Depression and nonadherence predict mortality in hemodialysis treated end-stage renal disease patients

The scientific evaluation of depression's impact on mortality in hemodialysis (HD) patients has yielded mixed results, with the more recent, more rigorous studies detecting a significant relationship. In this study, 130 HD patients from an urban North American hospital were evaluated for depressive affect and then observed for up to 5 years. In a corrected Cox regression model, which held constant age, gender, dialysis vintage, illness severity and diabetic status, depressive affect emerged as a modest but significant predictor of mortality (relative risk = 1.05, 95% confidence interval = 1.01-1.08). When the subjects were divided according to depressive affect severity, those with severe depressive affect had significantly shorter time to death (β = 0.452, P = 0.044). In a subgroup of 85 subjects, self-reported medication adherence was also predictive of mortality, with higher rates of nonadherence being associated with increased mortality risk. This paper lends support to the burgeoning literature on depression and reduced survival in HD populations, as well as begins the investigation of understanding the underlying mechanisms.