Abstract: Objective Appraise the quality of evidence-based guidelines on cognitive function training of Alzheimer's disease（AD） and provide clinical references for the AD with cognitive impairment. Methods Guidelines concerning AD were searched from GIN（Guidelines International Network）, NGC（National Guideline Clearinghouse）, NICE（National Institution for Health and Clinical Excellence）, RNAO（Registered Nurses' Association of Ontario）, SIGN（Scottish Intercollegiate Guidelines Network）, WHO（World Health Organization）, PubMed, Cochrane, CNKI, WanFang Data and VIP from creating database date to December 25^th 2017. Based on the AGREE（Appraisal of Guidelines Research and Evaluation） Ⅱ, quality of included guidelines were evaluated. Results A total of 10 guidelines were included. There was good consistency among quality evaluators of the guidelines （interclass correlation ceofficient 0.98~1.00）. The domains with higher scores include scope and purpose, participants and clarity. Domains with lower scores are preciseness, applicability and independence. A total of 20 recommendations concerning cognitive training, cognitive stimulation, cognitive rehabilitation and other cognitive interventions were extracted, and consistency was good. Conclusion The quality of the included guidelines is general. The scores in applicability, preciseness and independence are low, which is not conducive to the implementation of the guidelines in clinical practice. It is suggested that future guidelines should pay more attention. Multiple guidelines recommend cognitive interventions for AD. Domestic medical staff can combine clinical practice, apply evidence, and popularize cognitive interventions.