OBJECTIVES: The incidence of tuberculosis (TB) is high amongst foreign-born persons resident in developed countries. Vitamin D is important in the host defence against TB in vitro and deficiency may be an acquired risk factor for this disease. We aimed to determine the incidence and associations of vitamin D deficiency in TB patients diagnosed at an infectious diseases unit in London, UK. METHODS: Case-note analysis of 210 unselected patients diagnosed with TB who had plasma vitamin D (25(OH)D3) levels routinely measured. Prevalence of 25(OH)D3 deficiency and its relationship to ethnic origin, religion, site of TB, sex, age, duration in the UK, month of 25(OH)D3 estimation and TB diagnosis were determined. RESULTS: Of 210 patients 76% were 25(OH)D3 deficient and 56% had undetectable levels. 70/82 Indian, 24/28 East African Asian, 29/34 Somali, 14/19 Pakistani and Afghani, 16/22 Sri Lankan and 2/6 other African patients were deficient (with 58, 17, 23, 9, 6 and 1 having undetectable levels, respectively). Only 0/6 white Europeans and 1/8 Chinese/South East Asians had low plasma 25(OH)D3 levels. Muslims, Hindus and Sikhs all had equivalent rates of deficiency though Hindus were more likely to have undetectable levels (odds ratio 1.87, 95% CI 1.27-2.76). There was no significant association between 25(OH)D3 level and site of TB or duration of residence in the UK. There was no apparent seasonal variation in either TB diagnosis or 25(OH)D3 level. CONCLUSIONS: 25(OH)D3 deficiency commonly associates with TB among all ethnic groups apart from white Europeans, and Chinese/South East Asians. Our data support a lack of sunlight exposure and potentially a vegetarian diet as contributors to this deficiency.