OBJECTIVES: We aimed to compare the risk of atrial fibrillation (AF), stroke or acute myocardiac infarction (AMI) associated with the use of alendronate and raloxifene. We also focused on examining the effect of different dosage forms of alendronate (10 mg/daily dose versus 70 mg/weekly dose). METHODS: We used the National Health Insurance Research Database for a 8-year population-based retrospective cohort study. The study population was those women who first took alendronate or raloxifene during 2002-2006 and had a history of osteoporosis and vertebral or spinal fracture. The outcome measurements include AF, stroke, AMI, or until the end of the one-year follow-up period. Cox proportional hazards model was used to control the potential confounders. RESULTS: 9,609 women who had been prescribed either alendronate (6,494; 72.32%) or raloxifene (2,660; 27.68%) were included. Alendronate users were not at a higher risk of AF (HR 0.60; 95% CI, 0.42-0.85), stroke (HR 0.47; 95% CI, 0.39-0.57) or AMI (HR 0.51; 95% CI, 0.36-0.72) compared to the raloxifene group. When separating the alendronate by different dosage forms, users of alendronate 10 mg had a higher risk of AF (HR 1.66 95% CI, 1.12-2.46) and stroke (HR 1.56; 95% CL, 1.23-1.98) compared to the raloxifene while the alendronate 70 mg was at a significantly lower risk of AF (HR 0.28; 95% CI, 0.18-0.43), stroke (HR 0.23; 95% CI, 0.18-0.30) and AMI (HR 0.28, 95% CL, 0.18-0.41). The alendronate 70 mg group also demonstrated a lower risk of cardiovascular diseases to alendronate 10 mg group in AF, stroke, or AMI with HR of 0.17, 0.16 and 0.21, respectively. CONCLUSIONS: Alendronate 10 mg users had a higher risk of AF and stroke compared to raloxifene and a higher risk of all three cardiovascular outcomes compared to alendronate 70 mg. Further studies are required to investigate this relationship.