Growth hormone (GH) in human serum has been measured for more than 30 years and during this period it has been used diagnostically in growth disturbances and GH disorders. The first GH determinations were by radioimmunoassays using polyclonal antibodies. Due to the low specificity of these assays and the variable molecular appearance of GH, higher levels were obtained than with later assays employing two highly specific monoclonal antibodies. Because of this variability between assays, empiric GH cutoff levels had to be developed in different laboratories. However, with the growing need for more sophisticated diagnosis, especially in adulthood GH deficiency, knowledge of factors influencing the individual GH assays is required. Since GH deficiency in adulthood is now treated with GH substitution therapy, the relevance of factors influencing GH estimates has become increasingly evident, because no specific symptoms exist in adult GH deficiency.