Depressive symptoms in stroke patients: A 13 month follow-up study of patients referred to a rehabilitation unit

Abstract

Background Although depression is known to be frequently associated with stroke, it is nonetheless underdiagnosed and under-treated in this patient population. Its effect on outcome for stroke patients is thought to be substantial, but prediction is complicated by other pre- and post stroke factors. The aims of this study was to describe changes in depressive symptoms in elderly stroke patients across a timespan of one year, to examine risk factor for such changes and to explore whether depressive symptoms have any independent impact upon one year mortality and nursing home placement. Methods 194 patients diagnosed with an ischaemic or hemorrhagic stroke was recruited from the Stroke Rehabilitation Unit, Ullevaal University Hospital, Oslo, Norway during the period between March 2005 and August 2006 and followed up for a period of 13 months. Pre-stroke assessment was accomplished by means of the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE), the Frenchay Activities Index (FAI), the Barthel ADL Index and patient's medical history. Post-stroke assessment at inclusion and follow-up examination was performed with the Mini Mental State Examination (MMSE), the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), the Star Cancellation Test, the Barthel ADL Index, the modified Rankin Scale (mRS) and the National Institute of Health Stroke Scale (NIHSS). Information was collected from the patients' records. Results Institutionalization at 13 months was predicted by more depression (MADRS) and cognitive impairment (RBANS) at baseline, together with lower pre-stroke social activity levels (FAI). Two factors predicted death at 13 months: Cognitive impairment (MMSE) and greater age. The prevalence of depression was relatively unchanged from baseline (56%) to 13 month follow-up (48%). Among the patients who were depressed at baseline 55% still had MADRS score above six (persistent depression) at 13 months, while 35% in the non-depressed group at baseline had developed depression (incident depression). Persistent depression was significantly predicted by lower pre-stroke social activity levels (FAI) together with a more severe stroke (NIHSS) and worse overall function (mRS) at baseline. Incident depression was predicted by receipt of municipal home help before the stroke and a lower score on the delayed memory tasks on RBANS at baseline.