Measurement of 25-hydroxyvitamin D (25-OHD) is widely used for assessing vitamin D status. There has been a dramatic increase in 25-OHD requests over recent years prompting many laboratories to consider the use of automated immunoassays. To achieve higher throughput, these methods have abandoned the traditional solvent extraction of samples and are therefore more prone to non-specific interference. The Vitamin D External Quality Assessment Scheme (DEQAS) has revealed method-related differences in 25-OHD results, raising concerns about the comparability and accuracy of different assays. This paper highlights some of the pre-analytical, analytical and post-analytical issues which may influence the accuracy of 25-OHD assays and interpretation of results. Recent attention has focused on reconciling the relatively high results given by liquid chromatography-tandem mass spectrometry (LC-MS/MS) to those of the DiaSorin radioimmunoassay (RIA) on which clinical decision points have previously been based. Data is presented on 20 DEQAS samples which were analysed by an LC-MS/MS assay developed as a candidate reference measurement procedure by the US National Institute of Standards and Technology (NIST). The NIST results were on average 11.2% lower than those given by routine LC-MS/MS methods. If confirmed, these results suggest that most routine LC-MS/MS assays are perhaps overestimating 25-OHD by failing to resolve a molecule having the same mass as 25-OHD(3) and a similar fragmentation pattern. All 25-OHD assays should be monitored by a proficiency testing scheme and the results made available to clinicians and editors of scientific journals.