This study aimed to identify predictors of venous thromboembolism (VTE) in hospitalized cancer patients and develop a predictive model using demographic, clinical, and laboratory data. Our analysis showed that patient groups categorized under a very high risk, and high risk, patients with low hemoglobin levels and renal disease were at a significantly increased risk of developing VTE. We developed a VTE risk-assessment model (RAM) with moderate discriminatory performance, high specificity, and negative predictive value, indicating its potential utility in identifying patients without VTE risk. However, the model's positive predictive value and sensitivity were low due to the low prevalence of VTE within the analyzed population. Future studies are needed to analyze additional predictive factors, and to validate the effectiveness of our VTE RAM to safely rule out VTE, compare it with other VTE RAMs in hospitalized cancer patients, and address any limitations of our study.
Keywords: VTE RAM; VTE risk predictors; cancer patients; inpatient setting; venous thromboembolism.