More than 30 years after their introduction, growth hormone (GH) immunoassays showed the poorest inter-laboratory agreement of the various hormone assays evaluated in 1998 by the UK National External Quality Assessment Scheme, in which different laboratories using different assays reported that analyses of identical samples differed two- to threefold in value. There is therefore an urgent requirement and desire within the scientific community, particularly within centres diagnosing and treating GH deficiency and acromegaly, to resolve this problem and to develop a GH assay(s) that measures solely all of the relevant components of circulating GH immunoreactivity. The main confounders in the estimation of GH levels (now that the use of GH standards other than that recommended by the World Health Organization has largely been eliminated) are GH heterogeneity, anti-GH antiserum binding site specificity and interference from circulating high-affinity GH-binding protein (GHBP). The effects of these factors are closely related. The present study investigates these factors, focussing on the influence of GHBP and antibody binding site specificity on various assays for GH. The findings lead the authors to suggest that a solution to the problem may be to develop a GH assay that measures specifically and solely all serum 22 kDa GH, as this is the major circulating fraction and carries the dominant GH bioactivity.