A biomarker is defined as "a trait that is objectively measured and analysed as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention" by the National Institutes of Health Biomarkers Definitions Working Group in 1998. Biomarkers are now so widely used in scientific and clinical research, as well as clinical practice, that their inclusion as primary endpoints in clinical trials is nearly unquestioned. A large subgroup of medical indicators – that is, objective signals of medical state viewed from outside the patient – that can be assessed accurately and reproducibly is referred to as a "biomarker," a portmanteau of "biological marker." Medical signals are distinct from medical symptoms, which are limited to the signs of health or illness that patients perceive. Biomarkers are objective, quantifiable aspects of biological processes by definition. They may or may not correlate with a patient's experience and sense of well-being, and it's possible to conceive measurable biological traits that don't correspond to a patient's clinical state, or whose fluctuations are undetectable and have no impact on health. It's also easy to conceive observable biological features with such wide variation among populations that they're all but useless as accurate predictors of disease or absence.