Feng Xu, Yiwan Zhang, Danqing Tan, Yiming He, Qinfei Fei, Kai Xu
Objective: This study aimed to systematically evaluate the effects of exercise intervention on cognitive function in AD patients through meta-analysis, specifically investigating the dose-response relationships and moderating effects of various exercise prescription parameters including frequency, single-session duration, weekly total duration, total intervention duration, type, and intensity-on the magnitude of cognitive improvement. The findings are expected to provide scientific and actionable empirical evidence for designing precise and individualized exercise programs. Methods: Relevant literatures were systematically retrieved from eight databases up to August 2025, including PubMed, Web of Science, Embase, Emcare, Scopus, Cochrane Library, Ebsco, and SPORTDiscus. All included trials were randomized controlled trials (RCTs) involving adults aged 60 years and above. These studies examined the effects of exercise intervention on cognitive function in AD patients, compared with passive control groups without exercise. A multilevel meta-analysis was used to assess the impact of exercise on cognitive function outcomes in AD patients. Additionally, multivariate meta-regression analysis was employed to identify the exercise frequency (sessions per week) that maximizes cognitive improvement, as well as key moderating factors. Results: > 0.05). However, the cognitive benefits of exercise were significantly enhanced with increased weekly exercise frequency, with a notable strengthening trend observed when sessions exceeded five per week. Conclusion: Exercise intervention can effectively improve cognitive function in AD patients. "To optimize intervention outcomes, exercise prescription should consider higher frequency (e.g., more than five sessions per week may be associated with greater cognitive benefits) and the cumulative effect during the initial intervention phase (e.g., sustained beyond 12 weeks may represent a critical window for short-term improvement). In addition, a potential plateau in long-term benefits suggests the need for periodic program adjustments and multimodal exercise strategies. These findings should be interpreted as exploratory dose-response associations rather than definitive clinical prescriptions. Systematic review registration: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=1012816, identifier CRD420251012816.