The aims of this study were to evaluate the efficacy and safety of different doses of growth hormone (GH) treatment in prepubertal short children born small-for-gestational-age (SGA). Forty-eight children born SGA from Sweden, Finland, Denmark and Norway were randomly allocated to three groups: a control group of 12 children received no treatment for 2 y, one group was treated with GH at 0.1 IU/kg/d (n=16), and one group was treated with GH at 0.2 IU/kg/d (n=20). In total 42 children completed 2 y of follow-up, and 24 children from the treated groups completed 3 y of treatment. Their mean (SD) age at the start of the study was 4.69 (1.61) y and their mean (SD) height was -3.16 (0.70) standard deviation scores (SDS). The children remained prepubertal during the course of the study. No catch-up growth was observed in the untreated group, but a clear dose-dependent growth response was found in the treated children. After the third year of treatment, the group receiving the higher dose of GH, achieved their target height. The major determinants of the growth response were the dose of GH used, the age at the start of treatment (the younger the child, the better the growth response) and the family-corrected individual height deficit (the higher the deficit, the better the growth response). Concentration of insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 increased during treatment. An increase in insulin levels was found without negative effects on fasting glucose levels or glycosylated haemoglobin levels. GH treatment was well tolerated. In conclusion, short prepubertal children born SGA show a dose-dependent growth response to GH therapy, and their target height SDS can be achieved within 3 y of treatment given GH at 0.2 IU/kg/d. However, the long-term benefit of different regimens of GH treatment in children born SGA remains to be established.