빅데이터를 이용한 대기오염의 건강영향 평가 및 피해비용 추정 = Assessment of Human Health Effects of Air-Pollution Using Cohort DB and Estimation of Economic Costs in Korea(1)

다국어 초록 (Multilingual Abstract)

This study aims to assess the long-term effects of air-pollution on human health using cohort DB provided by the National Health Insurance Service and to estimate its economic costs in Korea. The ultimate goal is to construct an policy-evaluation framework based on the results from the assessment of health effects and economic costs. An integrated policy assessment framework is composed of 4 steps as illustrated in Figure 1. Step 1 is scoping stage where the policy scenarios are developed, which includes determining the end-points of health effects and population to be affected. Step 2 is health risk assessment. Our focus is the long-term effects of exposures to 5 major air pollutants . PM10, CO, O3, NO2 and SO2 . on respiratory diseases. The health risks are evaluated by estimating concentration-response (CR) functions. Individual CR functions are estimated depending on air-pollutants, provinces, and age groups. The results of health risks are presented in terms of the changes in mortality and morbidity associated with policy scenarios. Step 3 is monetary valuation of human health effects. Monetary valuation begins with estimating unit-values for health end-points which are mortality and morbidity in our cases. The unit values for mortality and morbidity are estimated with value of statistical life (VSL) and cost of illness (COI), respectively. The total economic costs are calculated, in turn, by multiplying the changes in mortality and morbidity by corresponding unit values. The final step is synthesis. In this step, uncertainty issues are considered and sensitivity analyses are conducted. In addition, qualitative information which cannot be incorporated in the procedure is specified for further consideration in the assessment. Although the proposed policy-evaluation framework is explained as a step-by-step procedure, working operation of the project is rather parallel and emphasizes the importance of feedbacks between the steps. Three-year project begins with cleaning and re-constructing cohort DB for the assessment of long-term health effects from exposure to 5 selected air-pollutants. The first-year (2015) preliminary results show that the health risks, measured by new hospital admissions of respiratory diseases using the time-varying survival analysis, are statistically higher in Seoul than nation-wide for all of the 5 pollutants and higher for the age group 15 and younger than the pooled group for CO, NO2 and SO2. In addition the health risks, measured by repetitive hospital admissions of respiratory diseases using panel data, are relatively higher for the age group 65 and older than the pooled group for PM10, CO, O3, and SO2. The COI for the respiratory diseases are calculated by including direct medical expenditures covered by insurance and individuals, personal out-of-pocket expenses such as transportation costs and costs of employing nursing services, and opportunity costs associated with lost productivity due to the illness. The main data sources are the cohort DB provided by the National Health Insurance Service and the Korea Health Panel Survey. In 2013, the total COI is estimated as 3,766 thousand won per hospital-admission episode of respiratory diseases where direct medical expenditures, personal out-of-pocket expenses, and opportunity costs associated with lost productivity comprise 51.1%, 15.2%, and 33.7%, respectively. Based on the preliminary results, study is planned to continue in the second year (2016) to refine the assessment of health effects by incorporating demographic and socio-economic variables in estimating CR functions. Estimation of VSL in Korean context using the stated preference method will be added to the economic analysis.