Health Insurance Review and Assessment Service

Health Insurance and Review Assessment (HIRA) in South Korea, also called National Health Insurance (NHI) data, is a repository of claims data collected in the process of reimbursing healthcare providers. Under the universal coverage system, having fee-for-services covering all citizens in South Korea, HIRA contains comprehensive and rich information pertaining to healthcare services such as treatments, pharmaceuticals, procedures, and diagnoses for almost 50 million beneficiaries. This corpus of HIRA data, which constitutes a large repository of data in the healthcare sector, has enormous potential to create value in several ways: enhancing the efficiency of the healthcare delivery system without compromising quality of care; adding supporting evidence for a given intervention; and providing the information needed to prevent (or monitor) adverse events. In order to actualize this potential, HIRA data need to actively be utilized for research. Thus understanding this data would greatly enhance this potential. We introduce HIRA data as an important source for health research and provide guidelines for researchers who are currently utilizing HIRA, or interested in doing so, to answer their research questions. We present the characteristics and structure of HIRA data. We discuss strengths and limitations that should be considered in conducting research with HIRA data and suggest strategies for optimal utilization of HIRA data by reviewing published research using HIRA data.

The claims data of the Health Insurance Review and Assessment Service (HIRA) is an important source of information for healthcare service research. The claims data of HIRA is collected when healthcare service providers submit a claim to HIRA to be reimbursed for a service that they provided to patients. To improve the accessibility of healthcare service researchers to claims data of HIRA, HIRA has developed the Patient Samples which are extracted using a stratified randomized sampling method. The Patient Samples of HIRA consist of five tables: a table for general information (Table 20) containing socio-demographic information such as gender, age and medical aid, indicators for inpatient and outpatient services; a table for specific information on healthcare services provided (Table 30); a table for diagnostic information (Table 40); a table for outpatient prescriptions (Table 53) and a table for information on healthcare service providers (Table of providers). Researchers who are interested in using the Patient Sample data for research can apply via HIRA's website (https://www.hira.or.kr).

OBJECTIVE: To review and update the conceptual framework, indicator content and research priorities of the Organisation for Economic Cooperation and Development's (OECD) Health Care Quality Indicators (HCQI) project, after a decade of collaborative work. DESIGN: A structured assessment was carried out using a modified Delphi approach, followed by a consensus meeting, to assess the suite of HCQI for international comparisons, agree on revisions to the original framework and set priorities for research and development. SETTING: International group of countries participating to OECD projects. PARTICIPANTS: Members of the OECD HCQI expert group. RESULTS: A reference matrix, based on a revised performance framework, was used to map and assess all seventy HCQI routinely calculated by the OECD expert group. A total of 21 indicators were agreed to be excluded, due to the following concerns: (i) relevance, (ii) international comparability, particularly where heterogeneous coding practices might induce bias, (iii) feasibility, when the number of countries able to report was limited and the added value did not justify sustained effort and (iv) actionability, for indicators that were unlikely to improve on the basis of targeted policy interventions. CONCLUSIONS: The revised OECD framework for HCQI represents a new milestone of a long-standing international collaboration among a group of countries committed to building common ground for performance measurement. The expert group believes that the continuation of this work is paramount to provide decision makers with a validated toolbox to directly act on quality improvement strategies.

