The study from the Chinese CHARLS cohort included 3,733 participants and examined the association of long-term cystatin C trajectories with cognitive decline in diabetics and non-diabetics. During the 3-year follow-up, there was a decline in cognitive function in 20.4% of participants. In non-diabetics, cumulative cystatin C increased the risk of cognitive decline, and maintaining a low cystatin C level was associated with a reduced risk compared with an intermediate level. In diabetics aged ≥ 65 years, maintaining a low cystatin C level was associated with an increased risk of cognitive decline compared to the median level. In non-diabetics aged ≥ 65 years, cumulative cystatin C increased the risk of falls. In non-diabetic and non-hypertensive individuals, maintaining a low cystatin C level was associated with a reduced risk of cognitive decline compared to the median level. The findings suggest that long-term changes in cystatin C may serve as an indicator of cognitive decline risk and support differentiated preventive strategies.