•We report that 65% of a cohort of older adults living in sheltered accommodation in London, UK, were vitamin D deficient (serum 25-hydroxyvitamin D concentrations <50nmol/L).•Sampling in winter/spring, non-white ethnic origin and lack of vitamin D supplement use were independently associated with lower vitamin D status.•None of a panel of 15 single nucleotide polymorphisms in DBP, DHCR7, CYP2R1, CYP27B1, CYP24A1 or VDR associated with serum 25-hydroxyvitamin D concentrations in this population.•Vigorous efforts should be made to improve uptake of vitamin D supplements among older adults in the UK in order to protect them against vitamin D deficiency. Despite the high prevalence of vitamin D deficiency among older adults in the UK, studies investigating the determinants of vitamin D status in this group are lacking. We conducted a cross-sectional study in 222 older adults living in sheltered accommodation in London, UK, who were screened for participation in a clinical trial of vitamin D supplementation for the prevention of acute respiratory infection. Details of potential demographic and lifestyle determinants of vitamin D status were collected by questionnaire and blood samples were taken for analysis of serum 25-hydroxyvitamin D (25OHD) concentration and DNA extraction. Fifteen single nucleotide polymorphisms (SNP) in 6 genes (DBP, DHCR7, CYP2R1, CYP27B1, CYP24A1, VDR) previously reported to associate with circulating 25(OH)D concentration were typed using Taqman allelic discrimination assays. Linear regression was used to identify environmental and genetic factors independently associated with serum 25(OH)D concentration. Mean serum 25(OH)D concentration was 42.7nmol/L (SD 22.0); 144/222 (64.9%) participants had serum 25(OH)D concentrations <50nmol/L. The following factors were independently associated with lower serum 25(OH)D concentration: non-white ethnicity (−8.6nmol/L, 95% CI −14.9 to −2.3, P=0.008); lack of vitamin D supplement consumption (−17.1nmol/L, 95% CI −23.3 to −10.9, P<0.001) vs. taking a daily supplement; sampling in Q1/January–March (−12.2nmol/L, 95% CI −21.5 to −2.9, P=0.01), and sampling in Q4/October–December (−10.3nmol/L, 95% CI −20.2 to −0.4, P=0.04) vs. sampling in Q3/July–September. None of the 15 SNP investigated independently associated with serum 25(OH)D concentration after correcting for multiple comparisons. In conclusion, vitamin D deficiency was highly prevalent among the older adults in this study; non-White ethnicity, lack of vitamin D supplement consumption and sampling in winter and spring independently associated with lower vitamin D status.