 1,902 (82.1) 408 (17.9) Alcohol drinking <0.001 <0.001 <0.001 No 36,695 (89.9) 4,029 (10.1) 18,020 (92.2) 1,489 (7.8) 18,675 (87.8) 2,540 (12.2) Yes 20,988 (84.7) 3,816 (15.3) 12,616 (88.3) 1,678 (11.7) 8,372 (79.7) 2,138 (20.3) Drug experience <0.001 <0.001 <0.001 No 57,354 (88.1) 7,621 (11.9) 30,396 (90.7) 3,060 (9.3) 26,958 (85.4) 4,561 (14.6) Yes 329 (58.3) 224 (41.7) 240 (68.4) 107 (31.6) 89 (41.6) 117 (58.4) Perceived stress <0.001 <0.001 <0.001 No 12,814 (97.3) 347 (2.7) 8,410 (97.3) 225 (2.7) 4,404 (97.2) 122 (2.8) Yes 44,869 (85.6) 7,498 (14.4) 22,226 (88.2) 2,942 (11.8) 22,643 (83.1) 4,556 (16.9) Academic performance <0.001 <0.001 <0.001 High 22,575 (89.4) 2,664 (10.6) 12,130 (91.4) 1,114 (8.6) 10,445 (87.0) 1,550 (13.0) Middle 16,657 (89.3) 1,969 (10.7) 8,524 (91.5) 777 (8.5) 8,133 (87.0) 1,192 (13.0) Low 18,451 (85.0) 3,212 (15.0) 9,982 (88.5) 1,276 (11.5) 8,469 (81.2) 1,936 (18.8) Perceived happiness <0.001 <0.001 <0.001 Happiness 41,180 (93.7) 2,776 (6.3) 22,555 (94.8) 1,232 (5.2) 18,625 (92.3) 1,544 (7.7) Average happiness 13,920 (83.2) 2,823 (16.8) 6,734 (87.1) 998 (12.9) 7,186 (79.7) 1,825 (20.3) Unhappiness 2,583 (53.1) 2,246 (46.9) 1,347 (58.6) 937 (41.4) 1,236 (48.1) 1,309 (51.9) Sadness during the past 12 months <0.001 <0.001 <0.001 No 46,894 (95.6) 2,099 (4.4) 25,960 (96.5) 925 (3.5) 20,934 (94.6) 1,174 (5.4) Yes 10,789 (65.2) 5,746 (34.8) 4,676 (67.7) 2,242 (32.3) 6,113 (63.3) 3,504 (36.7) Smoking <0.001 <0.001 <0.001 Non-smoker 49,778 (88.7) 6,239 (11.3) 24,322 (91.4) 2,213 (8.6) 25,456 (86.2) 4,026 (13.8) Former smoker 4,690 (85.6) 809 (14.4) 3,639 (89.0) 471 (11.0) 1,051 (75.6) 338 (24.4) Current smoker 3,215 (80.0) 797 (20.0) 2,675 (84.5) 483 (15.5) 540 (62.3) 314 (37.7) Smoking amount, cigarettes/day <0.001 < ,602 (81.9) 569 (18.1) 2,156 (86.4) 333 (13.6) 446 (64.5) 236 (35.5) (79.3) 112 (20.7) 366 (83.0) 78 (17.0) 61 (60.7) 34 (39.3) (63.0) 116 (37.0) 153 (68.4) 72 (31.6) 33 (47.8) 44 (52.2) Values are presented as number (%) using cross tabulation analysis in complex sampling design. a Calculated by Rao-Scott chi-square test. 로각각통계적으로유의한차이를보였다 (P<0.001). 스트레스와슬픔 / 절망감경험에따른자살생각도통계적으로유의한차이를보였는데, 스트레스를인지하는학생 (14.4%) 이인지하고있지않는학생 (2.7%) 보다자살생각비율이높았고, 슬픔 / 절망감을경험한적이있는학생 (34.8%) 이없는학생 (4.4%) 보다높은자살생각을보였다. 현재흡연자중

7 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 225 Table 3. Differences in suicidal planning according to general characteristics Total Boys Girls Variables Suicidal planning P a Suicidal planning P a Suicidal planning No Yes No Yes No Yes P a School <0.001 Middle school 30,835 (95.7) 1,384 (4.3) 16,115 (96.2) 627 (3.8) 14,720 (95.2) 757 (4.8) High school 32,059 (96.2) 1,250 (3.8) 16,414 (96.1) 647 (3.9) 15,645 (96.2) 603 (3.8) Residence type <0.001 <0.001 <0.001 With family 59,883 (96.2) 2,380 (3.8) 30,924 (96.5) 1,120 (3.5) 28,959 (95.9) 1,260 (4.1) None 3,011 (91.2) 254 (8.8) 1,605 (90.0) 154 (10.0) 1,406 (92.8) 100 (7.2) Economic status <0.001 <0.001 <0.001 High 23,274 (96.0) 970 (4.0) 12,798 (96.0) 526 (4.0) 10,476 (96.0) 444 (4.0) Middle 30,065 (96.8) 991 (3.2) 14,697 (97.0) 448 (3.0) 15,368 (96.5) 543 (3.5) Low 9,555 (93.4) 673 (6.6) 5,034 (94.2) 300 (5.8) 4,521 (92.6) 373 (7.4) Perceived health <0.001 <0.001 <0.001 Healthy 45,796 (97.0) 1,437 (3.0) 25,189 (96.9) 819 (3.1) 20,607 (97.2) 618 (2.8) Average health 13,521 (95.0) 702 (5.0) 5,832 (95.5) 274 (4.5) 7,689 (94.6) 428 (5.4) Unhealthy 3,577 (87.5) 495 (12.5) 1,508 (88.3) 181 (11.7) 2,069 (86.8) 314 (13.2) Physical activity for a week, days < <0.001 <1 22,937 (96.3) 880 (3.7) 9,092 (96.3) 349 (3.7) 13,845 (96.4) 531 (3.6) ,506 (95.9) 1,338 (4.1) 17,158 (96.3) 647 (3.7) 14,348 (95.3) 691 (4.7) 5 8,451 (95.1) 416 (4.9) 6,279 (95.5) 278 (4.5) 2,172 (94.0) 138 (6.0) Alcohol drinking <0.001 <0.001 <0.001 No 39,428 (96.8) 1,296 (3.2) 18,934 (97.0) 575 (3.0) 20,494 (96.6) 721 (3.4) Yes 23,466 (94.7) 1,388 (5.3) 13,595 (95.1) 699 (4.9) 9,871 (94.0) 639 (6.0) Drug experience <0.001 <0.001 <0.001 No 62,501 (96.2) 2,474 (3.8) 32,266 (96.4) 1,190 (3.6) 30,235 (96.0) 1,284 (4.0) Yes 393 (69.7) 160 (30.3) 263 (74.2) 84 (25.8) 130 (62.3) 76 (37.7) Perceived stress <0.001 <0.001 <0.001 No 12,936 (98.2) 225 (1.8) 8,468 (98.0) 167 (2.0) 4,468 (98.7) 58 (1.3) Yes 49,958 (95.4) 2,409 (4.6) 24,061 (95.6) 1,107 (4.4) 25,897 (95.3) 1,302 (4.7) Academic performance <0.001 <0.001 <0.001 High 24,337 (96.4) 902 (3.6) 12,792 (96.5) 452 (3.5) 11,545 (96.3) 450 (3.7) Middle 17,985 (96.6) 641 (3.4) 8,993 (96.7) 308 (3.3) 8,992 (96.5) 333 (3.5) Low 20,572 (94.9) 1,091 (5.1) 10,744 (95.3) 514 (4.7) 9,828 (94.4) 577 (5.6) Perceived happiness <0.001 <0.001 <0.001 Happiness 42,959 (97.7) 997 (2.3) 23,206 (97.5) 581 (2.5) 19,753 (97.9) 416 (2.1) Average happiness 15,947 (95.4) 796 (4.6) 7,390 (95.6) 342 (4.4) 8,557 (95.2) 454 (4.8) Unhappiness 3,988 (82.5) 841 (17.5) 1,933 (84.4) 351 (15.6) 2,055 (80.7) 490 (19.3) Sadness during the past 12 months <0.001 <0.001 <0.001 No 48,304 (98.6) 689 (1.4) 26,502 (98.6) 383 (1.4) 21,802 (98.6) 306 (1.4) Yes 14,590 (88.3) 1,945 (11.8) 6,027 (87.0) 891 (13.0) 8,563 (89.1) 1,054 (10.9) Smoking <0.001 <0.001 <0.001 Non-smoker 54,046 (96.5) 1,971 (3.5) 25,678 (96.7) 857 (3.3) 28,368 (96.2) 1,114 (3.8) Former smoker 5,217 (95.1) 282 (4.9) 3,936 (96.0) 174 (4.0) 1,281 (92.5) 108 (7.5) Current smoker 3,631 (90.3) 381 (9.7) 2,915 (92.0) 243 (8.0) 716 (83.7) 138 (16.3) Smoking amount <0.001 <0.001 < ,930 (92.4) 241 (7.6) 2,340 (93.9) 149 (6.1) 590 (86.5) 92 (13.5) (88.9) 58 (11.1) 405 (90.9) 39 (9.1) 76 (78.8) 19 (21.2) (72.4) 82 (27.6) 170 (74.1) 55 (25.9) 50 (67.7) 27 (32.3) Values are presented as number (%) using cross tabulation analysis in complex sampling design. a Calculated by Rao-Scott chi-square test. 20.0% 가자살생각을한적이있었으며, 이는비흡연자 11.3%, 과거흡연자 14.4% 와유의한차이를보였다 (P<0.001). 흡연량은 20개비이상핀학생중 37.0% 가자살생각을한적이있으며, 9개비이하, 개비를피운학생중자살생각을한적이있다고응답한학생보다각각 18.9%p, 16.3%p 더높게나 타났다 (P<0.001). 2) 일반적특성에따른자살계획및자살시도의차이 (Tables 3, 4) 거주형태, 경제상태, 주관적건강인지, 신체활동, 평생

8 226 Korean J Health Promot Vol. 17, No. 4, 2017 Table 4. Differences in suicidal attempt according to general characteristics Total Boys Girls Variables Suicidal attempt P a Suicidal attempt P a Suicidal attempt No Yes No Yes No Yes P a School <0.001 Middle school 31,399 (97.4) 820 (2.6) 16,432 (98.1) 310 (1.9) 14,967 (96.6) 510 (3.4) High school 32,599 (97.8) 710 (2.2) 16,705 (97.8) 356 (2.2) 15,894 (97.8) 354 (2.2) Residence type <0.001 < With family 60,901 (97.8) 1,362 (2.2) 31,482 (98.2) 562 (1.8) 29,419 (97.4) 800 (2.6) None 3,097 (94.0) 168 (6.0) 1,655 (92.9) 104 (7.1) 1,442 (95.4) 64 (4.6) Economic status <0.001 <0.001 <0.001 High 23,686 (97.7) 558 (2.3) 13,033 (97.8) 291 (2.2) 10,653 (97.5) 267 (2.5) Middle 30,501 (98.2) 555 (1.8) 14,936 (98.5) 209 (1.5) 15,565 (97.8) 346 (2.2) Low 9,811 (95.9) 417 (4.1) 5,168 (96.7) 166 (3.3) 4,643 (94.9) 251 (5.1) Perceived health <0.001 <0.001 <0.001 Healthy 46,469 (98.4) 764 (1.6) 25,620 (98.5) 388 (1.5) 20,849 (98.3) 376 (1.7) Average health 13,758 (96.7) 465 (3.3) 5,950 (97.3) 156 (2.7) 7,808 (96.2) 309 (3.8) Unhealthy 3,771 (92.2) 301 (7.8) 1,567 (91.8) 122 (8.2) 2,204 (92.5) 179 (7.5) Physical activity for a week, days <0.001 <1 23,307 (97.8) 510 (2.2) 9,256 (98.0) 185 (2.0) 14,051 (97.7) 325 (2.3) ,072 (97.6) 772 (2.4) 17,470 (98.1) 335 (1.9) 14,602 (97.0) 437 (3.0) 5 8,619 (97.1) 248 (2.9) 6,411 (97.6) 146 (2.4) 2,208 (95.8) 102 (4.2) Alcohol drinking <0.001 <0.001 <0.001 No 40,008 (98.2) 716 (1.8) 19,209 (98.4) 300 (1.6) 20,799 (98.0) 416 (2.0) Yes 23,990 (96.7) 814 (3.3) 13,928 (97.3) 366 (2.7) 10,062 (95.8) 448 (4.2) Drug experience <0.001 <0.001 <0.001 No 63,563 (97.8) 1,412 (2.2) 32,847 (98.1) 609 (1.9) 30,716 (97.5) 803 (2.5) Yes 435 (77.8) 118 (22.2) 290 (83.2) 57 (16.8) 145 (68.9) 61 (31.1) Perceived stress <0.001 <0.001 <0.001 No 13,031 (99.0) 130 (1.0) 8,544 (98.9) 941 (1.1) 4,487 (99.0) 39 (1.0) Yes 50,967 (97.3) 1,400 (2.7) 24,593 (97.6) 575 (2.4) 26,374 (97.0) 825 (3.0) Academic performance <0.001 <0.001 <0.001 High 24,731 (98.0) 508 (2.0) 12,992 (98.0) 252 (2.0) 11,739 (97.9) 256 (2.1) Middle 18,275 (98.2) 351 (1.8) 9,156 (98.4) 145 (1.6) 9,119 (97.9) 206 (2.1) Low 20,992 (96.8) 671 (3.2) 10,999 (97.5) 269 (2.5) 10,003 (96.0) 402 (4.0) Perceived happiness <0.001 <0.001 <0.001 Happiness 43,440 (98.9) 516 (1.2) 23,526 (98.9) 261 (1.1) 19,914 (98.8) 255 (1.2) Average happiness 16,254 (97.1) 489 (2.9) 7,539 (97.6) 193 (2.4) 8,715 (96.8) 296 (3.2) Unhappiness 4,304 (88.9) 525 (11.1) 2,072 (90.2) 212 (9.8) 2,232 (87.7) 313 (12.3) Sadness during the past 12 months <0.001 <0.001 <0.001 No 48,690 (99.4) 303 (0.6) 26,751 (99.5) 134 (0.5) 21,939 (99.3) 169 (0.7) Yes 15,308 (92.4) 1,227 (7.6) 6,386 (92.0) 532 (8.0) 8,922 (92.7) 695 (7.3) Smoking <0.001 <0.001 <0.001 Non-smoker 54,932 (98.1) 1,085 (1.9) 26,114 (98.4) 421 (1.6) 28,818 (97.8) 664 (2.2) Former smoker 5,329 (96.9) 170 (3.1) 4,023 (97.8) 87 (2.2) 1,306 (94.2) 83 (5.8) Current smoker 3,737 (92.7) 275 (7.3) 3,000 (94.6) 158 (5.4) 737 (85.1) 117 (14.9) Smoking amount cigarettes/day <0.001 <0.001 < ,017 (94.8) 154 (5.2) 2,407 (96.4) 82 (3.6) 610 (88.7) 72 (11.3) (90.4) 51 (9.6) 410 (92.6) 34 (7.4) 78 (79.0) 17 (21.0) (76.0) 70 (24.0) 183 (80.2) 42 (19.8) 49 (64.3) 28 (35.7) Values are presented as number (%) using cross tabulation analysis in complex sampling design. a Calculated by Rao-Scott chi-square test. 음주경험, 약물경험, 스트레스, 학업성적, 주관적행복, 슬픔 / 절망감경험에따른자살계획여부는모두유의한차이를보였다 (Table 3). 학교형태와남학생의신체활동에따른자살시도의차이를제외하고는자살시도에서도모두통계적으로유의한차이를보였다 (Table 4). 흡연량과의관계에서는 20개비이상피우는여학생중 32.3% 가자살 계획을세운적이있으며, 이는 9개비이하 (13.5%), 10-19개비 (21.2%) 를피우는여학생과통계적으로유의한차이를보였다 (P<0.001). 약물경험이있는학생중자살시도를한적이있는학생은 22.2% 로약물경험이없는학생중자살시도를한적이있는학생 (2.2%) 보다 20.0%p 더높게나타났으며, 유의한차이를보였다 (P<0.001). 행복한편이다 라

9 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 227 Table 5. Odds ratio for suicidal ideation according to smoking and daily smoking amount by logistic regression Variables Model 1 a Model 2 b Model 3 c Model 4 d Smoking Total Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P e <0.001 <0.001 < Boys Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P <0.001 <0.001 < Girls Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P < <0.001 <0.001 Smoking amount, cigarettes/day Smoker ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 < Boys ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 < Girls ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P Values are presented as adjusted odds ratio (95% confidence interval). Calculated by multivariate logistic regression analysis using complex sampling design. a Logistic regression for suicidal ideation with smoking and smoking amount without adjustment. b Logistic regression for suicidal ideation with smoking and smoking amount after adjustment for school, residence type, and economic status. c Logistic regression for suicidal ideation with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, and drug experience. d Logistic regression for suicidal ideation with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, drug experience, perceived stress, academic performance, perceived happiness, and sadness during the past 12 months. e P values are obtained by F-test. 고응답한여학생중자살시도를한적이있는여학생은 12.3% 였고 보통이다, 불행한편이다 라고응답한여학생중자살시도를한적이있는여학생이각각 1.2%, 3.2% 로주관적행복에따른자살시도의차이는통계적으로유의하였다 (P<0.001). 20개비이상담배를피운전체학생군에서는 24.0%, 9개비이하를피운군에서는 5.2% 가자살시도를한적이있다고응답하였고, 특히여학생에서 20개비이상담배를피운학생 (25.7%) 이 9개비이하를피운학생 (11.3%) 보다 24.4%p 더높게자살시도를한적이있다고응답했으며이는통계적으로유의한차이를보였다 (P<0.001). 3. 흡연여부및흡연량과자살관련행동과의연관성대상자의흡연여부및흡연량과자살관련행동과의연관성을알아보기위하여인구사회학적요인, 건강상태요인, 정신건강요인으로나누어혼란변수들을통제하고다변량로지스틱회귀분석을한결과는표 5-7과같다. 단계적으로요인을추가하여분석하였으며, 모형 1은통제없이분석하고, 모형 2는인구사회학적요인을통제하고분석하였다. 모형 3은인구사회학적요인과건강상태요인을통제하고분석하였으며, 모형 4는인구사회학적요인, 건강상태요인, 정신건강요인을통제하고흡연여부및흡연량과

10 228 Korean J Health Promot Vol. 17, No. 4, 2017 Table 6. Odds ratio for suicidal planning according to smoking and daily smoking amount by logistic regression Variables Model 1 a Model 2 b Model 3 c Model 4 d Smoking Total Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P e <0.001 <0.001 <0.001 <0.001 Boys Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Girls Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Smoking amount, cigarettes/day Smoker ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 <0.001 < Boys ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Girls ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P Values are presented as adjusted odds ratio (95% confidence interval). Calculated by multivariate logistic regression analysis using complex sampling design. a Logistic regression for suicidal planning with smoking and smoking amount without adjustment. b Logistic regression for suicidal planning with smoking and smoking amount after adjustment for school, residence type, and economic status. c Logistic regression for suicidal planning with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, and drug experience. d Logistic regression for suicidal planning with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, drug experience, perceived stress, academic performance, perceived happiness, and sadness during the past 12 months. e P values are obtained by F-test. 자살생각, 자살계획, 자살시도와의연관성을분석하였다. 또한전체학생과남학생, 여학생을나누어분석하였다. 1) 흡연여부및흡연량과자살생각과의연관성 (Table 5) 흡연여부와자살생각과의연관성을알아보기위해요인별통제없이분석한결과비흡연자에비해과거흡연자의교차비는 1.316배 (95% confidence interval [CI] ), 현재흡연자의교차비는 1.958배 (95% CI ) 높은것으로나타났다. 여학생군에서모든요인을통제하고분석한결과과거흡연자의교차비는비흡연자에비해 1.153배 (95% CI ), 현재흡연자의교차비는 1.634배 (95% CI ) 높았다. 흡연량과의연관성은모든요인을통제하고분석한결과, 9개비이하에비해 20개비이상피우는학생의교차비가 1.612배 (95% CI ) 높게나타났으며, 특히남학생군에서 9개비이하에비해 20개비이상피우는남학생이 1.849배 (95% CI ) 높게자살생각을하는것으로나타났다. 여학생은요인별통제후분석한결과흡연량이자살생각에통계적으로유의한영향을미치지않는것으로나타났다.

