Skip to main content
. 2009 Sep 1;114(21):4639–4644. doi: 10.1182/blood-2009-04-218040

Table 3.

Independent associations with erythropoietin (natural log) in multivariate analysis

Beta (95% CI) Standardized beta P
All patients, *N = 179
    Hemoglobin, g/L −3.1 (−3.6-−0.26) −0.69 <.001
    Hemoglobin F, % 0.02 (0.01-0.03) 0.27 <.001
Patients on hydroxyurea, n = 72
    Hemoglobin, g/L −2.7 (−3.8-−1.7) −0.58 <.001
    Hemoglobin F, % 0. 03 (0.01-0.05) 0.32 .005
Patients not on hydroxyurea, n = 107§
    Hemoglobin, g/L −3.3 (−3.9-−2.8) −0.80 <.001
    Hemoglobin F, % 0.01 (0-0.02) 0.12 .09
*

Includes patients for whom hemoglobin F and erythropoietin results were available. Variables entered into models were hemoglobin, hemolytic component, hemoglobin oxygen saturation, and hemoglobin F (%).

R-square = 0.51. If hemoglobin F category of 8% or higher versus lower than 8% is used in the model, P value for hemoglobin F = .001.

R-square = 0.34. If hemoglobin F category of 8% or higher versus lower than 8% is used in the model, P value for hemoglobin F = .015.

§

R-square = 0.61. If hemoglobin F category of 8% or higher versus lower than 8% is used in the model, P value for hemoglobin F = .3.