(P17-21-26) Time-Restricted Eating and Caloric Restriction Influences Cognition in Older Women: A Randomized Controlled Trial

Objectives: Circadian rhythms coordinate physiological processes with environmental cues such as meal timing and the light-dark cycle. Most Americans consume food over prolonged periods ( >12 h/d). Time-restricted eating (TRE) shortens the eating window and influences circadian biology that can improve cognitive performance. However, weight loss itself also improves cognition, and studies examining TRE with caloric restriction (CR) remain limited. We examined the effects of TRE+CR versus CR on cognition in postmenopausal women.

Methods: In this trial, women (BMI ≥25 kg/m2, 50-79 years old) were randomized to six months of TRE+CR (n=26) or CR (n=21). Groups were counseled to reduce calorie intake (-500 kcal/d), and TRE+CR self-selected a ≤ 9-hour eating window, while CR maintained their eating window. Baseline National Adult Reading Test (NART) was used to estimate premorbid levels of intelligence. The Pittsburg Sleep Quality Index (PSQI) questionnaire and Cambridge Neuropsychological Battery of Tests (CANTAB) were administered at weeks 0 and 24. CANTAB tests included memory and learning (Paired Associate Learning; PAL), executive function (Stockings of Cambridge; SOC; and Multitasking Test) and reaction time (RTI).

Results: The women (61±5 years, 72% white, 33±4 kg/m2) had an eating window of 12.3±0.9 h/d, which was reduced with TRE+CR (8.2±0.7 h/d), and maintained in the CR group (12.6±0.8 h/d) (p < 0.001). Mean NART error score (13.8±5.5) was within an expected range. Global PSQI score was unchanged after 6 months and did not differ between groups. There were similar reductions in body weight (-7.5±4.6 vs -6.1±5.4 kg) after 6 months. Spatial planning and problem-solving (SOC) improved over time with TRE (p < 0.05), but not CR and differed between groups (p < 0.05). Total errors during the PAL test decreased over time with TRE (p < 0.01), but not CR although the changes between groups were not significant. The multitasking test and RTI were not changed by the intervention. Reductions in the eating window predicted greater improvement in SOC (p < 0.01) and PAL (p < 0.05) after adjusting for relevant covariates.

Conclusions: TRE improved memory, learning, and problem-solving tests, independent of weight loss. Additional mechanistic analyses are warranted to explain these findings.

Funding Sources: NJ-IFNH Seed Grant; USDA-NIFA (NE1939)