ijsmr
|
|
Abstract: In assessing the performance of a diagnostic test, the widely used classification technique is the Receiver Operating Characteristic (ROC) Curve. The Binormal model is commonly used when the test scores in the diseased and healthy populations follow Normal Distribution. It is possible that in real applications the two distributions are different but having a continuous density function. In this paper we considered a model in which healthy and diseased populations follow half normal and exponential distributions respectively, hence named it as the Hybrid ROC (HROC) Curve. The properties and Area under the curve (AUC) expressions were derived. Further, to measure the distance between the defined distributions, a popular divergence measure namely Kullback Leibler Divergence (KLD) has been used. Simulation studies were conducted to study the functional behavior of Hybrid ROC curve and to show the importance of KLD in classification. Keywords: AUC, Exponential distribution, Half-Normal distribution, Hybrid ROC Curve, Kullback-Leibler Divergence.Download Full Article |
|
|
Abstract: Aim: The goal of this study is to specify the risks, family and environmental factors affecting smoking behavior and develop suggestions for Turkish individuals by considering sibling data. Materials and Methods:The data was collected by voluntary senior year students attending Kırıkkale University, Department of Statistics. The sample of 751 families was selected from families with at least two children. Each sibling’s socio-demographic information and behavioral phenotypes were collected using a survey from both siblings. We selected one of siblings randomly as ‘sibling1’ and defined the other sibling as ‘sibling2’. Hypothesis testing and multivariable clustered logistic regression models were used to evaluate the data and find the optimum model by using dependent sibling data. Results: Out of 1502 (751 pairs) siblings 843 (56.1%) were males, 659 (43.9%) were females. According to the survey results, 508 of the males (67.7%) and 242 of the females (32.3%) were smokers for a month or longer and smoked every day. The risk of smoking was 2.26 times higher in males than in females. Having a smoking sibling increased the risk of smoking 1.95 times, alcohol using increased the risk 2.11 times. We found that when the age difference between siblings is 0-7 years, having a same sex sibling who smokes increases one’s risk 4.7 times in females and 5 times in males; when the siblings are of different sexes, according to these age differences Conclusion: The survey showed that the gender and sibling’s and parent’s smoking both play a significant role on smoking behavior. But children seem to learn smoking from their siblings more than from parents. Having same sex sibling who smokes plays significant role in smoking behavior for both males and females. Keywords: Clustered logistic regression, FTND (Fagerström Test for Nicotine Dependence), Nicotine dependence, Paired data, Smoking status, Siblings, Turkey. |
|
|
Abstract: Disease risk prediction tools are used for population screening and to guide clinical care. They identify which individuals have particularly elevated risk of disease. The development of a new risk prediction tool involves several methodological components including: selection of a general modelling framework and specific functional form for the new tool, making decisions about the inclusion of risk factors, dealing with missing data in those risk factors, and performing validation checks of a new tool’s performance. There have been many methodological developments of relevance to these issues in recent years. Developments of importance for disease detection in humans were reviewed and their uptake in risk prediction tool development illustrated. This review leads to guidance on appropriate methodology for future risk prediction development activities. Keywords: Disease risk prediction, missing data, model validation, model updating, model utility.Download Full Article |
|
|
Abstract: Cluster-randomized clinical trials (CRT) are trials in which the unit of randomization is not a participant but a group (e.g. healthcare systems or community centers). They are suitable when the intervention applies naturally to the cluster (e.g. healthcare policy); when lack of independence among participants may occur (e.g. nursing home hygiene); or when it is most ethical to apply an intervention to all within a group (e.g. school-level immunization). Because participants in the same cluster receive the same intervention, CRT may approximate clinical practice, and may produce generalizable findings. However, when not properly designed or interpreted, CRT may induce biased results. CRT designs have features that add complexity to statistical estimation and inference. Chief among these is the cluster-level correlation in response measurements induced by the randomization. A critical consideration is the experimental unit of inference; often it is desirable to consider intervention effects at the level of the individual rather than the cluster. Finally, given that the number of clusters available may be limited, simple forms of randomization may not achieve balance between intervention and control arms at either the cluster- or participant-level. In non-clustered clinical trials, balance of key factors may be easier to achieve because the sample can be homogenous by exclusion of participants with multiple chronic conditions (MCC). CRTs, which are often pragmatic, may eschew such restrictions. Failure to account for imbalance may induce bias and reducing validity. This article focuses on the complexities of randomization in the design of CRTs, such as the inclusion of patients with MCC, and imbalances in covariate factors across clusters. Keywords: Experimental Design, Randomization, Cluster Randomized Trials, Multiple Chronic Conditions.Download Full Article |
|
|
Abstract: In medical research, frequently other important determinants, correlated with the key treatment variable, are omitted from the analysis. This omission yields biased and inconsistent estimates. For example, leaving out correlated (with, say, ) determinants of Y from regressions yield biased estimates of key parameters (say ). Instrumental variable estimation solves this problem by constructing similar triangles to retrieve consistent estimates. This article illustrates the geometry of correlated regressor bias, and the simple IV geometric solution. Keywords: Omitted variable bias (OVB), classical measurement error (CME), simultaneous equation models (SEM), instrumental variables, orthogonal projections.Download Full Article |