Background: Ferritin and soluble transferrin receptor (sTfR) concentrations are commonly used to assess iron deficiency (ID); however, they are influenced by multiple factors.

Objectives: We assessed associations between numerous variables and both ferritin and sTfR concentrations in Cambodian women and compared ID prevalence through the use of study-generated correction factors (CFs) for ferritin with those from a published meta-analysis.

Methods: Venous blood from 450 women (aged 18–45 y) was assessed for hemoglobin (Hb), ferritin, sTfR, retinol binding protein, folate, vitamin B-12, C-reactive protein, α-1 acid glycoprotein (AGP), and genetic Hb disorders. Linear regression was used to calculate geometric mean ratios (95% CIs) for ferritin and sTfR concentrations.

Results: The variant Hb EE genotype was associated with 50% (14%, 96%) and 51% (37%, 66%) higher geometric mean ferritin and sTfR concentrations, respectively, than was the normal Hb AA genotype; a 1-g/L increase in AGP was associated with 99% (50%, 162%) and 48% (33%, 64%) higher concentrations in the same variables, respectively. ID prevalence in nonpregnant women (n = 420) was 2% (n = 9) with the use of ferritin <15 μg/L and 18% (n = 79) with the use of sTfR >8.3 mg/L as criteria. ID prevalence with the use of sTfR was higher in women with the Hb EE genotype (n = 17; 55%) than in those with the Hb AA genotype (n = 20; 10%); and in women with the Hb AA genotype and chronic inflammation (n = 10; 18%) than in that group of women without chronic inflammation (n = 10; 7%) (P < 0.05). No differences in ID prevalence were found with the use of ferritin between women with Hb EE and AA genotypes (P = 1.0) or by chronic inflammation status (P = 0.32). There were no differences in mean ferritin concentrations among all 450 women when study-generated CFs were compared with those from the meta-analysis (P = 0.87).

Conclusions: Compared with sTfR, ferritin concentrations appear to reflect more accurately true ID in rural Cambodian women. The CFs from a published meta-analysis were appropriate for use in this population with a high prevalence of Hb disorders and inflammation.