• Communications for Statistical Applications and Methods
• /
• v.24 no.6
• /
• pp.583-604
• /
• 2017

The most popular regression model for the analysis of time-to-event data is the Cox proportional hazards model. While the model specifies a parametric relationship between the hazard function and the predictor variables, there is no specification regarding the form of the baseline hazard function. A critical assumption of the Cox model, however, is the proportional hazards assumption: when the predictor variables do not vary over time, the hazard ratio comparing any two observations is constant with respect to time. Therefore, to perform credible estimation and inference, one must first assess whether the proportional hazards assumption is reasonable. As with other regression techniques, it is also essential to examine whether appropriate functional forms of the predictor variables have been used, and whether there are any outlying or influential observations. This article reviews diagnostic methods for assessing goodness-of-fit for the Cox proportional hazards model. We illustrate these methods with a case-study using available R functions, and provide complete R code for a simulated example as a supplement.

• Nutrition Research and Practice
• /
• v.12 no.3
• /
• pp.258-264
• /
• 2018

BACKGROUND/OBJECTIVES: Diet quality scores or indices, based on dietary guidelines, are used to summarize dietary intake into a single numeric variable. The aim of this study was to examine the association between the modified diet quality index for Koreans (DQI-K) and mortality among Health Examinees-Gem (HEXA-G) study participants. SUBJECTS/METHODS: The DQI-K was modified from the original diet quality index. A total of 134,547 participants (45,207 men and 89,340 women) from the HEXA-G study (2004 and 2013) were included. The DQI-K is based on eight components: 1) daily protein intake, 2) percent of energy from fat, 3) percent of energy from saturated fat, 4) daily cholesterol intake, 5) daily whole-grain intake, 6) daily fruit intake, 7) daily vegetable intake, and 8) daily sodium intake. The association between all-cause mortality and the DQI-K was examined using Cox proportional hazard regression models. Hazard ratios and confidence intervals were estimated after adjusting for age, gender, income, smoking status, alcohol drinking, body mass index, and total energy intake. RESULTS: The total DQI-K score was calculated by summing the scores of the eight components (range 0-9). In the multivariable adjusted models, with good diet quality (score 0-4) as a reference, poor diet quality (score 5-9) was associated with an increased risk of all-cause mortality (hazard ratios = 1.23, 95% confidence intervals = 1.06-1.43). Moreover, a one-unit increase in DQI-K score resulted in a 6% higher mortality risk. CONCLUSIONS: A poor diet quality DQI-K score was associated with an increased risk of mortality. The DQI-K in the present study may be used to assess the diet quality of Korean adults.

• Journal of Korean Society of Forest Science
• /
• v.97 no.3
• /
• pp.326-331
• /
• 2008

This study was carried out to analyze the characteristics of each factors by using the quantification theory(I) for prediction of landslide hazard area. The results obtained from this study were summarized as follows; The stepwise regression analysis between landslide sediment ($m^3$ ) and environmental factors, factors affecting landslide sediment ($m^3$ ) were high in order of mixed (forest type), < 15 cm(soil depth), 801~1,200 m (altitude), $31{\sim}40^{\circ}$ (slope gradient), 46 cm < (soil depth), 1,201 m < (altitude) and s(aspect). According to the range, it was shown in order of soil depth (0.3784), altitude (0.2876), forest type (0.2409), slope gradient (0.1728) and aspect (0.1681). The prediction of landslide hazard area was estimated by score table of each category. The extent of prediction score was 0 to 1.2478, and middle score was 0.6239. Class I was over 1.1720, class II was 0.7543 to 0.1719, class III was 0.4989 to 0.7542 and class IV was below 0.4988.

• Journal of Preventive Medicine and Public Health
• /
• v.22 no.3 s.27
• /
• pp.355-367
• /
• 1989