11 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 229 Table 7. Odds ratio for suicidal attempt according to smoking and daily smoking amount by logistic regression Variables Model 1 a Model 2 b Model 3 c Model 4 d Smoking Total Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P e <0.001 <0.001 <0.001 <0.001 Boys Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Girls Former smoker ( ) ( ) ( ) ( ) Current smoker ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Smoking amount, cigarettes/day Smoker ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 <0.001 <0.001 <0.001 Boys ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P <0.001 < Girls ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) P < Values are presented as adjusted odds ratio (95% confidence interval). Calculated by multivariate logistic regression analysis using complex sampling design. a Logistic regression for suicidal attempt with smoking and smoking amount without adjustment. b Logistic regression for suicidal attempt with smoking and smoking amount after adjustment for school, residence type, and economic status. c Logistic regression for suicidal attempt with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, and drug experience. d Logistic regression for suicidal attempt with smoking and smoking amount after adjustment for school, residence type, economic status, perceived health, physical activity for a week, alcohol drinking, drug experience, perceived stress, academic performance, perceived happiness, and sadness during the past 12 months. e P values are obtained by F-test. 2) 흡연여부및흡연량과자살계획과의연관성 (Table 6) 흡연여부및흡연량과자살계획과의연관성을알아보기위해로지스틱회귀분석을한결과는표 6과같다. 요인별통제없이분석한결과비흡연자에비해과거흡연자의교차비는 1.403배 (95% CI ), 현재흡연자의교차비는 2.941배 (95% CI ) 높게자살계획을세운적이있는것으로나타났으며, 특히여학생군에서는비흡연자에비해과거흡연자가 2.083배 (95% CI ), 현재흡연자가 4.989배 (95% CI ) 높게자살계획을세운적이있는것으로나타났다. 모든요인을통제한후분석한결과, 전체학생군에서자살생각에대한교차비가과거흡연자는 (95% CI ), 현재흡연자는 (95% CI ) 으로나타났다. 흡연량과자 살계획과의연관성은 9개비이하에비해 개비를피우는학생군의교차비가 (95% CI ), 20개비이상피우는학생군에서의교차비가 (95% CI ) 였고, 특히남학생군에서 9개비이하를피우는남학생에비해 10-19개비를피우는남학생이 1.205배 (95% CI ), 20개비이상피우는남학생이 2.669배 (95% CI ) 높게자살계획을세운적이있는것으로나타났다. 여학생군에서는모든요인을통제한후분석한결과흡연량과자살계획이통계적으로유의한연관을나타내지않았다. 3) 흡연여부및흡연량과자살시도와의연관성 (Table 7) 모든요인을통제하고흡연여부와자살시도와의연관성을분석한결과비흡연자에비해과거흡연자가 1.211배

12 230 Korean J Health Promot Vol. 17, No. 4, 2017 Table 8. Post-hoc verification of smoking and smoking amount affecting suicidal ideation Variables N Mean (SD) P F Scheffe Smoking a<b<c Non-smoker a 56, (0.315) Former smoker b 5, (0.354) Current smoker c 4, (0.399) Smoking amount, cigarettes/day d=e<f 9 d 3, (0.384) e (0.406) 20 f (0.487) Abbreviation: SD, standard deviation. P-value <0.05. Table 9. Post-hoc verification of smoking and smoking amount affecting suicidal planning Variables N Mean (SD) P F Scheffe Smoking a<b<c Non-smoker a 56, (0.184) Former smoker b 5, (0.221) Current smoker c 4, (0.293) Smoking amount, cigarettes/day d=e<f 9 d 3, (0.265) e (0.310) 20 f (0.445) Abbreviation: SD, standard deviation. P-value <0.05. Table 10. Post-hoc verification of smoking and smoking amount affecting suicidal attempt Variables N Mean (SD) P F Scheffe Smoking a<b<c Non-smoker a 56, (0.138) Former smoker b 5, (0.173) Current smoker c 4, (0.253) Smoking amount, cigarettes/day d<e<f 9 d 3, (0.215) e (0.293) 20 f (0.423) Abbreviation: SD, standard deviation. P-value <0.05. (95% CI ), 현재흡연자가 2.215배 (95% CI ) 높게자살시도를한적이있는것으로나타났으며, 남학생군에서는각각 1.037배 (95% CI ), 1.882배 (95% CI ) 높게나타났다. 또한여학생군에서는각각 1.540배 (95% CI ), 3.070배 (95% CI ) 높게자살시도를한적이있다고나타났다. 흡연량과자살시도와의연관성을알아보기위해시행한로지스틱회귀분석결과전체학생군과남, 여학생군에서 9개비이하에비해 10-19개비, 20개비이상피우는학생이자살시도를더많이한것으로나타났다. 모든요인을통제한분석결과, 9개비이하보다 10-19개비, 20개비이상피우는학생이각각 1.549배 (95% CI ), 2.590배 (95% CI ) 높게자살시도를한적이있는것으로나타났고, 성별의차이는남학생이각각 1.702배 (95% CI ), 2.758배 (95% CI ) 높게, 여학생이각각 1.852배 (95% CI ), 2.467배 (95% CI ) 높게자살시도를한적이있는것으로나타났다. 4) 자살관련행동에영향을미치는흡연여부및흡연량에대한사후검정 (Tables 8-10) 다변량로지스틱회귀분석을통해흡연여부및흡연량이자살관련행동에영향을미치는요인임을알수있었다. 이후각각변수간의차이를검정하기위해일원분산분석을시행하고, Scheffe 사후검정을통해변수간대소를비교하였다. 자살관련행동에영향을미치는흡연여부변수인비흡연자, 과거흡연자, 현재흡연자간의차이를검정한결과에서모두비흡연자보다과거흡연자가높게나타났으며, 과거흡연자보다현재흡연자가높게나타났다. 자살생각, 자살계획에영향을미치는흡연량은 9개비이하군과 10-19개

13 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey 231 비군에서는차이가없었고, 20개비이상군은이군들보다높게나타났다. 자살시도에영향을미치는흡연량의크기비교를위한사후검정에서는 9개비이하군보다 10-19개비군이높게나타났고, 개비군보다 20개비이상군이높게나타나차이를보였다. 고 찰 본연구는 2016년제12차청소년건강행태온라인조사자료를이용하여청소년의흡연여부및흡연량과자살관련행동과의연관성을알아본연구로, 청소년사망의주요원인인자살에영향을미치는요인에대해분석하여자살에대한예방전략을수립하고자하였다. 본연구의결과에서는청소년의흡연은자살관련행동에영향을미치는요인이며, 또한흡연량이자살관련행동에영향을주는인자라는것을알수있었다. 흡연과자살생각과의관계를살펴보면, 흡연을하는학생중자살생각을한적이있는학생은 20.0% 였고, 흡연을하지않는학생중자살생각을한적이있는학생은 11.3% 로, 흡연을하는학생이자살생각을한적이있는경우가흡연을하지않는학생보다높았다. 또한, 본연구에서는여러인구사회학적요인, 건강상태요인, 정신건강요인들을통제한후청소년의흡연이자살생각에미치는영향을분석하였고, 그결과흡연을하는학생은비흡연자에비해자살생각이 1.143배높은것으로나타났으며, 특히여학생은 배높게자살생각을하는것으로나타났다. 이러한결과는남학생보다여학생에서흡연이자살생각이미치는영향이더높은것을볼수있었고, 이는선행연구의결과와도일치하였다. 5,23-26) 흡연은자살계획과도유의하게연관성이있었다. 현재흡연을하고있는학생은비흡연자에비해자살계획을세운경험이 1.646배높았고, 남학생은 1.405배, 여학생은 배높은것으로나타났다. 흡연과자살시도와의연관성을살펴보면, 현재흡연을하고있는학생은흡연을하지않는학생에비해자살시도를한경험이 2.215배높음을알수있었고, 특히여학생의경우 3.070배높게자살시도를한적이있는것으로나타났다. 쌍둥이아버지와쌍둥이자녀만을대상으로 26년간흡연과자살관련을추적조사한연구에서도현재흡연자이거나규칙적인흡연자는자살생각과자살계획, 자살시도가남자는 6.59배, 여자는 3.37배유의하게높은것으로나타나 18) 본연구결과와유사하였다. 흡연량과자살생각과의연관성에대해살펴보면, 9개비이하로흡연하는남학생에비해 20개비이상흡연하는남학생은자살생각을 1.849배높게하는것을알수있었다. 이는 18-20세에흡연한대상자들을 26년동안추적조사한 결과, 흡연량이증가함에따라자살위험성이증가하고, 특히하루에 20개비이상흡연하는군의경우, 초반 13년동안의추적에서자살위험성이 3.3배크게증가하는것으로나타난이전의연구와도유사한결과였다. 15) 또한, 정신과성인환자에서남성의경우흡연량이증가함에따라자살생각이증가하는결과를보여주는국내연구 27) 와도유사한경향을보였다. 하지만여학생은흡연량과자살생각과의관계가모든혼란변수들을통제한후분석한결과에서통계적으로유의하지않음을보였다. 이는여학생들은흡연이자살생각에영향을미치지만, 흡연량은영향을미치지않는것으로볼수있으며, 이전의연구에서여자청소년들은남자들에비해분노를적절히분출할수있는통로가적어흡연행동이공격성이나분노를표현할수있는수동적방법이라여겨진다고하였는데, 2) 여학생의경우흡연량과는상관없이흡연을하는행위자체에서자살생각의위험성이높아질수있음을유추해볼수있고, 따라서이러한차이는남학생과여학생의특성을고려한차별화된금연교육및자살예방프로그램의필요성을설명할수있을것으로보인다. 흡연량이자살계획과자살시도에미치는영향에대해살펴보면, 매일 9개비이하흡연하는학생에비해 10-19개비, 20개비이상흡연하는학생은각각 배, 배높게자살계획을세우는것을볼수있었고, 각각 1.549배, 2.590배높게자살시도를하는것으로나타났다. 이는 18세이하정신질환치료를받고있는흡연자와자살시도와의관계를알아보고자시행한연구결과에서비흡연자군보다하루 15개비이상흡연하는군에서자살시도경험발생률이 2.76배증가한다고나타난이전의연구와도유사한결과이다. 