We prospectively investigated relationships between blood markers of Fe, vitamin B12, folate, vitamin C and vitamin D status and subsequent all-cause mortality in 208 men and 191 women aged 75 years or over living in the community in Aberdeen, Scotland. The participants had been recruited for a cross-sectional study in 1999–2000 when they completed health and lifestyle questionnaires and had blood samples taken for analysis of serum ferritin, serum vitamin B12, erythrocyte folate, plasma vitamin C and serum 25-hydroxycholecalciferol. Mortality was ascertained on national databases up to December 2005, with a median time of follow up of 69·2 (range 1·0–79·9) months. Participants were divided into sex-specific quintiles of baseline levels for each nutrient, and hazard ratios were estimated with Cox proportional hazard models adjusted for age and sex with the significance of linear trends in the associations assessed by logistic regression. There was no significant association between blood markers of Fe, vitamin B12 or folate status at baseline and mortality, but vitamin D status at baseline was inversely related to mortality (P for trend < 0·001). For vitamin C there was no evidence of a linear trend but participants in the lowest quintile of plasma levels had a significantly higher risk of death than those in the highest quintile. Randomized controlled trials of lifestyle changes which improve vitamin status are needed to assess whether these associations could be causal.