• Health Policy and Management
• /
• v.26 no.4
• /
• pp.359-372
• /
• 2016

Background: The purpose of this study was to develop risk-adjustment models for acute stroke mortality that were based on data from Health Insurance Review and Assessment Service (HIRA) dataset and to evaluate the validity of these models for comparing hospital performance. Methods: We identified prognostic factors of acute stroke mortality through literature review. On the basis of the avaliable data, the following factors was included in risk adjustment models: age, sex, stroke subtype, stroke severity, and comorbid conditions. Survey data in 2014 was used for development and 2012 dataset was analysed for validation. Prediction models of acute stroke mortality by stroke type were developed using logistic regression. Model performance was evaluated using C-statistics, $R^2$ values, and Hosmer-Lemeshow goodness-of-fit statistics. Results: We excluded some of the clinical factors such as mental status, vital sign, and lab finding from risk adjustment model because there is no avaliable data. The ischemic stroke model with age, sex, and stroke severity (categorical) showed good performance (C-statistic=0.881, Hosmer-Lemeshow test p=0.371). The hemorrhagic stroke model with age, sex, stroke subtype, and stroke severity (categorical) also showed good performance (C-statistic=0.867, Hosmer-Lemeshow test p=0.850). Conclusion: Among risk adjustment models we recommend the model including age, sex, stroke severity, and stroke subtype for HIRA assessment. However, this model may be inappropriate for comparing hospital performance due to several methodological weaknesses such as lack of clinical information, variations across hospitals in the coding of comorbidities, inability to discriminate between comorbidity and complication, missing of stroke severity, and small case number of hospitals. Therefore, further studies are needed to enhance the validity of the risk adjustment model of acute stroke mortality.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.48-54
• /
• 2021

Sepsis is one of the leading causes of mortality globally, and it costs billions of dollars annually. However, treating septic patients is currently highly challenging, and more research is needed into a general treatment method for sepsis. Therefore, in this work, we propose a reinforcement learning method for learning the optimal treatment strategies for septic patients. We model the patient physiological time series data as the input for a deep recurrent Q-network that learns reliable treatment policies. We evaluate our model using an off-policy evaluation method, and the experimental results indicate that it outperforms the physicians' policy, reducing patient mortality up to 3.04%. Thus, our model can be used as a tool to reduce patient mortality by supporting clinicians in making dynamic decisions.

• Journal of Preventive Medicine and Public Health
• /
• v.37 no.2
• /
• pp.120-126
• /
• 2004

Objectives : This study sought to examine relationships between alcohol drinking and cardiovascular disease mortality and all-cause mortality. Methods : From March 1985 through December 1999, 2,696 males and 3,595 females aged 55 or over as of 1985 were followed up for their mortality until 31 December 1999. We calculated the mortality risk ratios by level of alcohol consumption. Among the drinker, the level of alcohol consumption was calculated by the frequency of alcohol comsumption and the type of alcohol. Cox proportional hazard model was used to adjust for confounding factors. Results : Among males, compared to abstainer, heavy drinker had significantly higher mortality in all cause(Risk ratio=1.35), cardiovascular disease(Risk ratio=1.52) and cerebrovascular disease(Risk ratio =1.66). Although not significant, moderate drinker had lower ischemic heart disease mortality(Risk ratio =0.38). Among females, there was no statistically significant association between alcohol comsumption and mortality. Conclusion : The results of this study suggest that alcohol drinking has harmful effect on all-cause mortality, cardiovascular disease mortality and cerebrovascular disease mortality among males, especially in heavy drinker among males. Minimal evidence on protective effect for cardiovascular disease mortality in low or moderate drinker is observed.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1169-1182
• /
• 2016

Cancer is a typical cause of death in Korea that becomes a major issue in health care. According to Cause of Death Statistics (2014) by National Statistical Office, SMRs (standardized mortality rates) in Busan were counted as the highest among all cities. In this paper, we used data of Busan Regional Cancer Center to estimate the extent of the cancer incidence rate and cancer mortality rate. The data are considered in small areas of administrative units such as Gu/Dong from years 2003 to 2009. All cancer including four major cancers (stomach cancer, colorectal cancer, lung cancer, liver cancer) have been analyzed. We carried out model selection and parameter estimation using spatial multi-level model incorporating a spatial correlation. For the spatial effects, CAR (conditional autoregressive model) has been assumed.

