Date Approved


Graduate Degree Type


Degree Name

Public Health (M.P.H.)

Degree Program

Public Health

First Advisor

Sarah Nechuta, MPH PhD

Academic Year



Deep vein thrombosis (DVT) is an often underdiagnosed disease that can be life threatening for individuals. Currently, the best practice approach to assess the risk of venous thromboembolism (VTE) in patients within a rehabilitation hospital is unknown. We examined how well the Caprini risk assessment (CRA) score predicts VTE events (deep vein thrombosis (DVT), pulmonary embolism (PE)) within an inpatient rehabilitation hospital. Between January 2015 and October 2020, VTE events, the CRA Score, and risk factors were collected from a chart review conducted at Mary Free Bed Rehabilitation Hospital (MFB) among 3,091 patients. Results from the unadjusted logistic regression model showed that patients who have scores in the moderate-risk (odds ratios (OR)=0.076; 95% confidence interval (CI): 0.010-0.587) and high-risk categories (OR=0.961; 95% CI: 0.476-1.943) have lower odds of being diagnosed with DVT or PE. The highest risk category ( ≥ 5) showed an association between the CRA score and DVT or PE (OR=1.031; 95% CI: 0.540-1.967). When adjusting for medical service and discharge location, findings were similar among all four risk categories. The area under the receiver operating characteristic (ROC) curve, was 0.57 for the unadjusted model and increased to 0.61 for the adjusted model. Although a small association between the CRA score and prediction of VTE was noted, the low incidence of these events in the patient population limited study sample size and precision. This may be because almost all patients were on prophylaxis medication.