28) 본연구에서도남학생과여학생의결과에차이는있지만, 흡연량이자살계획, 자살시도에영향을미치는요인으로분석되므로, 자살생각의다음단계인자살계획, 자살시도를한적이있는청소년이현재흡연자일경우에는흡연량에대한접근방법또한자살예방교육및금연교육의연구설계에의미있게도입되어야할것으로보인다. 본연구의제한점으로는자기기입식으로조사되어, 흡연, 음주, 약물과같은민감한질문에정직하게응답하지않았을가능성이있다. 그리고청소년건강행태온라인조사의문항으로는흡연경험및흡연량과최근 1년간의자살관련행동과의인과관계를충분히설명하지못하는단면연구의한계가있다. 또한종전의연구에서도자살생각, 자살계획, 자살시도에영향을미치는요인들에대해각각살펴보았지만세변수간의관계를나타내지못한제한점이있었고, 2) 본연구에서도세변수간의관계를파악하지는못하여이에대한추가적인연구가필요할것으로보인다. 하지만본연구는청소년의흡연여부와흡연량은자살관련행동에영향을미

14 232 Korean J Health Promot Vol. 17, No. 4, 2017 치는주요요인임을알수있었다. 이결과를고려하여청소년의자살예방프로그램에금연교육및흡연예방교육이함께이루어지고, 남 / 여학생의특성을고려한차별화된금연및자살예방프로그램이만들어져야할것이다. 요 약 연구배경 : 자살은우리나라 10대인구사망원인중 1위로알려져있으며, 흡연은청소년자살의위험요인중하나로보고되고있다. 본연구는청소년건강행태온라인조사를이용하여흡연및흡연량과자살관련행동과의연관성을알아보고자한다. 방법 : 2016년제12차청소년건강행태온라인조사를이용하여, 청소년의흡연여부및흡연량과자살생각, 자살계획, 자살시도여부를알아보았다. 조사참여자인 65,528명을대상으로현재흡연자, 과거흡연자, 비흡연자로나누고, 현재흡연자 4,012명을흡연량에따라 9개비이하군, 개비군, 20개비이상군으로분류하였고, 자살생각, 자살계획, 자살시도와의연관성을다변량로지스틱회귀분석을사용하여알아보았다. 결과 : 모든혼란변수들을통제하여분석한결과, 현재흡연자는비흡연자, 과거흡연자에비해자살생각과의교차비가 (CI ) 으로나타났으며, 자살계획, 자살시도의교차비도통계적으로유의하게높게나타났다. 또한 9개비이하군보다 20개비이상군에서자살관련행동의교차비는모두각각유의하게높게나타났다. 결론 : 본연구를통해청소년의흡연과흡연량은자살생각, 자살계획, 자살시도에영향을주는요인임을알수있었고, 이결과를고려한자살예방전략수립이필요할것이다. 중심단어 : 한국, 청소년, 자살, 행동, 흡연 REFERENCES 1. Korea National Statistical Office Annual report on the cause of death statistics [Internet]. Daejeon: Korea National Statistical Office; [Accessed Jun 28, 2017]. Available from: 2. Sohn SY. Factors affecting suicidal ideation, suicidal plan and suicidal attempt in Korean adolescents. J Korea Acad Industr Coop Soc 2014;15(3): Kim HS, Kim BS. A comparative study on suicidal ideation in the elderly and the adolescents. J Korean Gerontol Soc 2008; 28(2): Jeon HS. Effects of smoking, drinking, and drug use on the adolescent's suicidal ideation by using the data of the Korea youth risk behavior web-based survey through from 2008 to J Korean Soc Sch Health 2015;28(2): Ladame F, Jeanneret O. Suicide in adolescence: some comments on epidemiology and prevention. J Adolesc 1982;5(4): Hagh YH. A study on actural condition of junior high school students' suicidal ideation [dissertation]. Seoul: Ewha Womans University; Korean. 7. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-analysis. Br J Psychiatry 1997;170(3): Park H. Effect of sleep duration on suicidal ideation in Korean adolescents. J Korean Soc Sch Health 2015;28(1): Runeson BS, Beskow J, Waern M. The suicidal process in suicides among young people. Acta Psychiatr Scand 1996;93(1): Bettes BA, Walker E. Symptoms associated with suicidal behavior in childhood and adolescence. J Abnorm Child Psychol 1986;14(4): Kim HS, Kim BS. Verification of the structural relationship model of suicidal ideation to its related variables. Korean J Counseling Psychotherapy 2008;20(2): Kim BY, Lee CS. A meta-analysis of variables related to suicidal ideation in adolescents. J Korean Acad Nurs 2009;39(5): Wo C, Kim P. The multidimensional structure analysis of variables related to adolescents' suicidal