Background Maternal hemoglobin (Hgb) is considered an essential, modifiable risk factor for adverse pregnancy outcomes (APOs). Evidence for the effect of nutrition education on the Hgb levels of pregnant women in low-income countries, including Ethiopia, is inconclusive. This study aimed to assess the effect of nutrition education on the Hgb levels of pregnant women in urban settings in the Bale Zone, Southeast Ethiopia. Methods A community-based two-arm parallel cluster randomized controlled trial was carried out among 447 randomly selected pregnant women attending antenatal care (224 intervention and 223 control groups) at health facilities from February to December 2021. A multistage cluster sampling technique followed by systematic sampling was used to select the pregnant women. Pregnant women who took part in the intervention arm received six nutrition education sessions, whereas pregnant women in the control group received routine standard care. We used a pretested, interviewer-administered, structured questionnaire to collect the data. The Hgb level of pregnant women was measured by collecting a finger-prick blood sample using a HemoCue Hb 301. A generalized estimating equation (GEE) model was used to isolate the net effect of the intervention on Hgb, accounting for the clustering. Beta coefficients (beta) along with a 95% confidence interval (CI) were used for interpretations. Results The mean difference in Hgb levels between the intervention and control groups was 0.12 +/- 0.04 (P value < 0.002). The multivariable GEE linear model revealed that nutrition education significantly improved the Hgb levels of pregnant women [beta = 0.36, 95% CI: (0.30, 0.43)]. An increase in the consumption of a cup of coffee or tea decreased Hgb levels by 0.14 g/dL [beta = -0.14, 95% CI: (-0.23, -0.06)]. Conclusion The findings showed that a comprehensive nutrition education intervention using the health belief model (HBM) and theory of planned behaviour (TPB) designed to improve dietary diversity substantially improved hemoglobin (Hgb) levels among pregnant women. While we found no single dietary factor to be significant, in this group of pregnant women in Ethiopia, an increase in the daily consumption of a cup of coffee or tea decreased Hgb levels. As a consequence, pregnant women should be advised to limit their coffee or tea consumption. The study was registered on Clinicaltrials.gov retrospectively with the registration number PACTR202201731802989 on 24/01/2022.