• Journal of Environmental Health Sciences
• /
• v.36 no.1
• /
• pp.20-26
• /
• 2010

Temperature change has been shown to affect daily mortality even though different analytical methods produce different results. The effect of air pollution on the relationship between the temperature and the mortality is not large, although differences exist between temperature models. The aim of this study was to examine how the temperature change affected the daily mortality in Seoul by comparing the results from the temperature model using two study periods: one from 1994 to 2007 and the other from 1997 to 2007. Generally mean temperature, minimum temperature and Q10 temperature was derived as an optimal model, even though there are differences between age and cause of death. The analysis of threshold using total mortalities in all ages from 1994 to 2007 and from 1997 to 2007 showed that the number of the deaths increased 7.02% (95% CI: 6.06~7.98) and 2.51% (95% CI: 1.83~3.19), respectively as the mean temperature increased $1^{\circ}C$ from a threshold temperature of $27.5^{\circ}C$ and $25.7^{\circ}C$ respectively. These results indicated that the temperature has less effect on the number of death than does an extreme heat wave period.

• Health Policy and Management
• /
• v.10 no.2
• /
• pp.120-130
• /
• 2000

Studies on the relation between socio-economic factors and metermal and child health have found that poverty, lack of edcation, inappropriate health serives are affecting to maternal and child health. The Gender Related Development Index (GDI) focuses on equality between men and women as well as on the average achiement of all people taken together, using same cariables as the Human Development Index (HDI) which are life expectancy, literacy rate, and per capita GDP. This research is to inverstigate whether HDI and GDI are useful determinants for maternal infant mortality. Using 146 UN member countries date, we condented multiple regression analysis for maternal and infant mortality with three models which are Model(individual variables-literacy rate, per capita GDP), Model(HDI) and Model(GDI). The results showed that HDI and GDI are powerful determinants of both maternal and infant mortality, respectively HDI($\beta$=-1.18, t=3.3; $\beta$=1.04, t=5.1) GDI($\beta$=-1.44, t=3.9; $\beta$=1.28, t=6.5) The higher power in model with GDI for both maternal and infant mortalities represented that GDI was more powerful determinant of maternal and infant mortality, than HDI respectively HDI($R^2$=0.824, $R^2$=0.842), GDI($R^2$=0.834, $R^2$=0.865). In conclusion, the maternal and infant mortalities are explained by GDI than HDI and may be lower in the societies where there are less discimination between men and women.

• Korea journal of population studies
• /
• v.9 no.1
• /
• pp.14-31
• /
• 1986

This study shows trends and age-sex patterns of mortality in Korea, based on adjusted death registration data during 1956∼1980. Description on the data-the types of errors and their methods of adjustment-are presented elsewhere (Kim, 1986). Crude death rates have declined in 20 years to 7.0 in 1976∼1980, almost half the level of 13.0 in 1956∼1960. Mortality of females declined faster than males, especially at older ages. Substantial differences in provincial mortality are observed based on the expectation of life at birth. The difference between the highest and the lowest provinces is 9.2 years for males and 8.3 years for females during 1976∼1980. This study presents two interesting features of age-sex patterns of mortality in Korea. One feature is higher female mortality than males during childhood periods. However, infant mortality is higher for males than for females throughout the period, though the difference is much smaller than is expected from various model life tables. Another feature is the rapid rise in mortality for males after age 40. This pattern is found to be existed throughout the period 1956∼1980. Further studies are recommended to establish its causal linkage to traditional sex role (as main bread winner) and health behavior (smoking and drinking) of males and recent rapid industrialization in Korea.