STUDY DESIGN: Retrospective cohort study using national health insurance data. OBJECTIVE: To provide a longitudinal reoperation rate after surgery for lumbar herniated intervertebral disc (HIVD) disease, and to compare the reoperation rates of surgical methods. SUMMARY OF BACKGROUND DATA: Herniated intervertebral disc disease is the most common cause of lumbar spinal surgery. Despite improved surgical techniques and instrumentation, reoperation cannot be avoided. The reoperation rates were in the range of 6% to 24% in previous studies. A population-based study is less subject to bias; hence, a nationwide longitudinal analysis was warranted. METHODS: A national health insurance database was used to identify a cohort of patients who underwent first surgery for herniated intervertebral disc disease in 2003 and 18,590 patients were selected. Individual patients were followed for at least 5 years through their encrypted unique resident registration number. The primary endpoint was any type of second lumbar surgery. After adjusting for confounding factors, 5 surgical methods (fusion, laminectomy, open discectomy, endoscopic discectomy, and nucleolysis [including mechanical nucleus decompression]) were compared. Open discectomy was used as the reference method. RESULTS: Open discectomy was the most common procedure (68.9%) followed by endoscopic discectomy (16.1%), laminectomy (7.9%), fusion (3.9%), and nucleolysis (3.2%). The cumulative reoperation rate was 5.4% at 3 months, 7.4% at 1 year, 9% at 2 years, 10.5% at 3 years, 12.1% at 4 years, and 13.4% at 5 years. The reoperation rates were 18.6%, 14.7%, 13.8%, 12.4%, and 11.8% after laminectomy, nucleolysis, open discectomy, endoscopic discectomy, and fusion, respectively. Compared with open discectomy, the reoperation rate was higher after laminectomy at 3 months, whereas the other surgical methods had similar rates. CONCLUSION: The cumulative reoperation rate after 5 years was 13.4% and half of the reoperations occurred during the first postoperative year. With the exception of laminectomy, the reoperation rates of the other procedures were not different from that of open discectomy.

OBJECTIVE: The authors assessed the relationship between exposure to ambient particulate matter and suicide in urban settings during a 1-year period. METHOD: The association between particulate matter and suicide was determined using a time-stratified case-crossover approach in which subjects served as their own controls. All suicide cases (4,341) in 2004 that occurred in seven cities in the Republic of Korea were included. Hourly mean concentrations of particulate matter < or =10 microm in aerodynamic diameter (at 106 sites in the seven cities) and particulate matter < or =2.5 microm in aerodynamic diameter (at 13 sites in one city) were measured. The percent increase in suicide risk associated with an interquartile range increase in particulate matter was determined by conditional logistic regression analysis after adjusting for national holidays and meteorological factors. Subgroup analysis was performed after stratification by underlying disease (cardiovascular disease, diabetes mellitus, chronic obstructive pulmonary disease, cancer, and psychiatric illness). RESULTS: The largest associations were a 9.0% increase (95% CI=2.4-16.1) and a 10.1% (95% CI=2.0-19.0) increase in suicide risk related to an interquartile range increase in particulate matter < or =10 microm (average of 0 to 2 days prior to the day of suicide) and particulate matter < or =2.5 microm (1 day prior to the day of suicide), respectively. Among individuals with cardiovascular disease, a significant association between particulate matter < or =10 microm (average of 0 to 2 days prior to the day of suicide) and suicide was observed (18.9%; 95% CI=3.2-37.0). CONCLUSIONS: Conclusions: A transient increase in particulate matter was associated with increased suicide risk, especially for individuals with preexisting cardiovascular disease.

