Background/Aims: The diagnosis of growth hormone (GH) deficiency in children in Canada is based on clinical, auxological, radiographic and biochemical criteria which include response to provocative GH testing. The objective of this study was to determine the variability of GH assays used at Canadian pediatric hospitals. Methods: Pooled samples of patient sera were generated at a single center and tested at 15 laboratories across Canada using local GH assays. Additional samples were analyzed using the most commonly employed assays. Results: DSL and AutoDelfia<sup>©</sup> assays measured lower for all samples. Linear regression analysis demonstrated excellent correlation between Access 2<sup>©</sup>, AutoDelfia, and Immulite 2000<sup>©</sup>, however, Bland-Altman plots revealed bias when comparing these assays. Conclusions: Adequate comparison and standardization of commercially available GH immunoassays is necessary to ensure consistent interpretation of results.