A comparison of sevelamer and calcium-based phosphate binders on mortality, hospitalization, and morbidity in hemodialysis: a secondary analysis of the Dialysis Clinical Outcomes Revisited (DCOR) randomized trial using claims data

Abstract

Background: The Dialysis Clinical Outcomes Revisited (DCOR) trial, a large, randomized, multicenter, open-label study, compared effects of sevelamer with calcium-based phosphate binders on mortality and hospitalization in hemodialysis patients. Many patients were lost to follow-up, precluding intent-to-treat analysis by using prospective data collection.

Study design: Preplanned secondary analysis, intent-to-treat design for all outcomes, using Centers for Medicare & Medicaid Services (CMS) data.

Setting & participants: Participants were 18 years or older and on hemodialysis therapy for more than 3 months, with Medicare as primary payor. The trial was completed at the end of 2004.

Intervention: Sevelamer, calcium-based phosphate binders.

Outcomes: Mortality, morbidity, and hospitalization end points.

Measurements: DCOR subjects were linked to the CMS End-Stage Renal Disease database. Outcomes were evaluated through the CMS End-Stage Renal Disease enrollment and claims database; baseline characteristics and comorbid conditions were evaluated using CMS and case-report data.

Results: Groups were well balanced except for a greater percentage of calcium-group patients with atherosclerotic heart disease. Analyses were adjusted by using 10 baseline characteristics. All-cause (17.7 versus 17.4 deaths/100 patient-years; P = 0.8 unadjusted; P = 0.9 adjusted) and cardiovascular mortality (9.0 versus 8.2 deaths/100 patient-years; P = 0.3 unadjusted; P = 0.4 adjusted) did not differ significantly between treatment groups. First hospitalization, cause-specific multiple hospitalizations, first morbidity, and multiple morbidity rates also did not differ significantly. Multiple all-cause hospitalization rate (1.7 versus 1.9 admissions/patient-year; P = 0.03 unadjusted; P = 0.02 adjusted) and hospital days (12.3 versus 13.9 days/patient-year; P = 0.05 unadjusted; P = 0.03 adjusted) were lower in the sevelamer group.

Limitations: Outcome parameters and cardiovascular comorbidity assessments were derived from Medicare claims data; only subjects with Medicare-as-primary-payor status were included in hospitalization and morbidity analyses.

Conclusions: In this secondary analysis, treatment with sevelamer versus calcium-based binders did not affect overall mortality (primary outcome), cause-specific mortality, morbidity, or first or cause-specific hospitalization (secondary outcomes), but there was evidence for a beneficial effect on multiple all-cause hospitalizations and hospital days (secondary outcomes).