Importance: Current guidelines recommend ticagrelor as the preferred P2Y12 platelet inhibitor for patients with acute coronary syndrome (ACS), primarily based on a single large randomized clinical trial. The benefits and risks associated with ticagrelor vs clopidogrel in routine practice merits attention. Objective: To determine the association of ticagrelor vs clopidogrel with ischemic and hemorrhagic events in patients undergoing percutaneous coronary intervention (PCI) for ACS in clinical practice. Design, Setting, and Participants: A retrospective cohort study of patients with ACS who underwent PCI and received ticagrelor or clopidogrel was conducted using 2 United States electronic health record-based databases and 1 nationwide South Korean database from November 2011 to March 2019. Patients were matched using a large-scale propensity score algorithm, and the date of final follow-up was March 2019. Exposures: Ticagrelor vs clopidogrel. Main Outcomes and Measures: The primary end point was net adverse clinical events (NACE) at 12 months, composed of ischemic events (recurrent myocardial infarction, revascularization, or ischemic stroke) and hemorrhagic events (hemorrhagic stroke or gastrointestinal bleeding). Secondary outcomes included NACE or mortality, all-cause mortality, ischemic events, hemorrhagic events, individual components of the primary outcome, and dyspnea at 12 months. The database-level hazard ratios (HRs) were pooled to calculate summary HRs by random-effects meta-analysis. Results: After propensity score matching among 31 290 propensity-matched pairs (median age group, 60-64 years; 29.3% women), 95.5% of patients took aspirin together with ticagrelor or clopidogrel. The 1-year risk of NACE was not significantly different between ticagrelor and clopidogrel (15.1% [3484/23 116 person-years] vs 14.6% [3290/22 587 person-years]; summary HR, 1.05 [95% CI, 1.00-1.10]; P = .06). There was also no significant difference in the risk of all-cause mortality (2.0% for ticagrelor vs 2.1% for clopidogrel; summary HR, 0.97 [95% CI, 0.81-1.16]; P = .74) or ischemic events (13.5% for ticagrelor vs 13.4% for clopidogrel; summary HR, 1.03 [95% CI, 0.98-1.08]; P = .32). The risks of hemorrhagic events (2.1% for ticagrelor vs 1.6% for clopidogrel; summary HR, 1.35 [95% CI, 1.13-1.61]; P = .001) and dyspnea (27.3% for ticagrelor vs 22.6% for clopidogrel; summary HR, 1.21 [95% CI, 1.17-1.26]; P < .001) were significantly higher in the ticagrelor group. Conclusions and Relevance: Among patients with ACS who underwent PCI in routine clinical practice, ticagrelor, compared with clopidogrel, was not associated with significant difference in the risk of NACE at 12 months. Because the possibility of unmeasured confounders cannot be excluded, further research is needed to determine whether ticagrelor is more effective than clopidogrel in this setting.

The authors evaluated the incidence of hip fracture and subsequent mortality in Korea using nationwide data obtained from the Health Insurance Review and Assessment Service. This study was performed on patient population, aged 50-yr or older who underwent surgical procedures because of hip fracture (ICD10; S720, S721). All patients were followed using patient identification code to identify deaths. Crude hip fracture rates increased from 191.9/100,000 in 2005 to 207.0/100,000 in 2008 in women and from 94.8/100,000 in 2005 to 97.8/100,000 in 2008, in men respectively. Crude mortality within 12 months after hip fracture showed a similar trend (18.8% in 2005 and 17.8% in 2007). The mean of standardized mortality ratio of hip fracture was 6.1 at 3 months, 3.5 at 1 yr, and 2.3 at 2 yr post-fracture. The increasing incidence and the high mortality after hip fracture are likely to become serious public health problems and a public health program should begin to prevent hip fractures in Korea.

Financing and provision of long-term care is an increasingly important concern for many middle-income countries experiencing rapid population aging. We examine three countries (South Korea, Japan, and Germany) that use social insurance to finance medical care and have developed long-term care insurance (LTCI) systems. These countries have adopted different approaches to LTCI design within the social insurance framework. We contrast their financing systems and draw lessons regarding revenue generation, benefits design, and eligibility. Based on this review, it seems important for middle-income countries to start developing LTCI schemes early, before aging becomes a significant problem and substantial revenues are needed. Early financing also ensures that the service delivery system has time to adapt because most middle-income countries lack the infrastructure for providing long-term care services. One approach is to start with a limited benefit package and strict eligibility rules and expanded the program as the country develops sufficient experience and more providers became available. All three countries use some form of cost-sharing to discourage service overuse, combined with subsidies for poor populations to maintain appropriate access. A major policy choice is between cash benefits or direct provision of services and the approach will have a large impact on the workforce participation of women.

