Severe growth hormone (GH) deficiency (GHD) induces a well-defined clinical entity encompassing, amongst the most reported features, abnormalities of body composition, in particular increased fat mass, especially truncal, and reduced lean body mass. The results from virtually all treatment studies are in agreement that GH replacement improves the body composition profile of GHD patients by increasing lean body mass and reducing fat mass. More recently, the observations have been extended to adults with partial GHD, defined by a peak GH response to insulin-induced hypoglycaemia of 3–7 µg/l. These patients exhibit abnormalities of body composition similar in nature to those described in adults with severe GHD; these include an increase in total fat mass of around 3.5 kg and a reduction of lean body mass of around 5.5 kg. The increase in fat mass is predominantly distributed within the trunk. The degree of abnormality of body composition is intermediate between that of healthy subjects and that of adults with GHD. The impact of GH replacement on body composition in adults with GH insufficiency, although predictable, has not been formally documented. The skeleton is another biological endpoint affected by GH status: in adults with severe GHD, low bone mass has been reported using dual energy x-ray absorptiometry (DEXA) and other quantitative methodologies. The importance of low bone mass, in any clinical setting, is as a surrogate marker for the future risk of fracture. Several retrospective studies have documented an increased prevalence of fractures in untreated GHD adults. Hypopituitary adults with severe GHD have reduced markers of bone turnover which normalize with GH replacement, indicating that GH, directly or via induction of insulin-like growth factor-I, is intimately involved in skeletal modelling. Whilst the evidence that GH plays an important role in the acquisition of bone mass during adolescence and early adult life is impressive, the impact of GHD acquired later in adulthood is less clear. Recently we examined the relationship between bone mineral density (BMD) and age in 125 untreated adults with severe GHD using DEXA. A significant positive correlation was observed between BMD (z-scores) and age at all skeletal sites studied. Overall, few patients, except those aged less than 30 years, had significantly reduced bone mass (ie a BMD z-score of less than–2); correction of BMD to provide a pseudo-volumetric measure of BMD suggested that reduced stature of the younger patients may explain, at least in part, this higher frequency of subnormal BMD z-scores. Despite normal BMD, however, an increase in fracture prevalence may still be observed in elderly GHD adults as a consequence of increased falls related to muscle weakness and visual field defects.