Establishing reference intervals: Health as a relative concept

The use of reference values to diagnose or screen for disease implies that health is a relative concept; clinical examination, evaluation of laboratory data, and diagnostic imaging findings all require comparison to a "normal" standard. "Normality" itself is also relative. Values obtained from exotic species in the wild may vary significantly from those obtained from captive animals of the same species; similarly, sex and age, to say nothing of season and reproductive and nutritional status, often impose significant variation in "normal" values. Because health and disease are defined against "normal" or reference standards, the importance of appropriate reference values cannot be overstated. Unfortunately, in avian and reptile medicine, generation of appropriate sex-, age-, and species-specific reference values is often challenging owing to the difficulty in obtaining sufficient sample number and volume. Consequently, many clinical or diagnostic laboratories do not have their own reference values for most species, and many of the diagnostic or clinical laboratories that provide reference values have obtained those values from the literature or from other diagnostic facilities. Finding reference values for some species is sufficient victory that we often forget to ascertain the appropriateness or even validity of those values. The aims of this report are to provoke the clinician to address the appropriateness of the reference values in use, to consider the source of the reference values in clinical decision making, and to provide recommended guidelines for generating population-based reference values. Copyright (C) 2001 by W.B. Saunders Company.