The burden of venous thromboembolism (VTE) remains high in the United States (US). This study assesses the rate of VTE prophylaxis in a large real-world population of medically ill patients and identifies factors which confer VTE risk to this population. Discharges from the PharMetrics database were included if they were aged > or =40 years and had a hospitalisation claim (Jan 2001-Dec 2005) for cancer, congestive heart failure (CHF), severe infectious disease (SID), or lung disease. Discharges with incomplete records in the prior year to the index hospitalisation claim date were excluded. VTE rate, type (deep venous thrombosis [DVT] or pulmonary embolism [PE]), and time to VTE were compared between groups. Multivariate logistic regression analysis was used to identify independent predictors of VTE occurrence. A total of 158,325 patients were included in the study. Cancer patients had the highest incidence of VTE (7.6%), with the average for all patients being 5.6% (1.5% PE). VTE occurred most frequently post discharge, with the median time being 74 days. Post-discharge prophylaxis was provided to 13.1% of CHF patients and < 5% of all other patients. Independent predictors of VTE included a pre-index VTE (odds ratio [OR] 9.06, 95% confidence interval [CI] 8.28-9.91) and a primary diagnosis of cancer compared with a diagnosis of SID (OR 1.34, 95% CI 1.24-1.46). In conclusion, commercially insured medical patients in the US are at high risk of VTE following hospital discharge. One-quarter of medical patients who developed a VTE are at high risk of developing the more severe form of the disease, namely PE, with independent predictors of VTE in the post-discharge period including previous VTE and cancer.