ijsmr
Editor’s Choice : A Bayesian Approach for the Cox Proportional Hazards Model with Covariates Subject to Detection Limit
|
|
Abstract: The research on biomarkers has been limited in its effectiveness because biomarker levels can only be measured within the thresholds of assays and laboratory instruments, a challenge referred to as a detection limit (DL) problem. In this paper, we propose a Bayesian approach to the Cox proportional hazards model with explanatory variables subject to lower, upper, or interval DLs. We demonstrate that by formulating the time-to-event outcome using the Poisson density with counting process notation, implementing the proposed approach in the OpenBUGS and JAGS is straightforward. We have conducted extensive simulations to compare the proposed Bayesian approach to the other four commonly used methods and to evaluate its robustness with respect to the distribution assumption of the biomarkers. The proposed Bayesian approach and other methods were applied to an acute lung injury study, in which a panel of cytokine biomarkers was studied for the biomarkers’ association with ventilation-free survival. Keywords: Bayesian, Biomarker, Detection limit, Lung Injury, Proportional hazards models.Download Full Article |
Editor’s Choice : A Bayesian Shared Parameter Model for Analysing Longitudinal Skewed Responses with Nonignorable Dropout
|
|
Abstract: When the nature of a data set comes from a skew distribution, the use of usual Gaussian mixed effect model can be unreliable. In recent years, skew-normal mixed effect models have been used frequently for longitudinal data modeling in many biomedical studies. These models are flexible for considering skewness of the longitudinal data. In this paper, a shared parameter model is considered for simultaneously analysing nonignorable missingness and skew longitudinal outcomes. A Bayesian approach using Markov Chain Monte Carlo is adopted for parameter estimation. Some simulation studies are performed to investigate the performance of the proposed methods. The proposed methods are applied for analyzing an AIDS data set, where CD4 count measurements are gathered as longitudinal outcomes. In these data CD4 counts measurements are severely skew. In application section, different structures of skew-normal distribution assumptions for random effects and errors are considered where deviance informationcriterion is used for model comparison. Keywords: Bayesian approach, Longitudinal data, Markov Chain Monte Carlo, Missingness mechanism, Nonignorable missing data, Random effects model. |
Editor’s Choice : A Nonparametric Bayesian Approach to Estimating Malaria Prophylactic Effect After Two Treatments
|
|
Abstract: Two treatment regimens for malaria are compared in their abilities to cure and combat reinfection. Bayesian analysis techniques are used to compare two typical treatment therapies for uncomplicated malaria in children under five years, not only in their power to resist recrudescence, but also how long they can postpone recrudescence or reinfection in case of failure. We present a new way of analysing this type of data using Markov Chain Monte Carlo techniques. This is done using data from clinical trials at two different centres. The results which give the full posterior distributions show that artemisinin-based combination therapy is more efficacious than sulfadoxine-pyrimethamine. It both reduced the risk of recrudescence and delayed the time until recrudescence. Keywords: Bayesian clinical trial, conditional survival posterior, drug resistance, efficacy, recurrence time, uncomplicated malaria, sulfadoxine-pyrimethamine.Download Full Article |
Editor’s Choice : A Dynamical Study of Risk Factors in Intracerebral Hemorrhage using Multivariate Approach
|
|
Abstract: The purpose of this study is to investigate the effects of clinical covariates to the outcome of Intracerebral Hemorrhage (ICH) patients in terms of best fitted and excellent discriminate model of binary response variable. Clinical data of 985 patients with ICH have collected using the International classification of diseases, Ninth revision codes. The diagnosis of ICH was confirmed by neuro-imaging in all patients. Univariate analysis revealed that out of 88 covariates 46 were found to be significant (p<0.05). The multivariable analysis using multiple logistic regressions, exhibited a significant negative relationship between ICH and hypertension. The improvement among ICH patients having hypertension was 0.5 (p=0.001, ARR=0.5, 95% C.I. 0.3 – 0.8). The improvement among ICH patients using antihypertensive medicine was 1.3 (p = 0.016, ARR=1.3, 95% C.I. 1.1 – 1.5). Thus present study showed that ICH has strong relationship with use of antihypertensive medicine. The improvement of patients who were using antihypertensive medicine at the time of discharge was 3.0 times (p < 0.0001, ARR=3.0, 95% C.I. 2.7 – 3.2) as compared to those who did not use antihypertensive medicine. The change in ARR from 1.3 to 3.0 times shows that the use of antihypertensive medicine and ICH outcome variable are positively associated. The change in ARR of hypertensive range of SBP also indicates that the blood pressure range and ICH outcome variable are negatively associated. The neurological symptomatology, slurred speech and double vision are important factors of proposed statistical models. Moreover, a clear decrease was found in mental status from normal to coma in applicable model. Surgery is an important part of recovery, and estimated that the improvement among the ICH patients, who were treated with surgery, was 1.4 times with significant p-value in best fitted models. The complication of pneumonia during treatment of ICH subjects has highly significant negative association with outcome variable. Present Model has 0.892 area under the curve with sensitivity (0.852), specificity (0.793) and p-value (0.204). This indicates that the model gives the impression to fit quite well for predictive performance of the ICH outcome variable and the model is excellent model. Keywords: Intracerebral Hemorrhage, clinical covariates, multivariable analysis, logistic regression, discriminate model, sensitivity and specificity.Download Full Article |
|
|
Abstract: Background: An accurate tool with good discrimination for bleeding would be useful to clinicians for improved management of all their patients. Bleeding risk models have been published but not externally validated in independent clinical dataset. We chose the NCDR PCI score to validate within a large, multi-site community datasets. The aim of the study was to determine the diagnostic utility of this bleeding risk score tool. Methods: This is a large-scale retrospective analysis utilizing American College of Cardiology data from a 37-hospital health system. The central repository of PCI procedures between 6-1-2009 and 6-30-2012 was utilized to validate the NCDR PCI bleeding risk score (BRS) among 4693 patients. The primary endpoint was major bleeding. Discriminant analysis calculating the receiver operating characteristic curve was performed. Results:There were 143 (3.0%) major bleeds. Mean bleeding risk score was 14.7 (range 3 – 42). Incidence of bleeding by risk category: low (0.5%), intermediate (1.7%), and high risk (7.6%). Patients given heparin had 113 (3.7%) major bleeds and those given bivalirudin had 30 (2.1%) major bleeds. Tool accuracy was poor to fair (AUC 0.78 heparin, 0.65 bivalirudin). Overall accuracy was 0.71 (CI: 0.66-0.76). Accuracy did not improve when confined to just the intermediate risk group (AUC 0.58; CI: 0.55-0.67). Conclusion:Bleeding risk tools have low predictive value. Adjustment for anticoagulation use resulted in poor discrimination because bivalirudin differentially biases outcomes toward no bleeding. The current state of bleeding risk tools provides little support for diagnostic utility in regards to major bleeding and therefore have limited clinical applicability. Keywords: Major bleeding, bleeding risk model, anticoagulant, percutaneous coronary intervention, cardiovascular. |