Biostatistical Methods for Epilepsy Trials / 대한간질학회지

Clinical trial provide reliable basis for evaluating the efficacy and safty of new treatments. To proceed effectively with clinical trial requires an comprehension of the basic principles of clinical design and biostatistical methods. This review focuses on fundamentals of biostatistical theory, on studies of calculating sample size, on definitions for classsification of evidence in epilepsy trials and on examining biostatistical methods for evaluating efficacy of antiepileptic drugs. This review guide how to understand basic statistical concepts and types of study design for epilepsy trials.

Regression Methods for Overdispersed Dichotomous Response Data / 신경정신의학

In neuropsychiatrical research, many problems of statistical inference concern the relationship between the PTSD and traumatic experiences. The logistic model is widely used for modeling a relationship between the covariate and the magnitude of the PTSD. A common complication in the logistic model for dichotomous response data is overdispersion. In this study, two different methods for analyzing dichotomous response data are illustrated and compared. One method is the logistic regression approach, where the numbers of dichotomous responses are predicted by the logistic function of covariates. The other one is the overdispersed logistic regression approach, where the overdispersion is measured by a scale parameter in the variance function of the dichotomous response. In dichotomous response model, when reponses are overdispersed, the overdispersed logistic regression produces more appropriate standard errors of the regression coefficients and the 95% confidence intervals of odds ratios. Therefore, in neuropsychiatrical research, it is recommended to examine the overdispersion problems for their data set before applying the logistic regression model.

Regression Methods for Overdispersed Dichotomous Response Data / 신경정신의학

In neuropsychiatrical research, many problems of statistical inference concern the relationship between the PTSD and traumatic experiences. The logistic model is widely used for modeling a relationship between the covariate and the magnitude of the PTSD. A common complication in the logistic model for dichotomous response data is overdispersion. In this study, two different methods for analyzing dichotomous response data are illustrated and compared. One method is the logistic regression approach, where the numbers of dichotomous responses are predicted by the logistic function of covariates. The other one is the overdispersed logistic regression approach, where the overdispersion is measured by a scale parameter in the variance function of the dichotomous response. In dichotomous response model, when reponses are overdispersed, the overdispersed logistic regression produces more appropriate standard errors of the regression coefficients and the 95% confidence intervals of odds ratios. Therefore, in neuropsychiatrical research, it is recommended to examine the overdispersion problems for their data set before applying the logistic regression model.

Time-dependent Effect of Non-immunologic Factors on the Graft Survival and Graft Function in Haplotype Matched Living Donor Renal Transplant Recipients / 대한이식학회지

PURPOSE: In the analysis of risk factors affecting the renal graft survival and graft function, time-dependent effect of each risk factor should be differentiated from net effect of risk factor. We attempted to analyze the impact of immunologic and/or non-immunologic risk factors on the graft function and survival after renal transplantation among the recipients having same immunologic risks at the time of transplantation. METHODS: Three hundred ninety recipients who underwent haplotype matched living related donor kidney transplantation and have been regularly followed-up were retrospectively evaluated in a single center. All recipients were treated with cyclosporine-based double or triple regimens. The graft function was evaluated by serum creatinine (Scr) level and 24 hours urinary excretion of protein every year until 5 years after transplantation. The donor kidney weight/ recipient body weight ratio (KW/BW), donor age/ recipient age ratio (DA/RA), donor-recipient sex (D-R sex) relationship, and episodes of acute rejection (AR) within 1 year were regarded as the potential risk factors affecting the graft survival and function in this study. Kaplan-Meier method and Cox proportional-hazard model were used for survival analysis. ANOVA to evaluate time-point difference of graft function, and repeated measures ANOVA to evaluate the yearly difference of graft function were used. RESULTS: Only the episode of AR was a significant risk factor affecting the graft survival. However, each non-immunologic risk factors (KW/BW, DA/RA, D-R sex) and AR episode persistently showed statistically significant impact on Scr level until 5 years after transplantation. Recipients having lowest KW/BW (1st Q KW/BW) and highest DA/RA (4th Q DA/RA) had experienced accelerated increment of Scr level from 4th year after transplantation. From 3rd year after transplantation, there is a significant correlation between the numbers of non-immunologic risk factor the recipients having had and yearly increment of Scr level. However, episode of AR didn't influence the annual slope of Scr level even 4th year after transplantation. CONCLUSIONS: Non-immunologic risk factors had an detrimental effect on renal graft function, especially from 3rd year after transplantation. To have a better long-term graft function, non-immunologic risk factors should be considered from the time of live donor evaluation for transplantation. From the early period of transplantation, the recipients should be aware of the negative impact of overweight in terms of graft function and other metabolic derangement.