This study examined the physical and mental health factors affecting the industrial accidents of 142 injured and 1,212 uninjured workers in the shipbuilding industry from 1986 to 1988. The results acquired from the Todai Health Index (THI) and from analysis of the health examination were as follows: 1. Among the personal characteristics of the workers, the educational level of injured workers was significantly lower than that of the uninjured workers. 2. Among the physical characteristics, vision and $R\ddot{o}hrer$ Index of the injured workers were lower than those of the uninjured workers, and the difference was statistically significant. On the other hand, the differences in height, weight, hearing function, hematocrit, blood pressure, urine test, and X-ray findings were not statistically significant between the injured and uninjured workers. 3. The score of the THI questionnaire on the physical and mental health of the injured workers was higher than that of the uninjured workers, and the difference was statistically significant. 4. Form the THI score, the industrial workers had complained more about mental health than physical health and there was a statistically singinficant relation with the industrial accidents. 5. The relative risk expressed in terms of the odds ratio was 2.9 for poorer vision, 2.7 for a lower educational level, 2.2 for a higher THI score and 1.6 for overdrinking. 6 Educational level, vision, and the THI score were selected as significant factors influencing industrial accidents based on a log-linear model. According to the results of this model by logistic analysis, the odds ratio of industrial accidents was 1.8 for a lower educational level, 1.7 for poorer vision, and 1.6 for a higher THI score. 7 By event history analysis with the dependent variable as the duration of work at the time of the industrial accident, educational level, age, $R\ddot{o}hrer$ Index and THI score were the statistically significant variables selected, and the hazard rate of industrial accident occurrence was 0.24 for a lower educational level, 0.92 for age, 0.99 for a lower $R\ddot{o}hrer$ Index and 2.72 for a higher THI score. As we have seen, educational level and THI score were the most significant factors affecting the hazard rate of industrial accidents. Vision, $R\ddot{o}hrer$ Index, age, and drinking behavior were also statistically significant variables influencing industrial accidents. Therefore, in order to prevent industrial accidents, it is necessary to establish a health management plan for industry which can objectively evaluate not only the physical but also the mental health of the workers. If we use this type of study as a prospective study design, we can determine the relative risk of physical and mental health factors on industrial accidents. Furthermore, it is expected that this type of study will provide workers at high risk with more precise basic data for a health managment plan for industrial accident prevention.

• Korean Journal of Radiology
• /
• v.22 no.9
• /
• pp.1514-1524
• /
• 2021

Objective: To develop a radiomics risk score based on dynamic contrast-enhanced (DCE) MRI for prognosis prediction in patients with glioblastoma. Materials and Methods: One hundred and fifty patients (92 male [61.3%]; mean age ± standard deviation, 60.5 ± 13.5 years) with glioblastoma who underwent preoperative MRI were enrolled in the study. Six hundred and forty-two radiomic features were extracted from volume transfer constant (K^trans), fractional volume of vascular plasma space (V_p), and fractional volume of extravascular extracellular space (V_e) maps of DCE MRI, wherein the regions of interest were based on both T1-weighted contrast-enhancing areas and non-enhancing T2 hyperintense areas. Using feature selection algorithms, salient radiomic features were selected from the 642 features. Next, a radiomics risk score was developed using a weighted combination of the selected features in the discovery set (n = 105); the risk score was validated in the validation set (n = 45) by investigating the difference in prognosis between the "radiomics risk score" groups. Finally, multivariable Cox regression analysis for progression-free survival was performed using the radiomics risk score and clinical variables as covariates. Results: 16 radiomic features obtained from non-enhancing T2 hyperintense areas were selected among the 642 features identified. The radiomics risk score was used to stratify high- and low-risk groups in both the discovery and validation sets (both p < 0.001 by the log-rank test). The radiomics risk score and presence of isocitrate dehydrogenase (IDH) mutation showed independent associations with progression-free survival in opposite directions (hazard ratio, 3.56; p = 0.004 and hazard ratio, 0.34; p = 0.022, respectively). Conclusion: We developed and validated the "radiomics risk score" from the features of DCE MRI based on non-enhancing T2 hyperintense areas for risk stratification of patients with glioblastoma. It was associated with progression-free survival independently of IDH mutation status.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.2
• /
• pp.54-66
• /
• 2006

Korea Forest Service made the landslide hazard map for all mountainous districts over the country in May 2005. In this study, we selected landslide areas occurred in Jeonbuk from 02 August 2005 to 03 August 2005 as the study area. We extracted landslide areas using images taken by PKNU 3 System, which was developed by PE&RS Laboratory in Dept. of Satellite Information Sciences, Pukyong National University and verified the accuracy of landslide hazard map by overlaying landslide hazard areas extracted by PKNU 3 images. And we analyzed characteristics of an altitude, a gradient, an inclined direction, a flow length, a flow accumulation for landslide areas using mountainous terrain analysis and Stream Network analysis of ArvView 3.3. As a result of this study, it is necessary to adjust the unitage(%) by the class and to modify and improve the score table for prediction of landslide-susceptible area forming the foundation of making the landslide hazard maps.

