BACKGROUND - The relationship between severity of depressive symptoms reported by incident dialysis patients and first-year outcomes is not known.
METHODS - We evaluated the association between self-report of depressive symptoms in incident hemodialysis patients admitted at Fresenius Medical Care North America facilities between 1 January and 31 December 2006 and mortality or withdrawal from dialysis for up to 1 year after the initial survey. The impact of depression scores calculated from two Short Form-36 (SF-36) questionnaires was determined independently of the mental and physical component scores, case-mix and laboratory variables using stepwise Cox models.
RESULTS - We received 6415 SF-36 responses within 46±24 days of first dialysis from a cohort with a mean age of 62.3±15.2 years; 58% were diabetic, 45% were female and 69% were Caucasian. A 1-point increase in depression score was associated with unadjusted hazard ratio (HR) of 1.09 (1.03, 1.15) for mortality and 1.15 (1.05, 1.26) for withdrawal from dialysis. After adjustment, a 1-point increase in depression score had a mortality HR of 1.08 (1.01, 1.14) and for withdrawal 1.19 (1.08, 1.31).
CONCLUSIONS - Depressive symptoms reported within the first 90 days of dialysis were associated with greater dialysis withdrawal and mortality risk over the succeeding year. Whether further evaluation for and treatment of depression during this early vulnerable period may improve symptoms, increase survival and decrease premature withdrawal from dialysis requires confirmation in prospective clinical trials.