Background: The evidence regarding the association between added sugar (AS) intake and obesity remains inconsistent. The aim of this study was to investigate the association between changes in the percentage of energy intake from AS (EAS%) and changes in body weight in a cohort study of older Australians during 15 years of follow-up. In addition, associations were assessed according to whether EAS% intake was provided from beverage or non-beverage sources. Methods: Data were analyzed from the participants of the Blue Mountains Eye Study Cohort. Dietary data were collected at baseline (1992–94) and three five-yearly intervals using a 145-item food frequency questionnaire. Participants' body weight was measured at each time point. Five-yearly changes in EAS% intake and body weight were calculated (n = 1,713 at baseline). A generalized estimating equation (GEE) model was used to examine the relationship between the overall five-yearly changes in EAS% intake and body weight, adjusted for dietary and lifestyle variables. Results: In each time interval, the EAS% intake decreased by ~5% in the lowest quartile (Q1) and increased by ~5% in the highest quartile (Q4). The mean (SD) body weight change in Q1 and Q4 were 1.24 (8.10) kg and 1.57 (7.50) kg (first time interval), 0.08 (6.86) kg and −0.19 (5.63) kg (second time interval), and −1.22 (5.16) kg and −0.37 (5.47) kg (third time interval), respectively. In GEE analyses, the overall five-yearly change in EAS% intake was not significantly associated with body weight change (Ptrend = 0.837). Furthermore, no significant associations were observed between changes in EAS% intake from either beverage or non-beverage sources and changes in body weight (Ptrend for beverage sources = 0.621 and Ptrend for non-beverage sources = 0.626). Conclusion: The findings of this older Australian cohort do not support the association between changes in EAS% intake and body weight, regardless of AS food sources (beverage or non-beverage).