We review evidence for and against the use of erythrocyte indicators of health status and condition, parasite infection level and physiological stress in free-living vertebrates. The use of indicators that are measured directly from the blood, such as haemoglobin concentration, haematocrit and erythrocyte sedimentation rate, and parameters that are calculated from multiple measured metrics, such as mean cell volume, mean cell haemoglobin content or mean cell haemoglobin concentration is evaluated. The evidence for or against the use of any given metric is equivocal when the relevant research is considered in total, although there is sometimes strong support for using a particular metric in a particular taxon. Possibly the usefulness of these metrics is taxon, environment or condition specific. Alternatively, in an uncontrolled environment where multiple factors are influencing a metric, its response to environmental change will sometimes, but not always, be predictable. We suggest that (i) researchers should validate a metric's utility before use, (ii) multiple metrics should be used to construct an overall erythrocyte profile for an individual or population, (iii) there is a need for researchers to compile reference ranges for free-living species, and (iv) some metrics which are useful under controlled, clinical conditions may not have the same utility or applicability for free-living vertebrates. Erythrocyte metrics provide useful information about health and condition that can be meaningfully interpreted in free-living vertebrates, but their use requires careful forethought about confounding factors.