Background: hypovitaminosis D has a prevalence of 50,4% in older persons. This population has grown globally in an impressive way during the last 20 years. In addition, the deficiency of vitamin D (< 30 ng/ml) could raise the risk of mortality in this population. Objectives: to know whether vitamin D levels less than 30 ng/ml have an impact on mortality for overall causes, cardiovascular causes, cancer or injuries (falls) in a follow-up period of nine years in elderly subjects. Methods: a retrospective longitudinal study with nine years period of observation; 418 elderly subjects were followed, and they were recruited in 2004 to participate in a study about vitamin D supplementation and muscle strength. Results: vitamin D levels below 30 ng/dl increase by almost three times the probability of dying (OR = 2.77, IC = 1.81-3.85), while the same level of vitamin D is associated with a double risk of dying from cardiovascular causes (OR = 1.78, CI = 1.21-2.09) in the unadjusted model. For cancer mortality and falls, no significant results were obtained. Vitamin D would not act as a protective factor against mortality. Conclusions: vitamin D levels less than 30 ng/ml could be a risk factor for general mortality.