PURPOSE: Venous thromboembolism (VTE) is a morbid, costly, and potentially fatal complication in hospitalized medical patients. Prior epidemiologic studies have focused on narrow community populations to determine VTE incidence. The U.S. inpatient hospitalization database, Healthcare Cost and Utilization Project (HCUP), can be utilized to identify patients at risk for deep vein thrombosis (DVT) and pulmonary embolism (PE). Through a probability model that we developed, our aim is to estimate VTE events in major subgroups of acutely ill medical patients across all U.S. hospitals.
METHODS: We queried the HCUP database, using clinical classification diagnostic codes, to identify acutely ill medical patients over 40 years old with a hospital stay of at least 2 days. We then applied each population to a decision tree model. The model was populated with VTE event rates and prophylaxis prescribing rates. All event rate probabilities were derived from published literature. The model estimated symptomatic VTE events and VTE related deaths for the year 2003.
RESULTS: We found more than 8 million hospitalized acutely ill medical patients at risk for VTE events. The largest diagnosis patient groups were heart failure, respiratory failure, pneumonia, and cancer (Table 1), representing about 70% of the hospitalized population at-risk. Prophylaxis rates across groups ranged from 15.3% to 49.2 %. We estimate that 203,968 hospital acquired VTE events occur annually, affecting 3 of every 100 medical patients. Of these VTE events, there were 161,669 symptomatic DVTs, 42,299 symptomatic PEs, and 53,499 VTE-related deaths. Sensitivity analysis shows that improving prophylaxis from 28.8% to 95% will reduce the frequency of these events by approximately 40%.
CONCLUSION: Our study demonstrates that acutely ill medical patients represent a major population at risk for VTE events in USA hospitals. The largest populations warranting VTE prophylaxis are those with congestive heart failure, respiratory failure, pneumonia, and cancer.
CLINICAL IMPLICATIONS: Subgroups of acutely ill medical patients are at risk for VTE and warrant VTE prophylaxis. Clinicians should utilize pharmacologic or mechanical interventions to reduce this risk.
DISCLOSURE: John Fanikos, Consultant fee, speaker bureau, advisory committee, etc. I have served as a consultant and am a member of the Speakers Bureau for sanofi-aventis; No Product/Research Disclosure Information