ideation. Korean J Youth Studies 2011;18(4): Chin YR, Lee HY, So ES. Suicidal ideation and associated factors by sex in Korean adults: a population-based cross-sectional survey. Int J Public Health 2011;56(4): Taylor A, Dal Grande E, Gill T, Fisher L, Goldney R. Detecting determinants of suicidal ideation: South Australian surveillance system results. Int J Public Health 2007;52(3): Hemmingsson T, Kriebel D. Smoking at age and suicide during 26 years of follow-up-how can the association be explained? Int J Epidemiol 2003;32(6): DeBry SC, Tiffany ST. Tobacco-induced neurotoxicity of adolescent cognitive development (TINACD): a proposed model for the development of impulsivity in nicotine dependence. Nicotine Tob Res 2008;10(1): Scherrer JF, Grant JD, Agrawal A, Madden PA, Fu Q, Jacob T, et al. Suicidal behavior, smoking, and familial vulnerability. Nicotine Tob Res 2011;14(4): Kessler RC, Berglund PA, Borges G, Castilla-Puentes RC, Glantz MD, Jaeger SA, et al. Smoking and suicidal behaviors in the National Comorbidity Survey: replication. J Nerv Ment Dis 2007;195(5): Hemenway D, Solnick SJ, Colditz GA. Smoking and suicide among nurses. Am J Public Health 1993;83(2): Ministry of Education, Ministry of Health & Welfare, Korea Centers for Disease Control and Prevention. The 12th Korea youth risk behavior web-based survey 2016 [Internet]. Sejong: Ministry of Education, Ministry of Health & Welfare, Korea Centers for Disease Control and Prevention; [Accessed Nov 7, 2016]. Available from: new/pages/pds1.asp. 22. Chen M, Wang EK, Jeng YJ. Adequate sleep among adolescents is positively associated with health status and health-related behaviors. BMC Public Health 2006;6: Cho SJ, Jeon HJ, Kim JK, Suh TW, Kim SU, Hahm BJ, et al. Prevalence of suicide behaviors (suicidal ideation and suicide attempt) and risk factors of suicide attempts in junior and high school adolescents. J Korean Neuropsychiatr Assoc 2002;41(6):

15 Hye Won Kim, et al. The Relationship between Smoking and Suicidal Behavior in Korean Adolescents: 12th Korea Youth Risk Behavior Web-based Survey Woo CY, Park AC, Jeong HH. The structural relationship among human relations, stress, depression and suicidal ideation of adolescents by genders and level of school. Korean J Educ Psychol 2010;24(24): Park E. The influencing factors on suicide attempt among adolescents in South Korea. J Korean Acad Nurs 2008;38(3): Kwak SJ, Yi YJ, Jung HS. The analysis on factors related to suicide ideation of middle and high school students in Korea. J Korean Soc Sch Health 2009;22(2): Oh SM, Sohn JH, Chang SM, Seong SJ, Lee JY, Bae A, et al. Correlation between tobacco smoking and mental disorders including suicidal tendencies. J Korean Neuropsychiatr Assoc 2011;50(5): Riala K, Viilo K, Hakko H, Räsänen P; STUDY-Seventy Research Group. Heavy daily smoking among under 18-year-old psychiatric inpatients is associated with increased risk for suicide attempts. Eur Psychiatry 2007;22(4):