• Journal of Korean Neurosurgical Society
• /
• v.60 no.2
• /
• pp.239-249
• /
• 2017

Objective : The purpose of this study is to evaluate the associations between 30-day mortality and various radiological and clinical factors in patients with traumatic acute subdural hematoma (SDH). During the 11-year study period, young patients who underwent surgery for SDH were followed for 30 days. Patients who died due to other medical comorbidities or other organ problems were not included in the study population. Methods : From January 1, 2004 to December 31, 2014, 318 consecutive surgically-treated traumatic acute SDH patients were registered for the study. The Kaplan-Meier method was used to analyze 30-day survival rates. We also estimated the hazard ratios of various variables in order to identify the independent predictors of 30-day mortality. Results : We observed a negative correlation between 30-day mortality and Glasgow coma scale score (per 1-point score increase) (hazard ratio [HR], 0.60; 95% confidence interval [CI], 0.52-0.70; p<0.001). In addition, use of antithrombotics (HR, 2.34; 95% CI, 1.27-4.33; p=0.008), history of diabetes mellitus (HR, 2.28; 95% CI, 1.20-4.32; p=0.015), and accompanying traumatic subarachnoid hemorrhage (hazard ratio, 2.13; 95% CI, 1.27-3.58; p=0.005) were positively associated with 30-day mortality. Conclusion : We found significant associations between short-term mortality after surgery for traumatic acute SDH and lower Glasgow Coma Scale scores, use of antithrombotics, history of diabetes mellitus, and accompanying traumatic subarachnoid hemorrhage at admission. We expect these findings to be helpful for selecting patients for surgical treatment of traumatic acute SDH, and for making accurate prognoses.

• The Korean journal of internal medicine
• /
• v.39 no.2
• /
• pp.306-317
• /
• 2024

Background/Aims: To determine whether metformin, which is considered a host-directed therapy for tuberculosis (TB), is effective in improving the prognosis of patients with TB and diabetes mellitus (DM), who have higher mortality than those without DM. Methods: This cohort study included patients who were registered as having TB in the National Tuberculosis Surveillance System. The medical and death records of matched patients were obtained from the National Health Information Database and Statistics Korea, respectively, and data from 2011 to 2017 were collected retrospectively. We classified patients according to metformin use among participants who used diabetes drugs for more than 28 days. The primary outcome was all-cause mortality during TB treatment. Double propensity score adjustment was applied to reduce the effects of confounding and multivariable Cox proportional hazard models were used to estimate adjusted hazard ratio (aHR) with 95% confidence interval (CI). Results: The all-cause mortality rate during TB treatment was lower (9.5% vs. 12.4%, p < 0.01) in the metformin user group. The hazard of death due to all causes after double propensity score adjustment was also lower in the metformin user group (aHR 0.76, 95% CI 0.67-0.86, p < 0.01). There was no significant difference in mortality between metformin users and non-users for TB-related deaths (p = 0.22); however, there was a significant difference in the non-TB-related deaths (p < 0.01). Conclusions: Metformin use in patients with TB-DM co-prevalence is associated with reduced all-cause mortality, suggesting the potential for metformin adjuvant therapy in these patients.

• Communications for Statistical Applications and Methods
• /
• v.3 no.3
• /
• pp.113-123
• /
• 1996

Data-dependent (adaptive) choice of asymptotically efficient score functions for rank estimators and M-estimators of regression parameters in a linear regression model with left-truncated and right-censored data are developed herein. The locally adaptive smoothing techniques of Muller and Wang (1990) and Uzunogullari and Wang (1992) provide good estimates of the hazard function h and its derivative h' from left-truncated and right-censored data. However, since we need to estimate h'/h for the asymptotically optimal choice of score functions, the naive estimator, which is just a ratio of estimated h' and h, turns out to have a few drawbacks. An altermative method to overcome these shortcomings and also to speed up the algorithms is developed. In particular, we use a subroutine of the PPR (Projection Pursuit Regression) method coded by Friedman and Stuetzle (1981) to find the nonparametric derivative of log(h) for the problem of estimating h'/h.

• Journal of the Korea Safety Management & Science
• /
• v.10 no.3
• /
• pp.1-8
• /
• 2008

Risk is the probability of an adverse event given exposure to hazard. There are many reason for unsafety situation without safety operation. The reason is no safety evaluation system in small enterprise. And then this study purposes safety management activities that is evaluation system for total safety efficiency's maximization. Therefore, in this study, this model that can evaluate quantitative activities in small enterprise that maximize safety efficiency wishes to do design using balanced scorecard. In other words, this study aims to suggest a performance measurement model reflecting the characteristics of safety evaluation system, especially the model for return manufacturing related to safety, and to develop the S-BSC(Safety-Balanced ScoreCard) measurement model using a weight lifetime value to which a relative weight is applied by using AHP based on the BSC.