• Journal of the Korean Data and Information Science Society
• /
• v.21 no.2
• /
• pp.279-289
• /
• 2010

In this study, we propose a simple statistical model for determining the admission or discharge of 668 patients with a chief complaint of dyspnea. For this, we use 11 explanatory variables which are chosen to be important by clinical experts among 55 variables. As a modification process, we determine the discharge interval of each variable by the kernel density functions of the admitted and discharged patients. We then choose the optimal model for determining the discharge of patients based on the number of explanatory variables belonging to the corresponding discharge intervals. Since the numbers of the admitted and discharged patients are not balanced, we use, as the criteria for selecting the optimal model, the arithmetic mean of sensitivity and specificity and the harmonic mean of sensitivity and precision. The selected optimal model predicts the discharge if 7 or more explanatory variables belong to the corresponding discharge intervals.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.1
• /
• pp.125-133
• /
• 2013

In this paper, penalized binary logistic regression models are employed as statistical models for determining the discharge of 668 patients with a chief complaint of dyspnea based on 11 blood tests results. Specifically, the ridge model based on $L^2$ penalty and the Lasso model based on $L^1$ penalty are considered in this paper. In the comparison of prediction accuracy, our models are compared with the logistic regression models with all 11 explanatory variables and the selected variables by variable selection method. The results show that the prediction accuracy of the ridge logistic regression model is the best among 4 models based on 10-fold cross-validation.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.477-485
• /
• 2011

Park et. al (2010) proposed a statistical model for determining the admission or discharge of 668 patients with a chief complaint of dyspnea by the number of 11 blood tests belonging to the corresponding discharge intervals. Since this method does not take into consideration the importance of each blood test result, its performance might not be optimally good. In this study, we employ a quantification method to evaluate the importance of those blood test results, and then provide a new statistical mode that takes the importance into consideration. The results show that the performance of this new model is a little better than that of the model by Park et. al (2010).

• Journal of Digital Convergence
• /
• v.11 no.4
• /
• pp.351-359
• /
• 2013

This study was conducted to investigate the factors that affect the length of stay in hospital for schizophrenic patients. Of the data from the in-depth injury patient surveillance system, the final subject included 2,239 patients with schizophrenia in their final diagnosis. Using SPSS 18.0, a hierarchical regression analysis was performed by sequentially entering the explanatory variables by setting sociodemographic characteristics, discharge characteristics and hospital characteristics as explanatory variables and the length of stay in hospital as a dependent variable. The findings showed that the sociodemographic characteristics had the highest explanatory power and the explanatory power changed when the explanatory variable of the hospital characteristics was added, as opposed to the discharge characteristics. Male, type-1 medicaid, Chungcheong-do and the number of beds were found to be the factors that mostly affect the length of stay. Since this study used the secondary data, it has a limitation in terms of additional variables that could better explain the length of stay for schizophrenic patients. Nevertheless, it has an implication in that it investigated a large scale of data on a national level. For the effort of reducing the length of stay, it is suggested that an effort should be made at the national level, by focusing more on the hospital characteristics as well as the individual characteristics of patients.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.393-400
• /
• 2011

In a large scale survey, cluster sampling design in which a set of observation units called clusters are selected is often used to satisfy practical restrictions on time and cost. Especially, a two stage cluster sampling design is preferred when a strong intra-class correlation exists among observation units. The sample Primary Sampling Unit(PSU) and Secondary Sampling Unit(SSU) size for a two stage cluster sample is determined by the survey cost and precision of the estimator calculated. For this study, we derive the optimal sample PSU and SSU size when the population SSU size across the PSU are di erent by extending the result obtained under the assumption that all PSU have the same number of SSU. The results on the sample size are then applied to the $4^{th}$ Korea Hospital Discharge results and is compared to the conventional method. We also propose the optimal sample SSU (discharged patients) size for the $7^{th}$ Korea Hospital Discharge Survey.

