PHARMACOLOGICAL TREATMENT OF GERIATRIC DEPRESSION - KEY ISSUES IN INTERPRETING THE EVIDENCE

OBJECTIVE: A framework for critical appraisal of antidepressant trial literature involving geriatric subjects is presented. Among older adul ts, treatment decisions are complicated by comorbid medical and cognit ive illness, the variable course of recovery, and the overlap of depre ssive symptoms with other disease symptoms. Consumers of the literatur e on competing antidepressant therapies for older adults must consider disease and population-specific outcome assessment issues. DESIGN: An appraisal guide, adapted for geriatric depression, is developed from literature on methodological challenges of outcome assessment and publ ished clinical trials comparing competing antidepressant therapies in older subjects. CONCLUSIONS: The clinical utility of pharmacologic tre atment of depression can be difficult to assess because depression sca les in current use provide an important but limited perspective on tre atment outcome and because the scales vary in actual content and abili ty to detect change. The use of indicators of function, independence, and self-perceived well-being as outcomes offers additional patient re levance and should be included. Key considerations involve (1) general characteristics: whether the depression outcome measure is valid and reproducible for the aged population under study, whether the outcome measure is sensitive to treatment-related change, and whether the time -frame of outcome assessment is appropriate for the treatment goal; (2 ) cognitive impairment: how subjects with dementia are assessed, and w hether the impact of the intervention on caregivers and healthcare sys tems is considered as a relevant outcome; (3) pharmacologic issues: wh ether the comparator agent is a reasonable usual-care standard in the older adult and whether ascertainment for adverse effects is similar f or all agents under study; and (4) broader general health status issue s: whether functional or quality of life measures are used as outcome indicators.