Background <p>The effects of 15:9 time-restricted eating (TRE) on cognitive function and anthropometric parameters among the elderly with mild cognitive impairment (MCI) requires further investigation.</p> Objective <p>This assessor-blinded, pilot randomized controlled trial aimed to examine the effects of 12 weeks of 15:9 TRE on cognitive function and anthropometric parameters in older adults with MCI.</p> Methods <p>Forty-six eligible participants (mean age 74.20 ± 5.09 years) were recruited from a community elderly service center in Chengdu, Sichuan, China, and were randomly allocated to either the TRE group (<i>n</i> = 23) or the control group (<i>n</i> = 23). The TRE group followed 15:9 TRE for 12 weeks and the control group maintained eating ad libitum without any modifications. The primary outcome were changes in the Montreal Cognitive Assessment (MoCA) and the auditory verbal learning test-Huashan version (AVLT-H) and their subitems. The secondary outcome were changes in body weight, body mass index (BMI), waist circumference, calf circumference, and adherence. All outcomes, except for adherence, were measured consistently at baseline, 6 weeks, and 12 weeks.</p> Results <p>There was a significant improvement in the MoCA total score (from baseline to 6-week: β = 1.22, 95% CI: 0.33 to 2.10, <i>P</i> = 0.007; from baseline to 12-week: β = 1.70, 95% CI: 0.46 to 2.94, <i>P</i> = 0.007) and the attention score of MoCA (from baseline to 6-week: β = 0.65, 95% CI: 0.15 to 1.15, <i>P</i> = 0.011) in the TRE group compared with the control group. The AVLT-H recognition score (AVLT-H-REC-S) was significantly improved in the TRE group compared with the control group (from baseline to 12-week: β = 2.44, 95% CI: 0.63 to 4.24, <i>P</i> = 0.008). There was no significant difference in body weight, BMI, waist circumference, or calf circumference between the two groups. The mean adherence to TRE was more than 5 days per week in the TRE group.</p> Conclusions <p>Short-term improvements in MoCA total score and AVLT-H-REC-S were observed among elderly individuals with MCI following a 12-week 15:9 TRE. However, no significant effect of TRE on anthropometric parameters was found. These findings should be interpreted with caution due to the limited sample size. Larger-sample studies are needed to further clarify the underlying mechanisms by which TRE affects cognitive function in the elderly MCI population.</p> Trial registration <p>This trial was registered at the China Clinical Trial Registration Center on 5 November 2021, with registration number ChiCTR2100052766.</p> Graphical Abstract <p></p>