The use of a Bayesian forecasting model in the management of warfarin therapy after total hip arthroplasty.

This study was performed to compare the computer-based and physician-based management of warfarin therapy after total hip arthroplasty (THA). The computer-assisted and control groups of patients were placed on warfarin postoperatively and followed for a 1-month period. A significant difference (P<.05) was found between the mean number of days needed to reach therapeutic anticoagulation in the control group (4.7+/-3.0 days) and the experimental group (2.8+/-1.4 days) and the proportion of patients in each group who were discharged with a subtherapeutic international normalized ratio (INR) (INR <1.5). The computer-based management of warfarin therapy was more efficient than unaided physician-based management and therefore may lead to improved, cost-effective patient care by reducing length of hospital stay and complications attributable to nontherapeutic anticoagulation in THA patients.