Psychiatry has traditionally focused on the study of neurons and neurotransmitter physiology in the pathophysiology and treatment of psychiatric disorders. A growing literature highlights REDOX imbalance (a state in which demand for antioxidants surpasses their bioavailability) as a common pathophysiology for a diverse array of brain conditions (e.g., trichotillomania, schizophrenia, autism, Parkinson’s disease). REDOX imbalance is typically measured via plasma glutathione, as glutathione is critical to the adaptive antioxidant response in the brain. Accordingly, glutathione, its precursors, and/or metabolites serve as biomarkers of disease risk, therapeutic targets, and measures of treatment response. However, as with any emerging field, there are currently several different methods for collection, processing, storage, and calculation of summary measures of plasma glutathione metabolism, within and between preclinical and clinical research. The lack of evidence-based best-practice methodology hampers reproducibility (preclinical or clinical), and translation (between preclinical and clinical work). To address this methodological need, here we used a repeated measures within-subject design to investigate how sample preparation (type of anticoagulant used during blood collection, deproteinization status, and storage temperature) affects plasma glutathione levels. Accordingly, we collected whole blood from mice (N = 13), and then, using a commercially available kit, quantified glutathione in plasma stored in four different ways. Presuming that these preparation conditions and post-processing calculations are unimportant, we would expect to see no difference in glutathione levels and summary measures from the same sample. However, we found each of these variables to significantly alter quantified glutathione levels. Accordingly, we propose a vital, gold-standard methodology for both sample collection, processing, and storage of plasma used for glutathione quantification and for summary calculations of glutathione that can be used preclinically and clinically, thus yielding more streamlined translation.