BACKGROUND: The lack of epidemiologic information on osteoporotic hip fractures hampers the development of preventive or curative measures against osteoporosis in South Korea. We conducted a population-based study to estimate the annual incidence of hip fractures. Also, we examined factors associated with post-fracture mortality among Korean elderly to evaluate the impact of osteoporosis on our society and to identify high-risk populations. METHODS: The Korean National Health Insurance (NHI) claims database was used to identify the incidence of hip fractures, defined as patients having a claim record with a diagnosis of hip fracture and a hip fracture-related operation during 2003. The 6-month period prior to 2003 was set as a 'window period,' such that patients were defined as incident cases only if their first record of fracture was observed after the window period. Cox's proportional hazards model was used to investigate the relationship between survival time and baseline patient and provider characteristics available from the NHI data. RESULTS: The age-standardized annual incidence rate of hip fractures requiring operation over 50 years of age was 146.38 per 100,000 women and 61.72 per 100,000 men, yielding a female to male ratio of 2.37. The 1-year mortality was 16.55%, which is 2.85 times higher than the mortality rate for the general population (5.8%) in this age group. The risk of post-fracture mortality at one year is significantly higher for males and for persons having lower socioeconomic status, living in places other than the capital city, not taking anti-osteoporosis pharmacologic therapy following fracture, or receiving fracture-associated operations from more advanced hospitals such as general or tertiary hospitals. CONCLUSION: This national epidemiological study will help raise awareness of osteoporotic hip fractures among the elderly population and hopefully motivate public health policy makers to develop effective national prevention strategies against osteoporosis to prevent hip fractures.

BACKGROUND: Research evidence suggests that nurse staffing influences patient outcomes. OBJECTIVES: To examine the relationship between nurse staffing and patient mortality in Korean intensive care units (ICUs). METHODS: Using survey and administrative databases, this study included 27,372 ICU patients discharged from 42 tertiary and 194 secondary hospitals. Ownership (public vs. private), location (metropolitan city vs. province), size, specialization of ICUs (specialized vs. mixed), physician staffing and nurse staffing, and years of nurse experience were included as hospital and ICU characteristics. Nurse staffing was measured as the ratio of average daily census to the total number of full-time equivalent registered nurses in ICUs. Age, gender, 26 categories of primary diagnoses, 16 groups of comorbid conditions, and source of payment were used for risk adjustment. Mortality was defined as deaths that occurred in the hospital or on the date of hospital discharge. Using SAS GLIMMIX procedures, multivariate logistic regression analysis was employed separately for tertiary and secondary hospitals. RESULTS: In tertiary hospitals, a greater likelihood of dying was found among patients who were admitted to a mixed ICU (odds ratio [OR] = 1.61, 95% confidence interval [CI] = 1.14-2.26) and where there was no board-certified physician present for 4 or more hours per day (OR = 1.56, 95% CI = 1.20-2.01). In secondary hospitals, every additional patient per RN was associated with a 9% increase in the odds of dying (OR = 1.09, 95% CI = 1.04-1.14). Nurse experience had no significant relationship with mortality. DISCUSSION: Nurse and physician staffing and specialization of ICUs impacted patient mortality. Associations differed in tertiary and secondary hospitals. Further investigation of the impact of organizational environments in ICUs is needed.

Introduction Advances in the diagnosis and management of acute ischemic stroke (AIS) and the increased use of mechanical thrombectomy (MT) have improved the quality of care and prognosis of patients with AIS since 2015. We investigated the changing trends in mortality of patients with AIS in Korea before and after 2015. Materials and methods A retrospective cohort study was conducted using combined anonymized data from the Acute Stroke Assessment Registry of Korea and the Health Insurance Review &amp;amp; Assessment Service database. Patients with ischemic stroke with precise onset time and initial National Institute of Health Stroke Scale records were included. Results Patients receiving MT treatment increased from 256 (2.7%) pre-2015 to 1,037 (3.9%) post-2015 ( p &amp;lt; 0.001). Overall mortality significantly decreased from pre-2015 to post-2015. In pre-2015, intravenous thrombolysis (IVT) administered within 2 h significantly reduced 3-month mortality when compared with non-IVT. While, in post-2015, IVT administered within 2 h significantly reduced the 3-month, 1-year, 2-year, and 4-year mortality ( p &amp;lt; 0.05). MT only reduced 1-year mortality pre-2015; however, MT significantly reduced the 3-month, 1-year, and 2-year mortality post-2015 ( p &amp;lt; 0.05). Post-stroke antiplatelet and anticoagulant drugs significantly reduced the 3-month, 1-year, 2-year, and 4-year mortality post-2015. Discussion Since 2015, faster IVT has significantly reduced the short- and long-term mortality in patients with AIS; MT reduced the 3-month, 1-year, and 2-year mortality. Post-stroke antithrombotic medication has significantly lowered the 2- and 4-year mortality since 2015. Conclusions Changing trends in AIS management since 2015 have improved the prognosis of patients with AIS.

