Growth hormone therapy and short stature‐related distress: A randomized placebo‐controlled trial

CONTEXT: Growth hormone (GH) treatment of short healthy children is based on the belief that short stature is associated with psychosocial problems and a diminished quality of life. OBJECTIVE: To determine the impact of GH therapy on psychosocial well-being and the ability of psychological metrics to define short stature-related distress. METHODS: Sixty prepubertal boys with idiopathic short stature (age: 10.0 ± 1.4 years, height-SDS: -2.38 ± 0.3) were enrolled in this 4-year intervention study (1-year double-blinded, randomized, placebo-controlled [GH/placebo-2:1] and 3-year open-labelled GH therapy). Explicit (conscious/voluntary) psychological metrics (Pediatric Quality of Life Inventory [PedsQL], Silhouette Apperception Test [SAT], Rosenberg Self-Esteem Scale [RSES], Child Behavior Checklist [CBCL]) and implicit (unconscious/involuntary) psychological metrics (Single-Category Implicit Association Test for height [SC-IAT-H], Height Perception Picture Test [HPPT]). Psychosocial evaluations were performed at study entry, after 1 and 4 years. RESULTS: At study entry, PedsQL of boys with idiopathic short stature was lower than Israeli norms (P = 0.001). After 1-year blinded intervention, only the GH-treated boys improved their actual and anticipated adult height perception (SAT, P < 0.001 and P = 0.022) with reduced short stature-related distress (SC-IAT-H, P < 0.001). At study end, RSES and SC-IAT-H improved significantly (P < 0.001), with no change in PedsQL and CBCL. CONCLUSIONS: Our finding of improved psychosocial functioning only in the GH-treated boys after 1-year blinded intervention suggests that it was the GH therapy, rather than being enrolled in a clinical trial, which contributed to the outcome. Long-term open-labelled GH treatment significantly improved height perception and self-esteem. Future studies are needed to fully assess the relevance of complementing the routinely used explicit self-report measures with the implicit measures.