Considerable controversy exists about the use of growth hormone (GH) treatment in short children without classical GH deficiency (idiopathic short stature or ISS). ISS is a multifactorial disorder with many potential causes rather than a single diagnostic entity, and it is in essence a diagnosis of exclusion. A careful investigation of the components of the GH/insulin-like growth factor (IGF) axis and other potential causes of short stature must be carried out in patients with significant short stature before they are categorized as having the ISS syndrome. As a group, most patients with ISS can be expected to attain an adult stature within the normal range. The ISS patients who will not do so are generally those who have a poor initial predicted adult height (PAH) and present at a younger age. Those ISS patients with a shorter midparental height are also less likely to achieve a normal adult height. In addition, several studies have shown that, as a group, males with ISS do not attain their initial PAH. Most studies on the GH treatment of patients with ISS have shown a short-term increase in growth rate and relative height. However, the long-term outcome of GH treatment of ISS has been variously reported to be no significant change, a slight increase, or a significant increase. One of the largest and most optimistic experiences with GH treatment of ISS is the collaborative US study sponsored by Genentech. This group has treated 121 patients with ISS not due to classically defined GH deficiency for up to 9 years with GH (0.3 mg/kg per week). A total of 51 of these patients has now reached near final height, with bone age (BA) > 16 years for boys and ≥ 14 years for girls. The mean difference between the first available pretreatment PAH and the PAH at near-final height (ΔPAH) was 5.7 cm for the boys and 6.5 cm for the girls. The strongest predictors of response to treatment available before the start of GH treatment were bone age and pretreatment PAH. The increase in height seen with GH treatment of ISS was not enough to increase the mean expected final height to the midparental target height, or to completely correct a low pretreatment PAH. Whether this increase in expected final height in ISS patients with GH treatment is due to a correction of a subtle abnormality of GH secretion or action, or to the pharmacological effects of GH is unclear. The decision to use GH treatment in ISS is a complex issue which must involve the accuracy of the diagnosis of GH deficiency or abnormality in the GH/IGF axis, the PAH and BA of the patient, the potential benefits, and economic issues for the medical care system. In general, GH treatment should only be considered in young ISS patients with poor initial PAH and height prognosis, and GH treatment of older patients who have a normal PAH is not justified.