Analyses. Continuous variables were summarized as means
with standard deviations and categorical variables were summarized
as percentages. General linear models were used to
model the effect of iron and vitamin D intake on total body
iron stores and circulating vitamin D levels after controlling
for age, sex, race, and transfusion status. When the influence
of race was explored, racial categories were collapsed to compare
white vs non-white races (ie, Asian, black, mixed, other).
Subgroup analyses were also performed in the nontransfused
thalassemia group alone. All inferences are based on twotailed
tests with a threshold of .05 for declaring significance.
All analyses were conducted using Statistical Analysis
Software (version 9.1.3, 2006, SAS Institute).