The objective of this study was to describe the prevalence of vitamin D deficiency among antiretroviral treatment-naive, HIV-positive individuals. We reviewed records of consecutive antiretroviral treatment-naive patients, registering for care for the first time at a London clinic from 01 January 2008 to 31 December 2009. During this period, serum 25-hydroxycholecalciferol was measured routinely for all new patients. 25-hydroxycholecalciferol deficiency and severe deficiency were defined as <= 50 and <= 25 nmol/L, respectively. Among 253 patients (82% men, median age 36 years, 64% white ethnicity), 148 (58.5%) were 25-hydroxycholecalciferol-deficient, including 32 (12.6%) who were severely deficient. In all, 73.5% (61/83) patients of non-white ethnicity were 25-hydroxycholecalciferol-deficient compared with 50.7% (76/150) of those reporting white ethnicity (p< 0.001). Seven of eight (87.5%) patients with hypocalcaemia (< 2.12 nmol/L) were 25-hydroxycholecalciferol-deficient. The prevalence of 25-hydroxycholecalciferol-deficiency was higher in winter and spring vs. summer and autumn (89/129 [69.0%] vs. 59/124 [47.6%], p< 0.001). Serum 25-hydroxycholecalciferol deficiency was not associated with gender, CD4 count, HIV viral load or clinical stage. Serum 25hydroxycholecalciferol deficiency was common among antiretroviral treatment-nai I ve patients, with those of non-white ethnicity at highest risk. CD4 count, HIV viral load and HIV clinical staging do not help to identify those at risk, but low serum calcium should prompt investigation of 25-hydroxycholecalciferol levels.