We have performed stability studies on thyroid function tests, i.e., T4, T3, free T4, free T3 and thyroid stimulating hormone (TSH), over a five-day period on the sera of ten patients. We find that the levels for each analyte were stable; the mean coefficients of variation (CV) ranged from 3.2 % for TSH to 8.7% for T4. ANOVA statistical analysis of group means for each analyte over the five-day period indicated that there was no statistically significant difference in these means, confirming the reproducibility of values. On the other hand, the CVs are significantly larger than those found in our prior stability studies for other critical analytes such as electrolytes (sodium, potassium, chloride, calcium). In addition, significant variations in values over the five-day period occurred for T4 which showed the largest variations in values of the five analytes studied. In one patient, the values ranged from a low of <10 ug/dL to over 14 ug/dL over the five-day period. A two-tailed t test comparing the mean CV for TSH with the mean CVs of the other four analytes showed that the mean CV for TSH was statistically significantly lower than that for the other four analytes indicating that the assay for TSH yield the most precise results. These results parallel those obtained in our prior electrolyte study that showed very low CVs for sodium, potassium and chloride and a significantly higher CV for calcium. Since approximately half of total serum calcium is bound to albumin and since very high percentages of T4 and T3 (≥99%) are bound to thyroid binding globulin and albumin, protein binding of these analytes may introduce a possible source of assay imprecision.