medical records Patients are also able to reduce their direct and indirect healthcare costs through reduction of the need for healthcare clinic visits or hospitalization

Comorbid conditions appear to be common among individuals hospitalised with coronavirus disease 2019 (COVID-19) but estimates of prevalence vary and little is known about the prior medication use of patients. Here, we describe the characteristics of adults hospitalised with COVID-19 and compare them with influenza patients. We include 34,128 (US: 8362, South Korea: 7341, Spain: 18,425) COVID-19 patients, summarising between 4811 and 11,643 unique aggregate characteristics. COVID-19 patients have been majority male in the US and Spain, but predominantly female in South Korea. Age profiles vary across data sources. Compared to 84,585 individuals hospitalised with influenza in 2014-19, COVID-19 patients have more typically been male, younger, and with fewer comorbidities and lower medication use. While protecting groups vulnerable to influenza is likely a useful starting point in the response to COVID-19, strategies will likely need to be broadened to reflect the particular characteristics of individuals being hospitalised with COVID-19.

BACKGROUND: During the COVID-19 pandemic, South Korea has achieved perfect universal health coverage (UHC)-all Koreans receive UHC regardless of their socioeconomic status. The current study investigated whether socioeconomic disparities remained in COVID-19 health outcomes under UHC. METHODS: This retrospective, observational study included all 7,590 confirmed COVID-19 patients in South Korea up to 15 May 2020. We used the official medical claim database, and socioeconomic status was estimated by insurance type (National Health Insurance Service [NHIS] beneficiaries and Medical Aid [MA] recipients). Type of insurance is a well-known indicator of socioeconomic status. Prevalence (per one million), mortality rate (per one million), and case fatality rate were calculated. To determine the factors associated with case fatality rate, multivariable logistic regressions were performed. FINDINGS: The nationwide prevalence, mortality rate, and case fatality rate of COVID-19 was 144•4, 4•3 and 3•0%, respectively. MA recipients had higher prevalence (424•3 vs 136.3), mortality rate (28•3 vs 3•6), and case fatality rate (6•7 vs 2•7) than NHIS beneficiaries. However, the adjusted analysis showed that the type of insurance was not associated with higher odds of case fatality. INTERPRETATION: We found socioeconomic disparities in COVID-19 prevalence and fatality despite UHC. However, disparities in fatality were not due to socioeconomic status, but due to the poor underlying health conditions of the people. This result can be explained by a combination of UHC, rapid early testing and treatment, transmission-reducing behaviours, and regional preparedness. FUNDING: This research did not receive any funding.

