Serum Folate Concentration Corresponding to the Red Blood Cell Folate Threshold for Increased Risk of Neural Tube Defects Among Ethiopian Women of Reproductive Age

Neural tube defects (NTDs) are severe birth defects linked to insufficient maternal folate status during early pregnancy. Red blood cell (RBC) folate concentration is the recommended biomarker for predicting NTD risk, with a threshold of 748 nmol/L using the US Centers for Disease Control and Prevention (CDC) microbiological assay. However, in large surveys, serum folate concentration is often measured instead of RBC folate. This study aimed to (1) identify the serum folate concentration that corresponds to the RBC folate threshold for NTD risk using data from Ethiopian women of reproductive age (WRA); (2) evaluate the performance of the serum threshold for population-level NTD risk assessment; and (3) compare the serum folate cutoff determined with the Ethiopian data versus the cutoff derived from a Southern Indian population. Biomarker data from 1570 Ethiopian nonpregnant, non-lactating WRA who participated in the 2015 Ethiopian National Micronutrient Survey were analyzed. Serum and RBC folate concentrations were measured using the CDC microbiological assay. Serum folate cutoffs were identified using unadjusted and adjusted regression models and receiver operating characteristics (ROC) analysis. Adjusted models included the following covariates: age, body mass index (BMI), serum vitamin B₁₂ concentration, presence of inflammation, region, and site of residence. Three serum folate cutoffs were identified: 15.3 nmol/L (unadjusted model), 15.2 nmol/L (adjusted model), and 17.9 nmol/L (ROC analysis). Regression-based cutoffs balanced sensitivity (~76%) and specificity (~62%), while the ROC-derived cutoff improved sensitivity (~83%) but reduced specificity (~50%). Ethiopian cutoffs showed fair discriminatory performance (~AUC:0.7) while the Indian cutoff (~AUC:0.6) showed poorer performance. Because of suboptimal NTD risk prediction using serum folate, future surveys should use RBC folate when possible. In situations where it is not possible to measure RBC folate, population-specific serum folate cutoffs should be determined.