Abstract Background Fractures can represent a seminal event in later life associated with significantly higher rates of functional impairment, nursing home admission and mortality. The aim of this study therefore was to measure the cumulative incidence of hip and wrist fractures longitudinally in a population-representative sample of community-dwelling older people and examine opportunities to intervene prior to the fracture. Methods Participants aged ≥65 years from the Irish Longitudinal Study on Ageing were included and followed from Wave 1 to Wave 5 (8-year follow-up). Falls and Fractures (Hip and Wrist) and were elicited by self-report at each wave. Participants were also asked how steady they feel when standing / walking. Logistic regression models assessed the relationship between reported falls and unsteadiness, and subsequent fracture. Results 3,072 participants were included (Mean age 73 years, 53% female) and followed for a median of 6.3 years, with over 7% (228/3,072) reporting a fracture. Over one quarter (69/267) of fractures were preceded by an injurious fall requiring medical attention in the prior 24 months, 45% were preceded by a fall of any type (121/267). One third of participants who subsequently had a fracture (89/267), reported that they were unsteady on their feet when interviewed prior to the fracture. One fifth (19/89) of these participants accessed physiotherapy services prior to the fracture. The combination of either a fall or unsteadiness reported during the preceding 24 months significantly increased the likelihood of future fracture (Odds Ratio 2.62 (95% CI 2.05 – 3.35)). Conclusion Most fractures are predictable with almost two thirds heralded by falls or unsteadiness in the preceding 24 months. Falls prevention strategies, which can reduce further falls by 20-40%, targeted at those with unsteadiness and falls would likely prevent an important proportion of future fractures, and have a positive impact on health outcomes in this cohort.