Estimation of clinical and economic effects of prophylaxis against venous thromboembolism in medical patients, including the effect of targeting patients at high-risk

Abstract

Background: The clinical and economic effects of medical thromboprophylaxis (MT) using low molecular weight heparin in Australia are unknown. Aim: To estimate the effects of MT in Australia. Methods: A decision tree model of MT was populated with national data for medical admissions. The Prevention of Recurrent Venous Thromboembolism (PREVENT) trial was chosen as the primary data source because its design uniquely avoided bias caused by treatment of sub-clinical events. Clinical efficacy and costs were estimated compared with no prophylaxis, assuming full compliance and according to three definitions of eligibility. Effectiveness was estimated as thrombotic events saved, mortality from bleeding or pulmonary embolus (PE), cost and $/year of life saved. Model outputs were subjected to sensitivity analysis. Results: MT decreased thrombotic events, and the numbers avoided increased as eligibility broadened (deep vein thrombosis (DVT): 2597, 2771 and 3232 at restricted, intermediate and broad eligibility; PE: 454, 484 and 565 respectively). The annual cost of no prophylaxis was $88.7m. Costs were reduced at most restricted eligibility (-$7.9m), but increased by $3.0 and $32.1m at broader eligibility. PE deaths declined, but this was offset by deaths from haemorrhage, causing a net increase (158, 299 and 672 respectively). Estimates were sensitive to the incidence of venous thromboembolic event (VTE), case-fatality rates for PE and bleeds and the relative risk reduction for PE with prophylaxis. Conclusions: Under PREVENT trial conditions, MT avoids up to 3200 DVT and 565 PE events annually, but may increase mortality.