The majority of organic carbon (OC) burial in marine sediments occurs on continental shelves, of which an estimated 10–20% is associated with reactive iron (FeR). The association of OC with FeR (OC-FeR) is thought to facilitate preservation of organic matter (OM) in sediments and therefore represents an important carbon sink. The citrate-bicarbonate-dithionite (CBD) method is used to quantify OC-FeR in marine sediments by reductively dissolving FeR, thereby releasing bound OC. While the CBD method is widely used, it may be less efficient at measuring OC-FeR than currently thought, due to the incomplete reduction of FeR, resulting from the neutral pH conditions required to prevent OM hydrolysis. Additionally, the typical range of values reported for OC-FeR in marine sediments is narrow, despite variation in OM and FeR inputs, OM source types and chemical compositions. This suggests a limitation exists on the amount of OC that can become associated with FeR, and/or that the CBD method is limited in the OC-FeR that it is able to quantify. In assessing the efficiency of the CBD method, we aimed to understand whether methodological errors or (mis)interpretation of these extraction results may contribute to the apparent limitation on OC-FeR values. Here, we synthesised OC-FeR composites with a known FeR phase and known OM moieties, varying in carboxyl content, at neutral pH. These were spiked into OC-free marine sediment, and subject to a CBD extraction to investigate i) the efficiency of CBD for OC extraction; ii) the efficiency of CBD for FeR extraction; ii) how the OC moiety affects the physical parameters of associated FeR minerals; and iii) the impact of OM moiety on OC and Fe release. We show that the CBD method results in only partial dissolution of the most susceptible FeR phase (ferrihydrite) and therefore incomplete removal of bound OC. While as little as ~20% of Fe is released from OC-free ferrihydrite, structural disorder of the mineral phase increases with the inclusion of more OC, resulting in greater losses of up to 62% Fe for carboxyl rich OC-FeR complexes. In addition, our results show that the NaCl control step performed in the CBD method is capable of removing weakly bound OC from FeR, such that inclusion of this OC in the total OC-FeR fraction may increase marine sediments OC-FeR estimates by ~33%. Finally, we suggest that the structure of OC involved in OC-FeR binding can affect quantification of the OC-FeR pool. Our results have important implications for assessing the FeR bound OC fraction in marine sediments and the fate of this OC in the global carbon cycle.