경쟁위험모형을 이용한 뇌졸중 환자의 재발 및 사망 비교

Abstract

Background and Purpose : According to the World Health Organization (WHO), stroke is defined as a condition in which a local neurological disorder or a consciousness disorder resulting from cerebral vascular occlusion (cerebral infarction, cerebral ischemia), rupture of the cerebral blood vessels (cerebral hemorrhage) occurs and lasts for more than 24 hours. Stroke is often acute, and recurrence is a common disease. In addition, there are high risks such as increased risk of stroke due to Comorbidity such as hypertension and diabetes, increased cost of health insurance. Therefore, it is not only the prevention of stroke prevention first but also the second prevention after cerebral stroke. Although there are many studies on stroke (cerebral infarction, cerebral hemorrhage), there are still few studies using competing risk analysis. This study aims to contribute to the improvement of the national health by analyzing the recurrence and death of stroke in the stroke patients using the competing risk analysis method. Subject and Method : The data used in this study was compiled from the income quintile National Health Insurance Corporation sample cohort data from 2002 to 2013 for the case of hospitalization with stroke. When the first hospitalization is cerebral infarction, the variables of interest are recurrence and death of cerebral infarction, and the risk of competition is cerebral hemorrhage and other disease recurrence and death. When the first hospitalization is cerebral hemorrhage, the variables of interest are recurrence or death of the hemorrhagic stroke, and the risk of competition is cerebral infarction and recurrence of other diseases and death. All variables were categorical, sex, age> 40 years, income quintile, CCI index, hypertension and diabetes diagnosis, severity of disability for one year before and after stroke. The concept of landmark was used to exclude cases of hospitalization and outpatient treatment for stroke for one year before and after the first admission. Cause-specific hazard model and subdistribution hazard model were used for competing risk analysis. Results : First, in the results of competition risk analysis, when the variables of interest were relapse of cerebral infarction, In males, the income quintile was in the 0-2 quartile, and the severity of the disorder was elevated from normal to severe, with a higher risk of recurrent cerebral infarction than recurrent cerebral hemorrhage or other diseases. Especially, the risk of recurrence of cerebral infarction was high in the age group, 6.8 times in the 70s compared to 40s, and 8.1 times in the 80s and over. Second, When the variables of interest were recurrence of cerebral hemorrhage, the risk of relapse of cerebral hemorrhage was 2.3 times higher in women and 2.9 times higher in patients with severe disorder than in those with cerebral infarction or recurrence of other diseases. Third, When the variable of interest is the ischemic death, the risk of cerebral infarction is high, about 3.7 times greater than the age of 50s, about 10 times greater than that of the 70s, and about 14 times greater than the age of 80s. The CCI index has a higher risk of cerebral infarction death than 1 or 3 or more than cerebral hemorrhage or other disease deaths. Fourth, when the variable of interest is cerebral hemorrhage death, the risk of cerebral hemorrhage was higher than that of cerebral infarction or other diseases, about 1.1 times in the 60s, about 6 times in the 70s, and 10 times in the 80s. Cause-specific hazard model and subdistribution hazard model were used. The results of both models showed similar results with the exception of moderate to severity of disability of cerebral hemorrhage Conclusion : In this study, we conducted a competition risk analysis using the landmark concept on recurrence and death of cerebral infarction using the National Health Insurance sample cohort. There were statistically significant differences in sex, age, income level, CCI Index, and severity of disability for the variables of interest. In the case of cerebral hemorrhage, there was statistically significant difference only in sex and age variables. Hypertension and diabetes, known as risk factors for stroke, were not statistically significant in both stroke recurrence and death. Based on the results of this study, it is expected that various landmark time for stroke research will be analyzed by analyzing various variables and clinically meaningful points for stroke research, and further studies using competing risk analysis will be conducted.