OBJECTIVES: The acetylcholinesterase inhibitors (ChEIs) donepezil, rivastigmine and galantamine are recommended for symptomatic treatment of mild to moderate Alzheimer’s Disease (AD). Despite AChEI efficacy being demonstrated in randomised clinical trials (RCTs), this study design has limited validity in relation to real-world patient care. Observational studies of routine patient care can be valuable sources of adverse event (AE) data in chronic conditions such as AD. A systematic review was undertaken to compare the safety of the AChEIs in treating AD in routine clinical care. METHODS:  Cochrane Library, MEDLINE, and EMBASE searches were conducted together with searches of selected bibliographies and conference proceedings to identify head-to-head, non-randomised AChEI studies. Two reviewers independently extracted data from relevant articles. RESULTS:   Twelve studies (N=6 prospective; N=6 retrospective) met the pre-specified inclusion criteria. Due to study design heterogeneity, a narrative data analysis was conducted. Four studies reporting total AE data found consistently fewer AEs in donepezil versus other AChEI patients, the difference being statistically significant in the largest study (N=5462; p<0.001). In three of four studies, fewer donepezil-treated patients withdrew due to AEs compared to patients receiving the other two AChEIs, with a statistically significant difference reported in the largest study (N=407; p<0.01). In four studies reporting total gastrointestinal (GI) AE data, donepezil was consistently associated with a lower incidence of GI AEs compared to rivastigmine, with three of these reporting a lower incidence for donepezil compared to galantamine. In the largest study reporting total GI AEs (N=5462), the incidence was donepezil 6%, rivastigmine 14%, and galantamine 24%.  In all studies, low numbers of CNS and cardiovascular AEs were recorded, with similar incidences of events found across the different AChEIs. CONCLUSIONS: In routine clinical settings, mild to moderate AD patients who received donepezil had fewer total and GI AEs versus patients treated with rivastigmine or galantamine.