• Journal of Information Technology Applications and Management
• /
• v.13 no.4
• /
• pp.273-290
• /
• 2006

This study suggests a frequency matrix technique to predict personal credit rate more efficiently using incomplete data sets. At first this study test on multiple discriminant analysis and logistic regression analysis for predicting personal credit rate with incomplete data sets. Missing values are predicted with mean imputation method and regression imputation method here. An artificial neural network and frequency matrix technique are also tested on their performance in predicting personal credit rating. A data set of 8,234 customers in 2004 on personal credit information of Bank A are collected for the test. The performance of frequency matrix technique is compared with that of other methods. The results from the experiments show that the performance of frequency matrix technique is superior to that of all other models such as MDA-mean, Logit-mean, MDA-regression, Logit-regression, and artificial neural networks.

• The Journal of the Korean dental association
• /
• v.51 no.9
• /
• pp.511-517
• /
• 2013

To assess the association between oral health and general health, this study examined the relationship between chewing difficulty and twelve chronic health conditions such as hypertension, hyperlipidemia, diabetes, cerebro- and cardiovascular disease, musculoskeletal disease, respiratory disease, eye/nose/throat disease, stomach/intestinal ulcer, renal dysfunction, thyroid disease, depression, and cancer in Korea. The study population was 3,066 elders aged 65 years old and more from the fourth Korean National Health and Nutrition Examination Survey. Chewing difficulty was measured on a 5-point Likert scale. Chronic conditions were assessed by self-reported questionnaire. Confounders were age, gender, education, income, smoking, drinking, and obesity. Chi-square test, general linear model, and multiple logistic regression model were done with complex sampling design. Musculoskeletal disease (adjusted odds ratio=1.33), respiratory disease (adjusted odds ratio=1.52), and cancer (adjusted odds ratio=1.58) were independently associated with chewing difficulty. Multiple chronic conditions with more than 4 chronic disease showed significant association with chewing difficulty (adjusted odds ratio=1.37).

• Journal of Korean Public Health Nursing
• /
• v.23 no.2
• /
• pp.219-231
• /
• 2009

Purpose: This study compared health behaviors between single-member and multiple-member households by gender in adults. Method: Study subjects were 21,757 adults 20-65-years-of-age from the Third Korea National Health and Nutrition Examination Survey in 2005. The relationship between health behaviors and residence characteristics was assessed using multiple logistic regression analysis while controlling for sociodemographic characteristics. Results: Younger women living alone were more likely to drink than older women living with others. Women who were in multiple-member households and had a low income were less likely to exercise regularly, to sleep well, and to have regular health examinations. Age, educational level, and income were significantly associated with smoking, exercise, and health examination rather than residence characteristics in men. Conclusion: Residence characteristics should be considered in developing a public health program to reduce or increase modifiable health behaviors as well as age, gender, education, and income.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.148-156
• /
• 2019

This study identified the effects of Depression, Anxiety, and Quality of Sleep on Excessive Daytime Sleepiness (EDS) in nursing students. Totally, 213 students of a nursing college located in C do, Korea, were enrolled for the study, which was accomplished through a questionnaire. Data were collected from May 20 to 30, 2019. The collected data were compiled using the SPSS/WIN 25.0 statistic program by applying Chi-square test, Paired t-test, Pearson's Coefficient, and Multiple logistic regression. The following results were obtained. The mean score for EDS was 8.95(±4.56), and Prevalence of EDS was determined to be 40.8%. The mean scores obtained for the affecting factors were Depression 10.05(±7.85), STAI-S 46.09(±9.50), STAI-T 46.47(±9.93), and Quality of Sleep 6.51(±2.95). Depression, STAI-S, STAI-T, and Quality of Sleep were significantly different for EDS and Non-EDS students (t=1.955, p=0.024; t=5.446, p<0.001; t=1.716, p=0.007; t=12.168, p<0.001; respectively). Multiple logistic regression revealed that factors associated with EDS were STAI-S and Quality of Sleep (adjusted odds ratio=1.04, 95% Confidence Interval=1.01-1.08; adjusted odds ratio=1.16, 95% Confidence Interval=1.04-1.29; respectively). These findings indicate the necessity to improve the quality of sleep and manage anxiety alleviation of nursing students, to reduce excessive daytime sleepiness.

