ABSTRACT
This chapter discusses the concepts of reliability and validity. The assessment validity of concentration biomarkers is a necessary yet challenging step, which involves the use of statistical methodology and complex study design to combine knowledge on behavioral, metabolic, laboratory, physiological, and nutritional aspects of the measured quantities. Dietary biomarkers have become increasingly popular because they can provide information about nutrients' bioavailability, defined as the available effective internal dose after absorption and precirculatory metabolism. It should be noted that the reliability of biomarker levels is based on model or on its extension, in analogy with classical measurement error model for recovery biomarkers. The absence of systematic errors in recovery biomarkers that could determine error correlations among replicates from the same subjects has been the object of research. Estimating the reproducibility is the first move of the assessment of a biomarker, a necessary step before further evaluating other quantities related to its performance, validity first of all, but also specificity and sensitivity.
