The objective of the study was to determine the association between AAC and neuromuscular function over 5 years. Participants in this study were ambulant women over 70 years old residing in Perth, Western Australia who participated in the Calcium Intake Fracture Outcomes Study, a randomised controlled trial of calcium supplementation. 1046 women (mean age = 74.9 ± 2.6 years; BMI = 27.1 ± 4.4 kg/m2) were included. Lateral spine images captured during bone density testing were scored for AAC (AAC24; 0–24) at baseline. Severe AAC (AACsev) was defined using established cut points (AAC24 ≥ 6). At baseline and follow-up, isometric grip strength was assessed using a dynamometer. Mobility was assessed by the Timed-Up-and-Go (TUG) test. Using pre-defined criteria, muscle weakness was considered as grip strength < 22 kg and poor mobility defined as TUG > 10.2 s. A subset of women had appendicular lean mass (ALM) determined by dual-energy X-ray absorptiometry at baseline and follow-up (n = 261). AACsev was evident in 193 (18.5%) women. Average decline in grip strength after 5 years was greater in those with AACsev than those without (3.6 ± 3.7 vs. 2.9 ± 4.2 kg; p = 0.034). This remained significant after adjustment for age, treatment allocation, diabetes, smoking history, renal function, medical record-derived prevalent vascular disease, BMI and physical activity (β = − 0.184; 95% confidence interval: − 0.361, − 0.008; p = 0.040). AACsev was not associated with 5-year changes in TUG or ALM in univariable or multivariable analyses (all p > 0.05). In older women, severe aortic calcification was associated with greater 5-year decline in muscle strength, but not TUG or ALM. These findings support the concept that vascular disease may have an effect on the loss of muscular strength.