Evaluación de la validez y confiabilidad de una prueba diagnóstica

Abstract

This review is intended to explain briefly, the significance of the validity of a diagnostic test by using specific indicators such as: the sensitivity, and the specificity, which are considered as intrinsic properties of the test. Additionally, the fundamental concepts of positive and negative predictive values of a diagnostic test in the clinical practice, are described and highlighted, and their connection to the prevalence of a particular disease. The likelihood ratio, meaning how many times a healthy person suffers from a disease, in other words, the percentage of ill people diagnosed with a given test result, divided by the percentage of healthy individuals with the same results, Ideally, abnormal test results should be much more typical in ill individuals, than in those who are healthy (high likelihood ratio), and normal test results should be more frequent in healthy people rather than in sick people (low likelihood ratio). Furthermore, the receiver operating characteristic curve (ROC) is described as a methodology to categorize, in a dichotomy form, as positive or negative those tests with values in continue scale. Finally, the reliability or reproducibility, which is related to the amount of error in any measurement, is analyzed. A more formal definition of reliability is variability between subjects, divided by inter-subject variability plus measurement, which could give as the best reliability of the test in clinical practice.