Table of Contents Author Guidelines Submit a Manuscript
Computational and Mathematical Methods in Medicine
Volume 2016 (2016), Article ID 7141050, 7 pages
http://dx.doi.org/10.1155/2016/7141050
Research Article

Alternative Confidence Interval Methods Used in the Diagnostic Accuracy Studies

Department of Biostatistics and Bioinformatics, Faculty of Medicine, Mersin University, 33343 Mersin, Turkey

Received 29 February 2016; Revised 11 May 2016; Accepted 5 June 2016

Academic Editor: Po-Hsiang Tsui

Copyright © 2016 Semra Erdoğan and Orekıcı Temel Gülhan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background/Aim. It is necessary to decide whether the newly improved methods are better than the standard or reference test or not. To decide whether the new diagnostics test is better than the gold standard test/imperfect standard test, the differences of estimated sensitivity/specificity are calculated with the help of information obtained from samples. However, to generalize this value to the population, it should be given with the confidence intervals. The aim of this study is to evaluate the confidence interval methods developed for the differences between the two dependent sensitivity/specificity values on a clinical application. Materials and Methods. In this study, confidence interval methods like Asymptotic Intervals, Conditional Intervals, Unconditional Interval, Score Intervals, and Nonparametric Methods Based on Relative Effects Intervals are used. Besides, as clinical application, data used in diagnostics study by Dickel et al. (2010) has been taken as a sample. Results. The results belonging to the alternative confidence interval methods for Nickel Sulfate, Potassium Dichromate, and Lanolin Alcohol are given as a table. Conclusion. While preferring the confidence interval methods, the researchers have to consider whether the case to be compared is single ratio or dependent binary ratio differences, the correlation coefficient between the rates in two dependent ratios and the sample sizes.