• Journal of Preventive Medicine and Public Health
• /
• v.34 no.3
• /
• pp.244-252
• /
• 2001

Objective : Health insurers and policy makers are increasingly examining the hospital mortality rate as an indicator of hospital quality and performance. To be meaningful, a risk-adjustment of the death rates must be implemented. This study reviewed 5 severity measurement methods and applied them to the same data set to determine whether judgments regarding the severity-adjusted hospital mortality rates were sensitive to the specific severity measure. Methods : The medical records of 584 patients who underwent coronary artery bypass graft surgery in 6 general hospitals during 1996 and 1997 were reviewed by trained nurses. The MedisGroups, Disease Staging, Computerized Severity Index, APACHE III and KDRG were used to quantify severity of the patients. The predictive probability of death was calculated for each patient in the sample from a multivariate logistic regression model including the severity score, age and sex to evaluate the hospitals' performance, the ratio of the observed number of deaths to the expected number for each hospital was calculated. Results : The overall in-hospital mortality rate was 7.0%, ranging from 2.7% to 15.7% depending on the particular hospital. After the severity adjustment, the mortality rates for each hospital showed little difference according to the severity measure. The 5 severity measurement methods varied in their statistical performance. All had a higher c statistic and $R^2$ than the model containing only age and sex. There was a little difference in the relative hospital performance evaluation by the severity measure. Conclusion : These results suggest that judgments regarding a hospital's performance based on severity adjusted mortality can be sensitive to the severity measurement method. Although the 5 severity measures regarding hospital performance concurred, more often than would be expected by chance, the assessment of an individual hospital mortality rates varied by the different severity measurement method used.

• Journal of Preventive Medicine and Public Health
• /
• v.44 no.6
• /
• pp.242-248
• /
• 2011

Objectives: Busan is reported to have the highest mortality rate among 16 provinces in Korea, as well as considerable health inequality across its districts. This study sought to examine overall and cause-specific mortality and deprivation at the town level in Busan, thereby identifying towns and causes of deaths to be targeted for improving overall health and alleviating health inequality. Methods: Standardized mortality ratios (SMRs) for all-cause and four specific leading causes of death were calculated at the town level in Busan for the years 2005 through 2008. To construct a deprivation index, principal components and factor analysis were adopted, using 10% sample data from the 2005 census. Geographic information system (GIS) mapping techniques were applied to compare spatial distributions between the deprivation index and SMRs. We fitted the Gaussian conditional autoregressive model (CAR) to estimate the relative risks of mortality by deprivation level, controlling for both the heterogeneity effect and spatial autocorrelation. Results: The SMRs of towns in Busan averaged 100.3, ranging from 70.7 to 139.8. In old inner cities and towns reclaimed for replaced households, the deprivation index and SMRs were relatively high. CAR modeling showed that gaps in SMRs for heart disease, cerebrovascular disease, and physical injury were particularly high. Conclusions: Our findings indicate that more deprived towns are likely to have higher mortality, in particular from cardiovascular disease and physical injury. To improve overall health status and address health inequality, such deprived towns should be targeted.

• Journal of Preventive Medicine and Public Health
• /
• v.45 no.1
• /
• pp.14-20
• /
• 2012

Objectives: The aim of this study was to evaluate and quantify the risk of both individual and combined health behaviors on premature mortality in middle aged men in Korea. Methods: In total, 14 533 male subjects 40 to 59 years of age were recruited. At enrollment, subjects completed a baseline questionnaire, which included information about socio-demographic factors, past medical history, and life style. During the follow-up period from 1993 to 2008, we identified 990 all-cause premature deaths using national death certificates. A Cox proportional hazard regression model was used to estimate the hazard ratio (HR) of each health risk behavior, which included smoking, drinking, physical inactivity, and lack of sleep hours. Using the Cox model, each health behavior was assigned a risk score proportional to its regression coefficient value. Health risk scores were calculated for each patient and the HR of all-cause premature mortality was calculated according to risk score. Results: Current smoking and drinking, high body mass index, less sleep hours, and less education were significantly associated with all-cause premature mortality, while regular exercise was associated with a reduced risk. When combined by health risk score, there was a strong trend for increased mortality risk with increased score (p-trend < 0.01). When compared with the 1-9 score group, HRs of the 10-19 and 20-28 score groups were 2.58 (95% confidence intervals [CIs], 2.19 to 3.03) and 7.09 (95% CIs, 5.21 to 9.66), respectively. Conclusions: Modifiable risk factors, such as smoking, drinking, and regular exercise, have considerable impact on premature mortality and should be assessed in combination.