OBJECTIVE: To develop a tool for assessing the performance of primary care services in South Korea from the patient's perspective and to test the validity of the tool under the conceptual framework of the recently developed definition of primary care in Korea. DESIGN: Item development for questionnaire and a cross-sectional survey for tool validation at 16 primary care clinics. All family physicians included in this study were required to have practiced at their current clinic for at least 2 years. PARTICIPANTS: A nine expert panel was assembled for tool development and patients (or guardians) who had visited their primary care clinic on six or more occasions over a period of more than 6 months participated in the survey. MAIN OUTCOME MEASURES: Scores corresponding to each domain of primary care. RESULTS: A total of 722 effective data sets were used for the analysis. Five items were eliminated from the preliminary 30-item tool after expert discussions at two seminars. Another four items were eliminated by principle component analysis. For each of the four domains (comprehensiveness, coordination function, personalized care, and family/community orientation), tests of scaling assumptions were well satisfied by all Likert-scaled measures. On the other hand, 'first contact' turned out to be a composite domain with five independent single-item scales. CONCLUSION: The Korean primary care assessment tool (version 1) consists of four multi-item scales and one composite scale. Widespread application of this tool will provide an empirical basis for the measurement, monitoring and continuous improvement of primary care in South Korea.

BACKGROUND: Medication adherence has been identified as a major factor influencing health outcomes in patients with type 2 diabetes. OBJECTIVES: We assessed the relationship between initial adherence to oral antihyperglycemic medications and subsequent health outcomes. RESEARCH DESIGN AND SUBJECTS: This was a retrospective cohort study of 40,082 patients enrolled in Korea's national health insurance program, who were 20 years of age or older and first diagnosed with type 2 diabetes in 2004. The patients were followed up for 3 years using claims data to measure adherence to oral antihyperglycemic medications for the initial 2 years after diagnosis and to investigate hospitalization, mortality, and healthcare costs in the third year of follow-up. RESULTS: After adjusting for confounders, nonadherence in the first 2 years after prescription increased the risk for hospitalization in the third year compared with adherence over the same period [odds ratio (OR)=1.26, 95% confidence interval=1.08-1.47]. Nonadherence in at least one of the 2 years increased the risk for hospitalization compared with adherence in both years. In addition, nonadherence during both the first and second years was associated with statistically significantly greater risks for mortality during this period than was adherence (odds ratio=1.40, 95% confidence interval=1.01-1.95). Medication adherence decreased healthcare costs compared with nonadherence (β=-0.127; P<0.001). CONCLUSIONS: Because improved medication adherence can produce better health outcomes for diabetes patients and also save national healthcare resources, government-sponsored healthcare policies to improve medication adherence among newly diagnosed diabetes patients are urgently required.

OBJECTIVES: To compare the performance of three International Statistical Classification of Diseases, 10th Revision translations of the Charlson comorbidities when predicting in-hospital among patients with myocardial infarction (MI). METHODS: MI patients > or =20 years of age with the first admission during 2006 were identified(n=20,280). Charlson comorbidities were drawn from Heath Insurance Claims Data managed by Health Insurance Review and Assessment Service in Korea. Comparisions for various conditions included (a) three algorithms (Halfon, Sundararajan, and Quan algorithms), (b) lookback periods (1-, 3- and 5-years), (c) data range (admission data, admission and ambulatory data), and (d) diagnosis range (primary diagnosis and first secondary diagnoses, all diagnoses). The performance of each procedure was measured with the c-statistic derived from multiple logistic regression adjusted for age, sex, admission type and Charlson comorbidity index. A bootstrapping procedure was done to determine the approximate 95% confidence interval. RESULTS: Among the 20,280 patients, the mean age was 63.3 years, 67.8% were men and 7.1% died while hospitalized. The Quan and Sundararajan algorithms produced higher prevalences than the Halfon algorithm. The c-statistic of the Quan algorithm was slightly higher, but not significantly different, than that of other two algorithms under all conditions. There was no evidence that on longer lookback periods, additional data, and diagnoses improved the predictive ability. CONCLUSIONS: In health services study of MI patients using Health Insurance Claims Data, the present results suggest that the Quan Algorithm using a 1-year lookback involving primary diagnosis and the first secondary diagnosis is adequate in predicting in-hospital mortality.

