Background: Mortality rates for individuals on chronic hemodialysis remain very high; therefore, strategies are needed to identify individuals at greatest risk for mortality so preventive strategies can be implemented. One such approach is to stratify individuals by self-reported mental health and physical function. Examining these parameters at baseline, and over time, may help identify individuals at greater risk for mortality.
Methods: We enrolled 14,815 individuals with end-stage renal disease (ESRD) and followed these individuals for up to 2 years. The mean age was 61.0 +/- 15.4 years (range, 20 to 96 years) and 31% were African Americans. The SF-36 Health Survey was administered 1 to 3 months after hemodialysis initiation and 6 months later. We examined the associations between the initial SF-36 Health Survey mental component summary (MCS) and physical component summary (PCS) scores and mortality during the follow-up period, and examined the associations between 6-month decline in PCS and MCS scores and subsequent mortality. We also examined the interactions between age and MCS and PCS scores. The general population-based mean of each of these scores was 50 with a standard deviation of 10. The main outcome measurement was death.
Results: Self-reported baseline mental health (MCS score) and physical function (PCS score) were both independently associated with increased mortality, and 6-month decline in these parameters was also associated with increased mortality. The multivariate hazard ratios for 1-year mortality for MCS scores of less than 30, 30 to 39, and 40 to 49 were 1.48 (95% CI, 1.32 to 1.64), 1.23 (95% CI, 1.14 to 1.32) and 1.18 (95% CI, 1.10 to 1.26) compared with a MCS score of 50 or more. The hazard ratios for PCS scores of less than 20, 20 to 29, and 30 to 39 were 1.97 (95% CI, 1.64 to 2.36), 1.62 (95% CI, 1.36 to 1.92), and 1.32 (95% CI, 1.11 to 1.57) compared with a PCS score of 50 or more. Six-month decline in self-reported mental health (hazard ratio, 1.07; 95% CI, 1.02 to 1.12, per 10-point decline in MCS score) and physical function (hazard ratio, 1.25; 95% CI, 1.18 to 1.33, per 10-point decline in PCS score) were also both significantly associated with an additional increase in mortality beyond baseline risk. We also found a significant interaction between age and physical function (P = 0.02). Specifically, there was a graded response between the PCS score category and mortality in most age strata, but this relationship was not observed in the oldest age (85 years old or older).
Conclusion: In individuals newly initiated on chronic hemodialysis, self-reported baseline mental health and physical function are important, independent predictors of mortality, and there is a graded relationship between these parameters and mortality risk. Following these parameters over time provides additional information on mortality risk. One must also consider age when interpreting the relationship between physical function and mortality.