Measurement of insulin-like growth factor I (IGF-I) levels is used during the assessment of a child for the presence of growth hormone (GH) deficiency and to monitor the efficacy of GH replacement therapy. In either case, robust normative data are required to allow IGF-I values to be expressed as standard deviation scores, enabling comparison between individuals and assessment of change over time. However, IGF-I levels in healthy children are affected by a number of parameters, including age, gender, pubertal status, height, nutrient intake, body composition, intercurrent illness and ethnicity, and the generation of such data requires the collection of samples from significant numbers of healthy children. As external quality assurance schemes for IGF-I and an international standard based on authentic recombinant IGF-I are not widely used, it is imperative for the clinician to understand the performance characteristics and limitations of the IGF-I assay used and to be aware of the source and quality of control data. It must also be recognized that IGF-I measurement is only one component of the diagnostic process and has its limitations, as tissue sensitivity to circulating serum IGF-I levels will differ between individuals.