Introduction: Venous thromboembolism (VTE) occurs frequently in at-risk hospitalized patients, and prophylaxis of VTE is significantly underused. We sought to preliminarily assess the validity of Caprini risk assessment model, a famous individual VTE risk assessment model, in Chinese hospitalized patients with VTE. Materials and Methods: We undertook a retrospective study combined with a follow-up study among 347 confirmed VTE patients from a Chinese hospital. Results: Compared with the other two risk assessment models (RAMs), Caprini model can classify much more VTE patients into high or highest risk level and the differences were statistically significant (Caprini model vs Kucher model, p<0.0001; Caprini model vs the Padua Prediction Score, p<0.0001). Caprini model exhibited much more effect at assessing patient's VTE risk among surgical patients than nonsurgical patients(average risk score, 5.71 +/- 2.54 vs 4.36 +/- 2.51, p<0.0001; by Wilcoxon rank sum test, p=0.001 in favor of the prediction effect of the RAM in surgical patients). Kaplan-Meier analysis showed that patients classified into low and highest risk level by Caprini model had increased hazard for VTE recurrence when compared with patients classified into moderate and high risk level, but the result was not statistically significant (p=0.222). Conclusions: Our study preliminarily suggests that the Caprini risk assessment model is a practical and effective tool to assess the risk of VTE among unselected Chinese inpatients and may also be useful in predicting the risk of VTE recurrence. However, future studies with control group and prospective validation of the model in Chinese inpatients are needed. (C) 2012 Elsevier Ltd. All rights reserved.