연구목적 : 본 연구는 지역사회건강조사를 통해 파악된 우리나라 주요 건강행태(흡연, 음주, 신체활동, 구강건강) 및 건강수준(고혈압 및 당뇨병, 심뇌혈관질환 사망률)의 시군구별 변이를 초래하는 관련요인을 규명하고자 수행하였다. 연구방법 : 연구단위는 2009년 기준 보건소가 설치된 253개 시군구이며, 지역사회건강조사, 통계청, 지방자치단체 기초통계연보 등 2차 자료를 통해 지역특성 DB를 구축하였다. 지역간 변이요인에 대한 규명은 생태학적 분석(선형회귀분석)을 통해 수행하였다. 연구결과 : `남자현재흡연율`은 `지역박탈지수`, `조이혼율`, `인구천명당주점수`, `체질량지수`, `월간음주율`과 양의 관련성을 `고교졸업률`, `주민대상금연교육수`와 음의 관련성을 보였다. `월간음주율`은 `현재흡연율`, `비만율`, `우울감경험률`, `월소득`이 높을수록, `인구천명당주점수`와 `인구1인당공원면적`이 낮을수록 높았다. `1인당자동자등록대수`가 많을수록 `걷기실천율`이 낮았다. 노인의 `저작불편호소율`은 `지역박탈지수`, `주관적구강건강상태`, `경제적이유로인한 치과진료미수진율`, `인구 십만명당치과병원수`로 나타났다. `고위험음주율`, `비만율`, `스트레스인지율`, `노인인구비율`이 높은 지역에서, 그리고 `기초생활수급자율`, `중등도신체활동실천율`이 높을수록 `고혈압평생의사진단경험률`이 낮았다. 또한 `당뇨병평생의사진단경험률`은 `기초생활수급자율`, `고혈압평생의사진단경험률`과 양의 관련성을, `검강검진수검률`과는 음의 관련성을 보였다. `현재흡연률`, `재정자주도`, `실업률`은 `심장질환사망률` 및 `뇌혈관질환사망률`과 모두 관련성이 있었고, 그 밖에 `고혈압평생의사진단 경험률`과 `당뇨병평생의사진단경험률`은 `심장질환사망률`과 `성비`는 `뇌혈관질환사망률`과 관련성을 보였다. 마지막으로 `사고손상경험률`은 도시지역은 `가구균등화소득`, `필요의료서비스미치료율`, `비만율`, 농촌지역은 `노인인구비율`, `투표율`, `우울감경험률`과 관련이 있었다. 특히 `지역박탈지수`는 관련요인을 통제 후 도시지역에서만 `사고손상경험률`과 관련성을 나타냈다. 결론 : 본 연구는 주요 건강행태 및 건강수준의 지역간 변이를 줄이는 정책 마련에 근거자료로 활용할 수 있을 것이다.

Objective : This study was performed to investigate the magnitude and pattern of small area variations defined by Si-gun-gu and to explore the factors related to these variations in the prevalences of smoking, drinking, physical activity, oral health, hypertension and diabetes mellitus (DM) management, and cardiovascular death rates, using 2009 Community health survey (CHS) data. Methods : The unit of analysis was 253 counties surveyed at 2009 CHS. Analysis dataset was constructed of secondary data extracted from Korean Statistical Information Service (KOSIS) and data from CHS. Magnitude and pattern of area variations in the given health behaviors and health status were examined with Extremal Quotient (EQ) and Coefficients of Variation (CV) of Small Area Variation Analysis (SAVA). Ecologic analysis using linear regression model was used to explore the related factors with these variations. Results : In all dependent variables, the variations between areas were statistically significant. Variations in smoking rates appear to be positively related with several variables, such as regional deprivation index, crude divorce rate, obesity rate, and number of bars per 1,000 person, and negatively with level of educational attainment, number of smoking cessation program in that area. The variables related with high monthly drinking rate were number of bars per 1,000 person, higher budget for welfare, and less park area (m2) per person. As the number of registered motor vehicles per person increases, the obesity rates became higher and rates of walking exercise decreased. The related factors with rates of physician diagnosed hypertension were proportion of elderly people aged 65 or over and of those with basic needs. Age-standardized cardiovascular disease death rates were found to be related with unemployment rate and low financial independence of that area. The variables related with total injury experience rate were monthly household equalization income, Unmeet health care need, obesity in urban, and Elderly percent, Vote rate, Depressive symptom in rural. In particular, regional deprivation index was associated only in urban after adjusting related factors. Conclusions : These results will help us to set up policy development for population-based health promotion through reducing the gap of health-related indices between area. Further researches are needed to maximize academic value of CHS data.