Impaired GH activity at target tissues, occurring when GH action is blocked or during suboptimal GH replacement therapy, may result in a pathological state associated with lowering of IGF-I, but not GH levels. Such a state represents functional but not necessarily actual GH deficiency (GHD). The aim of this study was to identify a range of IGF-I values commensurate with GHD, which could be used to determine the risk of functional GHD during the treatment of adult GH disorders. Centrally measured baseline IGF-I data from the Kabi International Metabolic Study European GHD database were analyzed. Inclusion criteria were adult-onset GHD and two or more additional anterior pituitary hormone deficits. Adults with childhood-onset GHD and cured acromegaly were excluded. The cohort was stratified into six gender-based age ranges. Baseline IGF-I measurements from 376 females (median age, 48 yr; range, 21-77 yr) and 434 males (median age 52 yr; range 21-80 yr) were analyzed. Data were not normally distributed and are presented as medians (quartiles). The median serum IGF-I and IGF-I SDS in males were 94.0 microg/liter (64 and 141) and -1.52 (-2.53 and -0.456; n = 434). Both were significantly greater than the equivalent values of females, which were 73 microg/liter (46 and 103.5) and -2.30 (-3.28 and -1.328; n = 376; P < 0.0001 for both). Age and gender-related 90th and 95th percentiles for IGF-I SDS were determined to generate risk estimates for functional GHD, which, in conjunction with the clinical status of the patient, may be used to aid dose titration during treatment of GH disorders in adulthood.