OBJECTIVES: To evaluate the relationship between concomitant medications that are potentially inappropriate and the occurrence of venous thromboembolic (VTE) complications or bleeding in elderly veterans undergoing major orthopedic surgery (MOS) for one year after surgery. METHODS: The Veterans Affairs national database was used to identify all major orthopedic surgery patients over age 65 years between March 2003 and March 2004. The 2003 Beers criteria were used to select drugs that are likely to increase risks of bleed, interfere with ambulation, or increase the risk of fall. Logistic regression was used to study the relationship between the medications identified and VTE or bleed. We controlled for age, gender, obesity, cancer, and congestive heart failure in the logistic models. Due to the potential minimal impact of antihistamines, and potential protective effect of aspirin and clopidogrel, we analyzed the impact of the remaining drugs and divided patients by 0 or 1 drug versus 2 or more. RESULTS: Our final list included 43 drugs. A total of 5089 patients were elderly, and 3918 (77%) of them used at least one identified medication (mean ± standard deviation =1.6 ± 0.8). Mean age was 75.6 ±6.3. Our regression models found that patients using aspirin were less likely (P=0.008) than those not using aspirin to have VTE (odds ratio, 95% confidence interval (0.72, 0.56-0.92)), but did not affect the risk of bleed (p=0.43). After removing antihistamines, aspirin, and clopidogrel from the model, we found that patients using more than one inappropriate drug were more likely (p=0.04) to have VTE than those using one or none (1.62, 1.01-2.6). CONCLUSION: The medications identified were commonly used among elderly veterans undergoing MOS. The use of aspirin was associated with a decreased risk of VTE. Using more than one inappropriate drug was associated with an increase risk of VTE.