In order to examine the effect of growth hormone (GH) on psychological parameters in adult hypopituitarism, 20 adults were studied, mean age 47 years (range 20-69) with hypopituitarism and confirmed GH deficiency (GH < 3 micrograms/l following clonidine 0.15 mg/m2). Subjects were randomly allocated to either GH-up to 0.25 U/kg per week in daily doses) or placebo for 3 months before crossover to the opposite treatment arm. Psychological parameters were assessed by self-reported questionnaires (Disease Specific Questionnaire (DSQ), the Symptom Checklist-90 (SCL-90), and the Social Adjustment Scale (SAS)) at baseline and monthly intervals to the end of the trial. Both SAS and SCL-90 total scores showed a significant decline with respect to time (p = .03 and .013 respectively), although this was not a function of active treatment. DSQ showed a trend to decline (p = .06), but no effect of active treatment. None of the subscale components of the SCL-90 showed any significant change on active treatment. Improvement in psychological parameters occurs as a function of enrolment to a clinical trial, rather than active GH replacement in adult hypopituitarism.