Objective: To determine whether changes in fat and lean mass over time, quantified using dual-energy x-ray absorptiometry (DXA), are related to incident cardiovascular events. Previous studies using surrogate anthropometric methods have had inconsistent findings. Study design: Prospective, longitudinal observational study of women aged 40 to 80 randomly selected from the electoral roll and stratified into decades: 40–49, 50–59, 60–69 and 70–79 years. Main outcome measures: Changes in anthropometric measurements (body mass index and waist-to-hip ratio) and DXA-quantified fat mass and lean mass between the first and fifth years of the study. Incident cardiovascular events recorded from the sixth to the 12th year. Results: In total 449 participants (87.9 %) were analyzed. A 10 % or greater decrease in total fat mass index was associated with a 67 % lower likelihood of any cardiovascular event (OR = 0.33, 95%CI 0.15–0.71); no association was observed for an increase. A 10 % or greater decrease in abdominal fat mass index was associated with a 62 % lower likelihood of incident stroke (OR = 0.38, 95%CI 0.16–0.91); no association was observed for an increase. A 10 % or greater decrease in appendicular lean mass index resulted in increased odds ratio of 2.91 for incident peripheral artery events (OR = 2.91, 95%CI 1.18–7.20). Conclusions: Reducing fat mass for women in midlife and beyond may decrease the risk of cardiovascular events. An increase in fat mass may not contribute to additional cardiovascular events. A reduction in limb muscle mass may provide an independent marker for cardiometabolic risk and peripheral artery disease. No independent association was found using anthropometric measurements and incident cardiovascular events.