Objectives: To measure the prevalence of high scores on the Geriatric Depression Scale in an inpatient unit for older people and assess whether administration of this instrument increased the use of antidepressant medication.
Design: Randomized controlled trial.
Setting: Inpatient unit for assessment, treatment and rehabilitation of older adults in a district hospital.
Subjects: Consecutive admissions to the inpatient unit were approached. Of 198 people, 100 gave consent and were randomized to receive the intervention.
Interventions: The Geriatric Depression Scale and the Folstein Mini-mental State Examination were administered to the intervention group. The Nottingham Instrumental Activities of Daily Living questionnaire and the Folstein Mini-mental State Examination were administered to the placebo group. The scores of these instruments and a copy of the completed instrument were placed in the subject's case notes. For those subjects randomized to receive the Geriatric Depression Scale an interpretation of the score was written in the case notes.
Main outcome measures: The primary outcome measure was whether antidepressant medication was listed on the discharge summary for the admission. Other outcome measures were death, readmission and use of antidepressant medication three months after administration of the instruments.
Results: There was a prevalence of Geriatric Depression Scale scores greater than 10 of 52% in the 50 people administered this instrument. Six out of 46 people administered the Geriatric Depression Scale, who were discharged in the study period, were on antidepressant medication at discharge. Three out of 47 people administered the placebo instrument were on antidepressant medication at discharge. The absolute difference in proportions was 6.7%, 95% confidence interval 19 to – 5.3%.
Conclusions: A high proportion of patients admitted to an inpatient unit for assessment, treatment and rehabilitation scored in the depressed range on the Geriatric Depression Scale; however, use of the Geriatric Depression Scale in this clinical setting did not increase the use of antidepressant medication. This may be because the instrument is too nonspecific.
