<strong>ABackground </strong> An elevated homocysteine level has been indicated as a risk factor for cardiovascular disease, cognitive decline, and fractures. Supplementation of vitamin B12 and folic acid in order to normalize homocysteine levels might be of substantial public health importance as this might reduce the risk for several age-related conditions. This thesis focuses on two health outcomes frequently associated with elevated homocysteine levels and low levels of vitamin B12 and folate: osteoporosis and cognitive decline later in life. <strong>Methods </strong> Findings are presented in the context of a model which links dietary intake to biomarkers of nutritional status and subsequently to health outcomes. Two systematic reviews with meta-analyses investigated the current status of knowledge about the association of vitamin B12 intake and status with cognitive function, and the association of homocysteine, vitamin B12 and folate status with bone health. Baseline data of the B-PROOF study were used to assess 1) the association of vitamin B12 intake with status according to four biomarkers (vitamin B12, holotranscobalamin (holoTC), methylmalonic acid (MMA) and homocysteine), 2) the mutual association among these four vitamin B12 biomarkers and 3) the association between homocysteine, vitamin B12 biomarkers, folate and cognitive function. The effect of 2-year daily vitamin B12 (500 μg) and folic acid (400 μg) supplementation on fracture risk was assessed in the B-PROOF study, a large (N=2919) randomized controlled trial in elderly people (aged ≥65 years) with an elevated homocysteine level (≥12.0 µmol/L). <strong>Results </strong> The systematic review of the literature showed no or inconsistent associations of vitamin B12 intake with cognitive function. Furthermore, serum vitamin B12 was not associated with risk of dementia, global cognition or memory. Studies on MMA and holoTC reported significant associations with risk of dementia, Alzheimer’s disease and global cognition. A meta-analysis showed that serum/plasma vitamin B12 per 50 pmol/L was borderline significantly associated with a lower fracture risk (RR=0.96, 95% CI = 0.92-1.00) and that homocysteine was significantly associated with a higher fracture risk (RR=1.04, 95% CI = 1.02-1.07). Meta-analyses regarding vitamin B12, folate and homocysteine levels and BMD did not show significant associations. In the B-PROOF study a doubling of vitamin B12 intake was associated with 9% higher levels of vitamin B12, 15% higher holoTC, 9% lower MMA and 2% lower homocysteine, saturation of biomarkers occurs with dietary intakes of >5 μg B12. Levels of MMA and homocysteine were higher when vitamin B12 levels were below 330 pmol/L and when holoTC levels were below 100 pmol/L, with a steep elevation when levels of vitamin B12 and HoloTC were below 220 and 50 pmol/L respectively. At baseline, levels of vitamin B12 and holoTC were not associated with cognitive function in any cognitive domain. Levels of homocysteine (β= -0.009), folate (β= 0.002), MMA (β= -0.163) and the wellness score – a vitamin B12 biomarker combination score - (β= 0.048) were significantly associated with the domain of episodic memory. Additionally, homocysteine (β= -0.015) and the wellness score (β= 0.103) were significantly associated with the domain information processing speed. The B-PROOF intervention did not lower the risk of fracture in the total population (HR=0.84, 95% CI = 0.58-1.22). Per protocol subgroup analysis of elderly aged >80 years, showed a lower risk of fracture in the intervention group (HR=0.28, 95% CI 0.10-0.74). We observed more cancer cases in the intervention group (HR=1.55, 95% CI = 1.04-2.30) compared to the placebo group. We cannot rule out the possibility of accelerated cancer progression as a possible negative side effect. <strong>Conclusion </strong> Our literature reviews and observational data confirm an association of levels of homocysteine, vitamin B12 and folate with cognitive function and fracture risk in elderly. Supplementation with vitamin B12 and folic acid did not lower the risk of fracture in the total study population. Though positive effects on fracture incidence emerged in elderly aged >80 years, these benefits should be weighed against potential risks.