highly significantP<2x10 -16
Specifically, adjusting for African ancestry in a multiple regression model largely eliminated the highly significant effect of Black ethnicity on serum creatinine levels (decreased from P<2x10 -16 to a marginally significant P=0.05) whilst the highly significant effect of African ancestry on creatinine remained essentially unchanged when adjusted for Black ethnicity (P<2x10 -16 before and after adjustment).