Abstract Background There is no evidence on the specific beneficial association of the main types of olive oil consumption with frailty. Objective The aim was to assess the relationship between olive oil consumption and incident frailty in community-dwelling older adults. Design Prospective cohort. Setting Participants were recruited in 2008–10 and follow-up through 2013. Subjects In total, 1,896 older adults aged 60+. Methods At baseline, olive oil and other food consumption was collected using a validated dietary history. Incident frailty was defined as having at least three of the following five Fried-based criteria: low physical activity, fatigue, slow walking, muscle weakness and unintentional weight loss. Analyses were performed with logistic regression and adjusted for the major confounders. Results Over a mean follow-up of 3.5 years, 135 incident frailty cases were identified. The odds ratio (95% confidence interval) of frailty across sex-specific tertiles of total olive oil consumption (12.7, 20 and 30.8 g/day, respectively) were: 1 (ref.), 0.52 (0.32, 0.83) and 0.47 (0.29, 0.78), P trend 0.003. When differentiating by olive oil types, the results held for virgin but did not for common (refined) olive oil. Conclusion The highest total olive oil consumption (~3 tablespoons), especially if virgin, was associated with half the risk of frailty as the lowest consumption (~1 tablespoon) among older adults. This study suggests that virgin olive oil should be the preferent culinary olive oil type for frailty prevention. If confirmed in other settings, small doses of virgin olive oil could be added as a simple geriatric nutritional advice on the prevention of frailty.