• Health Policy and Management
• /
• v.9 no.4
• /
• pp.41-64
• /
• 1999

The major concern for this research is to discuss and to offer some solutions to bring the effectiveness of existing notifiable diseases reporting system over the physicians' attitudes of reporting, the actual condition of performance and the reasons of inertia in notifiable diseases reporting through examining the physicians of medical institutions in nationwide such as pediatrics, internal medicine and family medicine. The actual conditions of notifiable communicable diseases(NCD) reporting was surveyed by mail objectifying an internal medicine, pediatrics and family medicine in nationwide on the basis of stratified random sampling method divided into the classification of medical institutions and areas. As a result of survey. the rate of respondents showed 145 persons from physicians, 105 persons from hospitals. 120 persons from general hospitals, and 51 persons from tertiary hospitals. The total number of respondents were 421 and was rated 59.0 %. The analysis of collected survey went through a descriptive analysis primarily to grasp physicians' attitudes on the notifiable communicable diseases reporting, and then upon the dependent variables. Following are major findings obtained form the data analysis. 1. The results of a descriptive analysis on physicians' attitudes towards reporting NCD were as follows: First, the respondents who didn't know that yellow fever is reporting NCD were 11.0% of clinic, 10.5% of hospital. 5.0% of general hospital. 11.8% of tertiary hospital. and in case of hepatitis B, were 26.9% of clinic, 35.2% of hospital. 35.0% of general hospital. 23.5% of tertiary hospital. Second, The rate of physicians' knowledge on penalties of not reporting the NCD by their medical institution were 35.2% of clinic, 45.7% of hospital. 36.7% of general hospital. 62.7% of tertiary hospital. Third, among the no-reporting physicians in whole, the major reason of not reporting NCD were uncertainty of diagnosis(78.9%), no need to report(46.4%), no adequate actions from PHC(29.1%), no knowledge of the cases being notifiable ones in the order of their frequencies(30.4%), meddling from PHC(29.1%), concerning of patient's privacy(26.3%). 2. To analyze the characteristics related to the physicians' behaviors to report NCD, univariate and multiple logistic regression analyses were applied to the variables related to physician, 4 medical facility, PHC, and reporting system. The result were as follows: First, the result of the univariate analysis on physicians' attitude to report NCD and characteristics related to reporting in odds ratio was in the case of hospital. 3.4 times higher positive responses on physicians' attitude to report NCD came up as compared to the clinic. Second, the result of the univariate analysis on physicians' action of reporting NCD and characteristics related to reporting by the classification of medical institutions showed that the odds ratio of hospital was 2.3 times, the odds ratio of general hospital was 2.0 times, the odds ratio of tertiary was 6.8 times significantly higher than clinic. And the medical institution with significantly higher positive attitudes rate by multiple logistic regression analysis was hospital that rated 2.5 times significantly higher than clinic. Also in the PHC related characteristics of reporting, the rate of action in reporting NCD was significantly higher in medical institution that were endowed with the good condition of reporting. In multiple logistic regression analysis, the medical institution that has a good conditions of reporting showed a significantly higher positive rate on the action of reporting than the others.

• Journal of Korean Neurosurgical Society
• /
• v.63 no.5
• /
• pp.640-648
• /
• 2020

Objective : This study aimed to assess the relationship between increased intracranial pressure (ICP) and mastoid effusions (ME). Methods : Between January 2015 and October 2018, patients who underwent intracranial surgery and had ICP monitoring catheters placed were enrolled. ICP was recorded hourly for at least 3 days. ME was determined by the emergence of opacification in mastoid air cells on follow-up brain imaging. C-reactive protein (CRP) levels, presence of endotracheal tube (ETT) and nasogastric tube (NGT), duration of intensive care unit (ICU) stay, duration of mechanical ventilator application, diagnosis, surgical modalities, and presence of sinusitis were recorded. Each factor's effect on the occurrence of ME was analyzed by binary logistic regression analyses. To analyze the independent effects of ICP as a predictor of ME a multivariable logistic regression analysis was performed. Results : Total of 61 (53%) out of 115 patients had ME. Among the patients who had unilateral brain lesions, 94% of subject (43/50) revealed the ipsilateral development of ME. ME developed at a mean of 11.1±6.2 days. The variables including mean ICP, peak ICP, age, trauma, CRP, ICU stays, application of mechanical ventilators and presence of ETT and NGT showed statistically significant difference between ME groups and non-ME groups in univariate analysis. Sex and the occurrence of sinusitis did not differ between two groups. Adding the ICP variables significantly improved the prediction of ME in multivariable logistic regression analysis. Conclusion : While multiple factors affect ME, this study demonstrates that ICP and ME are probably related. Further studies are needed to determine the mechanistic relationship between ICP and middle ear pressure.

