A decision-analytic model incorporating a Markov process to assess the incremental cost and effectiveness of venous thromboembolism (VTE) prevention strategies was used. Modeling was carried out using a hypothetical cohort of medical patients at risk of VTE. The model compared clinical effectiveness (primary and recurrent VTE, death), safety (adverse events), and direct medical costs between patients receiving enoxaparin prophylaxis, unfractionated heparin (UFH) prophylaxis, and no prophylaxis (n = 10,000 for each arm). Monte Carlo simulation was performed to identify changes in inputs that would affect the results. The estimated incidence ofVTE at two years (including recurrent VTE) was 6.8% with enoxaparin prophylaxis, 7.9% with UFH prophylaxis, and 17.9% with no prophylaxis. Two-year mortality occurred in 15.7% of enoxaparin patients and 16.0% of UFH patients, with the incidences of major bleeding in these groups being 0.7% and 1.2%, respectively. However, both enoxaparin and UFH prophylaxis were associated with higher rates of major bleeds than no prophylaxis (0.6%). Total average costs per patient were (US dollars) $1,264 (for enoxaparin prophylaxis, $1,585 for UFH prophylaxis, and $2,245 for no prophylaxis). No realistic parameter changes resulted in enoxaparin prophylaxis being more costly than UFH prophylaxis. For the healthcare payer, considering all direct medical costs associated with VTE up to two years after an admission for acute illness, prophylaxis with enoxaparin was more effective and less costly than UFH. This identifies enoxaparin as a potentially favorable VTE prophylaxis regimen compared with UFH and no prophylaxis in at-risk medical patients.