• Proceedings of the KAIS Fall Conference
• /
• 2010.05b
• /
• pp.1061-1063
• /
• 2010

본 연구는 2008년 1월 1일부터 12월 31일 까지 1년 동안 대전에 위치한 일개 대학병원 퇴원요약 정보를 활용하여 환자 개인요인 측면과 의료적 측면에서 전원 환자의 특성과 요인을 파악하여 전원환자 관리의 자료로 이용하고자 한다. 연구결과 퇴원환자 30,793명중 타 병원으로 전원 된 990명 전체가 대상이었으며, 성별로 보면 남자가 여자보다 높은 분포를 보였다. CART기법에 의한 의사결정나무는 치료결과, 입원경로 등이 타 의료 기관으로 전원되는 특성을 보여주고 있음을 알 수 있다. 결과적으로 의료적인 측면보다는 개인적인 측면으로 환자들이 전원을 하는 비율이 높았다. 이와 같이 전원환자 특성 및 요인 분석이 병원의 지속적인 환자 관리와 환자와의 신뢰를 형성하고 꾸준한 병원방문을 유도할 수 있을 것으로 생각된다.

• Clinical and Experimental Pediatrics
• /
• v.49 no.7
• /
• pp.757-762
• /
• 2006

Purpose : Pneumonia is one of the most common infections in children who visit emergency departments(ED), but standard clinical guidelines for children with pneumonia in Korea have not been studied. This study was performed to collect and evaluate a data-base of children with pneumonia for establishing clinical guidelines in ED. Methods : This study reviewed 304 children who were diagnosed and treated for pneumonia in the ED at one tertiary hospital between January 2003 and December 2003 retrospectively by reviewing the charts and analyzing the clinical characteristics, laboratory findings, and radiologic findings between an admission group and a discharge group. Results : The 2 year-5 year age group was the top of age distribution and the peak incidence of monthly distribution was December. Two hundred forty seven(81.3 percent) children were hospitalized(admission group), and the mean length of hospitalization was $7.24{\pm}3.24$ days. The most common indications of admission were fever, tachypnea and an age of less than three months. There was statistical differences in the outpatient department follow-up between the two groups(85.8 percent in admission group vs 35.1 percent in discharge group). Conclusion : More prospective studies are needed to establish clinical standard guidelines for children with pneumonia. This will be helpful in ED management and will aid the prevention of pneumonia.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.389-400
• /
• 2011

This study was carried out to analysis factors related to in-hospital mortality of community-acquired peumonia using administrative database. The subjects were 5,353 community-acquired pneumonia inpatients of the Korean National Hospital Discharge Injury Survey 2004-2006 data. The data were analyzed using chi-squared test and decision tree model in the data mining technique. Among the decision tree model, C4.5 had the best performance. The critical factors on in-hospital mortality of communityacquired pneumonia are admission route, respiratory failure, congenital heart failure including age, comorbidity, and bed size. This study was carried out using the administrative database including patients' characteristics and comorbidity. However further study should be extensively including hospital characteristics, regional medical resources, and patient management practice behavior.

• Journal of Digital Convergence
• /
• v.11 no.6
• /
• pp.221-233
• /
• 2013

This study aims to develop the severity-adjusted length of stay(LOS) model for acute stroke patients using data from the hospital discharge survey and propose management of length of stay(LOS) for acute stroke patients and using for Hospital management. The dataset was taken from 23,134 database of the hospital discharge survey from 2004 to 2009. The severity-adjusted LOS model for the acute stroke patients was developed by data mining analysis. From decision making tree model, the main reasons for LOS of acute stroke patients were acute stroke type. The difference between severity-adjusted LOS from the decision making tree model and real LOS was compared and it was confirmed that insurance type and bed number of hospital, location of hospital were statistically associated with LOS. And to conclude, hospitals should manage the LOS of acute stroke patients applying it into the medical information system.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.10
• /
• pp.4910-4918
• /
• 2013

This study aims to design a Severity-Adjusted LOS(Length of Stay) Model in order to efficiently manage LOS of AMI(Acute Myocardial Infarction) patients. We designed a Severity-Adjusted LOS Model with using data-mining methods(multiple regression analysis, decision trees, and neural network) which covered 6,074 AMI patients who showed the diagnosis of I21 from 2004-2009 Korean National Hospital Discharge in-depth Injury Survey. A decision tree model was chosen for the final model that produced superior results. This study discovered that the execution of CABG, status at discharge(alive or dead), comorbidity index, etc. were major factors affecting a Sevirity-Adjustment of LOS of AMI patients. The difference between real LOS and adjusted LOS resulted from hospital location and bed size. The efficient management of LOS of AMI patients requires that we need to perform various activities after identifying differentiating factors. These factors can be specified by applying each hospital's data into this newly designed Severity-Adjusted LOS Model.