• The Journal of Society for e-Business Studies
• /
• v.20 no.3
• /
• pp.1-10
• /
• 2015

The main objective of this study was to identify the factors that can maximize the effect of preventing technology leakage by government support. Therefore we used the 2013 small business technology protection capabilities and level of research which is conducted by the Small and Medium Business Administration, and have analyzed the presence of small business technological assets leakage protection eand skills. Multiple logistic regression analysis was performed to identify 1,518 small companies (43 big companies are excluded) which are divided into 155 technological assets leaked small business and non-leaked 1363 small business. The most important factors associated with technology leakage were entrant control system, security audit, employee absence of security activities and important data protection measures. This result shows that if the government can support more for these details, technological asset leakage prevention effect is expected to be maximized.

• Health Policy and Management
• /
• v.33 no.3
• /
• pp.264-272
• /
• 2023

Background: Private health insurance supplements the coverage of national health insurance in Korea. In this situation, the subject of the study is to identify the healthcare utilization of people with physical and mental illnesses according to private health insurance. Methods: This study used data from the Korea Health Panel Survey 2018. The study population consisted of 813 individuals with physical and mental illnesses (PMI). Multiple logistic regression analysis and binominal logistic regression analysis were conducted about the utilization of emergency, inpatient, and outpatient medical services of people with PMI depending on enrollment in private health insurance (PHI). Results: The results of this study indicated that individuals with PHI utilized emergency and outpatient medical services less frequently compared to those without PHI. Conversely, having PHI was associated with a higher utilization of inpatient medical services compared to not having PHI. Binomial logistic regression analysis revealed that individuals with PMI who had PHI exhibited a higher frequency of visits to emergency and outpatient medical services compared to those without PHI. However, the significance of this trend was not observed in the case of emergency medical services. On the other hand, individuals with PMI who had PHI showed a lower frequency of visits to inpatient medical services compared to those without PHI. Conclusion: In conclusion, there was a significant relationship between having PHI and the utilization of medical services in people with PMI. There is a need for a follow-up study considering the type of mental illnesses, length of stay, and health outcome of people with PMI depending on having PHI.

• Korean Journal of Community Nutrition
• /
• v.25 no.3
• /
• pp.226-235
• /
• 2020

Objectives: Decreasing muscle strength in old age has become a significant health problem because it increases the risk of falls or fractures and transfers to other diseases. The precise role of dietary protein intake in preventing or reducing muscle weakness is unclear. This study examined the relationship between handgrip strength and protein intake in Korean female elderly. Methods: This was a cross-sectional study that used data from the Seventh Korean National Health and Nutrition Examination Surveys (KNHANES) on female subjects aged 65 years and older. Low handgrip strength (LHGS) was defined as a handgrip strength below than 18 kg. Dietary intake data were obtained using the 1-day 24-hour recall method. Multiple regression was performed to test whether there is an independent relationship between the grip strength and protein intake, and the association between protein intake and LHGS was confirmed through multiple logistic regression. Results: The mean age of the 2,083 elderly females was 73.3 ± 0.1 years, and the prevalence of LHGS was 35% (n=734). Elderly women with an LHGS consumed less energy, total protein, and animal-based protein than those in the normal group. A multiple regression analysis after adjusting for covariate revealed a significant positive association between the handgrip strength and energy, protein, and animal-based protein intake. Multiple logistic regression analysis showed that the odds ratio (OR) of LHGS in female elderly with the highest quartiles of consumption of energy [OR, 0.65; 95% confidence interval (CI), 0.43-0.82; P for trend=0.004], and animal-based protein [OR, 0.59; CI, 0.40-0.87; P for trend=0.037] were significantly lower than those in the lowest quartiles. Conclusions: The energy intake and animal-based protein intake were negatively associated with the LHGS. These results suggest that adequate energy intake and protein intake, particularly those from animal-based sources, for elderly women in Korea are beneficial in lowering the risk of LHGS.

• Korean Journal of Health Education and Promotion
• /
• v.32 no.2
• /
• pp.53-63
• /
• 2015

Objectives: The purpose of this study was to investigate association of depressive symptom, fatalism with selected health behaviors among people aged 50-69 with no physical limitation in Seoul. Methods: In 2012, data were collected using a household based interview survey in Seoul. One person in each selected household aged between 50 and 69 was selected and responded. Data analysis was based on 1,190 subjects who answered they do not have any physical activity limitation. Results: Multiple logistic regression analysis showed significant association among variables including perceived health, depressive symptom and fatalism scores on some health behaviors. Multiple regression analysis showed that selected health risk behaviors(current smoking, monthly alcohol consumption, no regular health exam in two years) were significantly associated with depressive symptom and fatality scores in addition to demographic variables. The final regression model's adjusted R square was about 0.235. Conclusion: Demographic variables such as gender, age and socioeconomic status were significant variables in health behaviors and these behavioral factors were associated with perceived health, depressive symptom and fatalistic views. As a conclusion, depressive symptom and fatalism should be monitored and intervened in health education practice.