Exercise- and dietary pattern-based lifestyle interventions and neurocognitive function in adults: a Bayesian multilevel network and dose-response meta-analysis.

Mingyue Jiao, Qingmei Li, Hanping Liang, Binbin Zhang, Mengshi Li, Jian Li, Mohd Taib Harun, Fangfang Wen

Journal: Journal of neurology 2026;273(7):

PMID: 42334484

Abstract

OBJECTIVE

To systematically review the effects of lifestyle interventions focused on exercise and dietary patterns on neurocognitive function in adults, and to examine their relative effects and dose-response relationships.

METHODS

PubMed, Embase, Web of Science, and the Cochrane Library were searched from database inception to March 20, 2026. A total of 23 randomized controlled trials (RCTs) involving 12,286 participants were included. Bayesian multilevel pairwise, network, and dose-response meta-analyses were conducted to evaluate overall cognition, memory, and executive function. RoB 2 and CINeMA were used to assess the credibility of the evidence.

RESULTS

Combined interventions yielded small, but statistically significant cognitive benefits (overall cognition: SMD = 0.18; memory: SMD = 0.10; executive function: SMD = 0.09). These benefits were driven primarily by differences between the combined intervention and control or usual care, as well as exercise alone; compared with diet alone, the combined intervention still showed a small advantage for executive function. Network meta-analysis suggested that AE_MIND and AE_RT_MED were more beneficial for overall cognition, AE_RT_MIND and AE_RT_MED for memory, and AE_RT_CR and AE_MIND for executive function. Exploratory dose-response analyses suggested possible curvature in some spline-based models; however, comparison with simpler monotonic specifications did not provide robust evidence for a clinically interpretable non-linear or monotonic dose-response relationship.

CONCLUSION

Combined lifestyle interventions may produce statistically significant but modest improvements in cognitive function in adults, with benefits varying across cognitive domains. However, current evidence remains insufficient to support a standardized precision prescription, and high-quality RCTs with long-term follow-up are needed to determine the optimal combinations and dosages.

Address: School of Teacher Education, West Campus, Hezhou University, 3261 Xiaohe Avenue, Babu District, Hezhou, 542899, Guangxi, China.; School of Tourism and Sports Health, Hezhou University, Hezhou, 542899, Guangxi, China.; Faculty of Sports and Leisure, Guangdong Ocean University, Zhanjiang, 524088, Guangdong, China.; Faculty of Education and Liberal Arts, INTI International University, 71800, Nilai Negeri , Sembilan, Malaysia.; Faculty of Education and Liberal Arts, INTI International University, 71800, Nilai Negeri , Sembilan, Malaysia.; Faculty of Education and Liberal Arts, INTI International University, 71800, Nilai Negeri , Sembilan, Malaysia.; Department of Physical Education, Shanxi Institute of Science and Technology, Jincheng, 048000, Shanxi, China.; School of Teacher Education, West Campus, Hezhou University, 3261 Xiaohe Avenue, Babu District, Hezhou, 542899, Guangxi, China. [email protected].

DOI: 10.1007/s00415-026-13918-7

MeSH Terms: Humans, Bayes Theorem, Cognition, Exercise, Executive Function, Cognitive Enhancement, Life Style, Diet, Adult, Randomized Controlled Trials as Topic