데이터 마이닝을 이용한 당뇨환자의 관리요인에 관한 연구

Abstract

본 연구의 목적은 당뇨환자 관리와 관련된 요인을 규명하는데 있다. 2005년 국민건강 영양조사에 참여한 20세 이상의 성인 당뇨환자를 대상으로 하였다. 데이터마이닝 기법을 이용하여 로지스틱 회귀모형, 의사결정나무, 신경망 모형으로 당뇨환자관리모형을 개발한 결과 의사결정나무가 가장 설명력이 뛰어났다. 당뇨인지율과 관련된 요인으로는 연령, 거주지 및 직업이었고 중 연령이 가장 중요한 요인으로 나타났다. 당뇨치료율과 관련된 요인으로는 당뇨인지여부, 거주지 및 직업이었고 그 중 당뇨인지여부가 가장 중요한 변수로 나타났다. 당뇨환자의 관리프로그램은 당뇨환자의 특성별 군집으로 분류하고 그에 따라 관리해야 한다. The Objectives: The purpose of this study is to identify the factors related to management of DM in Korea. Methods: The subjects selected by using data of National Health and Nutrition Survey(NHANS) in 2005 were 415 adults, aged 20 and older, and diagnosed with DM. This study used data mining algorithms. This study validated the predictive power of data mining algorithms by comparing the performance of logistic regression, decision tree, and Neural Network on the basic of validation, it was found that the model performance of decision tree was the best among the above three techniques. Result: First, awareness of DM was positively associated with age, residential area, and job. The most important factor of DM awareness is age. Awareness rate of DM with 52 age over is 76.1%. Among the <TEX>${\geq}52$</TEX> age group, an important factor is family history. Among patients who are 52 years or over with family history of DM, an important factor is job. The awareness rate of patients who are 52 age over, family, history of DM, and professionals is 95.0%. Second, treatment of DM was also positively associated with awareness, region, and job. The most important factor of DM treatment is DM awareness. Treatment rate of patients who are aware of DM is 84.8%. Among patients who have awareness of DM, an important factor is region. The awareness rate of patients who are aware of DM in rural area is 10.4%. Conclusion: Finally, the result of analysis suggest that DM management programs should consider group characteristic of DM patients.