DESIGN: Prospective cohort study.

SETTING: Community-dwelling older adults interviewed at baseline (2013-2016) and follow-up (2020-2022) as a part of the Malaysian Elders Longitudinal Research study were included.

PARTICIPANTS: Participants who attended face-to-face follow-up visits.

PRIMARY AND SECONDARY OUTCOME MEASURES: Fall occurrence over 12 months preceding the follow-up visit was determined. Anthropometric, bioimpedance analysis and physical performance measurements were obtained at both time points. Participants were categorised into three groups according to changes in weight and body composition using≥5% increase or decrease in weight to determine loss or gain.

RESULTS: Of the 225 participants, aged 71.8±6.8 years, 128 (56.9%) were women. Weight gain was associated with increased fall risk at follow-up compared with stable weight (adjusted rate ratio, aRR (95% confidence interval, CI)=2.86 (1.02-8.02)) following adjustments for age and body mass index (BMI), but this relationship was attenuated by low baseline percentage lean body mass (%LBM) in women. The association was strenghtened after adjusting for age, BMI, and low muscle strength (aRR (95% CI)=2.89 (1.01-8.28)). Weight change did not influence falls risk in men. No difference was observed with changes in percentage body fat and %LBM over time with fall occurrence for both genders.