Table 4.

Anemia-related events for HCV+ versus HCV− patients (DOPPS phases 1–5, 1996–2015)

EventNo. of EventsCrude RateaUnadjusted HRb (95% CI)Adjusted HRc (95% CI)
Hemoglobin <8.5 g/dld8615112.31.10 (1.02 to 1.19)e1.12 (1.03 to 1.21)e
Transfusionf326849.91.31 (1.15 to 1.48)e1.36 (1.20 to 1.55)e
Gastrointestinal bleedg209323.61.28 (1.10 to 1.48)e1.32 (1.13 to 1.54)e
  • HCV+, hepatitis C virus-positive; HCV−, hepatitis C virus-negative; DOPPS, Dialysis Outcomes and Practice Patterns Study; HR, hazard ratio; 95% CI, 95% confidence interval.

  • a First-events (hemoglobin <8.5 g/dl, transfusion, gastrointestinal bleed) (per 1000 patient years).

  • b Stratified by DOPPS country (United States black and nonblack patients considered as separate strata) and phase and accounting for facility clustering.

  • c Stratified by DOPPS country (United States black and nonblack patients considered as separate strata) and phase and accounting for facility clustering; adjusted for age, sex, time on dialysis, 13 summary comorbidities and hepatitis B infection, and albumin, phosphorus, and creatinine levels.

  • d n=60,319 patients; restricted to patients with baseline hemoglobin ≥8.5 g/dl.

  • e P<0.05.

  • f n=48,253 patients; excludes DOPPS phase 1 where transfusions were not included as an option on the hospitalization worksheet. In phases 2 and 3, transfusions were only ascertained on the hospitalization worksheet (as inpatient or outpatient events); in phases 4 and 5, transfusion events were additionally captured every month on the interval summary; however, dialysis units may not be aware of or record all blood transfusions that patients receive when transfusions occur in a hospital. Large United States dialysis organizations in DOPPS phases 4 and 5 did not provide transfusion events.

  • g n=63,377 patients; large United States dialysis organizations in DOPPS phases 4 and 5 were excluded because of differential reporting of hospitalization cause. Facilities reporting zero hospitalizations during DOPPS follow-up or having <5 cumulative patient-years of hospitalization follow-up were excluded.