Venous thromboembolism (VTE) is a complex multifactorial disease, mainly manifested by deep vein thrombosis (DVT) and pulmonary embolism (PE). VTE events increase the length of hospitalization and treatment costs and seriously affect the quality of life of patients, so it is increasingly appreciated to identify high-risk patients with VTE and take preventive measures. The Padua prediction score (PPS) and Caprini risk assessment model (RAM) are widely used in clinical practice as common risk assessment scales in medical and surgical departments, respectively. And D-dimer levels have been considered as a well indicator to rule out acute VTE. Previous epidemiological studies on VTE have found the risk of VTE is significantly higher in hospitalized patients than in the general population and the prophylaxis decisions vary among countries, hospitals and departments, indicating current in-hospital VTE prevention strategies are far from optimal and it's imperative to regionalized control of VTE. Therefore, a single-institution-based risk profile study of in-hospital VTE patients is designed to explore current situation of VTE occurrence and predictive efficacy of widely used risk assessment models as well as D-dimer in one of the general hospitals in Beijing, China.

In this study, the investigators recruited adult inpatients attending Tsinghua Chang Gung Hospital in Beijing from June 2018 to April 2020, screened for patients with a hospital stay over 3 days and new-onset of VTE during hospital stay, and identified the in-hospital VTE patients after excluding patients who presented for DVT and/or PE. Similarly, negative patients were selected by not having a VTE during hospital stay. The diagnosis of VTE was confirmed as the occurrence of a critical value alert during hospitalization, which was predefined as ultrasound/radiology report of DVT or/and PE. Data was collected from the TEACH database of Beijing Tsinghua Chang Gung Hospital affiliated with Tsinghua University.

Patients baseline characteristics were extracted from database. VTE risk assessment was conducted with either Caprini risk assessment model (surgical patients) or Padua prediction score (medical patients). The evaluation was started upon admission and regularly repeated depending on the patient's condition. The last evaluation was conducted before discharge. The D-dimer values during hospitalization were also collected.

The rate of in-hospital VTE, distributions of in-hospital VTE patients and the receiver operating characteristic curves were analyzed.

Inclusion Criteria:

1. Age 18 and over, male or female
2. Inpatients with a hospital stay over 3 days
3. With or without new-onset of VTE during their stay

Exclusion Criteria:

1. Patients who presented for DVT and/or PE
2. Patients who were admitted in emergency department

The investigators recruited adult inpatients attending Tsinghua Chang Gung Hospital in Beijing from June 2018 to April 2020.