Background: Untreated or insufficiently treated Graves disease in pregnancy may pose risks to both mother and fetus. Antithyroid drugs (ATDs) are the treatment mainstay, but the potential teratogenic effect of these drugs has prompted clinicians to question the safe management of this vulnerable population. Objective: To examine the association between maternal prescriptions for ATDs and congenital malformations in live births. Design: Nationwide cohort study. Setting: Korean National Health Insurance database. Participants: A cohort of 2 886 970 completed pregnancies linked to live-born infants in 2 210 253 women between 2008 and 2014. Intervention: Maternal prescriptions for ATDs in the first trimester. Measurements: The risk for overall and organ-specific congenital malformations in offspring, with logistic regression models used to control for potential confounders. Results: 12 891 pregnancies (0.45%) were exposed to ATDs during the first trimester. The prevalence of malformations in exposed offspring was 7.27%, compared with 5.94% in offspring of women who were not prescribed ATDs during pregnancy (P < 0.001) (adjusted odds ratio, 1.19 [95% CI, 1.12 to 1.28]). Absolute increases in the prevalence of congenital malformations per 1000 live births were 8.81 cases (CI, 3.92 to 13.70 cases) for propylthiouracil alone, 17.05 cases (CI, 1.94 to 32.15 cases) for methimazole (MMI) alone, and 16.53 cases (CI, 4.73 to 28.32 cases) for propylthiouracil and MMI, compared with pregnancies without ATD prescriptions. In the MMI group, a high cumulative dose (>495 mg) during the first trimester was associated with an increased risk for malformations compared with a low dose (1 to 126 mg) (adjusted odds ratio, 1.87 [CI, 1.06 to 3.30]). Limitation: The study used a prescription claims database to assess ATD exposure. Conclusion: Exposure to ATDs during the first trimester was associated with increased risk for congenital malformations, particularly for pregnancies in which women received prescriptions for MMI or both ATDs. Primary Funding Source: None.

We sought to assess continuity of care for elderly patients in Korea and to examine any association between continuity of care and health outcomes (hospitalization, emergency department visits, health care costs). This was a retrospective cohort study using the Korea National Health Insurance Claims Database. Elderly people, 65-84 yr of age, who were first diagnosed with diabetes mellitus (n=268,220), hypertension (n=858,927), asthma (n=129,550), or chronic obstructive pulmonary disease (COPD, n=131,512) in 2002 were followed up for four years, until 2006. The mean of the Continuity of Care Index was 0.735 for hypertension, 0.709 for diabetes mellitus, 0.700 for COPD, and 0.663 for asthma. As continuity of care increased, in all four diseases, the risks of hospitalization and emergency department visits decreased, as did health care costs. In the Korean health care system, elderly patients with greater continuity of care with health care providers had lower risks of hospital and emergency department use and lower health care costs. In conclusion, policy makers need to develop and try actively the program to improve the continuity of care in elderly patients with chronic diseases.

Exposure to radon gas is the second most common cause of lung cancer after smoking. A large number of studies have reported that exposure to indoor radon, even at low concentrations, is associated with lung cancer in the general population. This paper reviewed studies from several countries to assess the attributable risk (AR) of lung cancer death due to indoor radon exposure and the effect of radon mitigation thereon. Worldwide, 3-20 % of all lung cancer deaths are likely caused by indoor radon exposure. These values tend to be higher in countries reporting high radon concentrations, which can depend on the estimation method. The estimated number of lung cancer deaths due to radon exposure in several countries varied from 150 to 40,477 annually. In general, the percent ARs were higher among never-smokers than among ever-smokers, whereas much more lung cancer deaths attributable to radon occurred among ever-smokers because of the higher rate of lung cancers among smokers. Regardless of smoking status, the proportion of lung cancer deaths induced by radon was slightly higher among females than males. However, after stratifying populations according to smoking status, the percent ARs were similar between genders. If all homes with radon above 100 Bq/m(3) were effectively remediated, studies in Germany and Canada found that 302 and 1704 lung cancer deaths could be prevented each year, respectively. These estimates, however, are subject to varying degrees of uncertainty related to the weakness of the models used and a number of factors influencing indoor radon concentrations.