Pick a Topic:
Find info on a:
ClotCare complies with the HONcode standard for trustworthy health
ClotCare is a member organization of the Coalition to Prevent Deep Vein Thrombosis. Click here to learn more about the Coalition to Prevent Deep Vein Thrombosis and DVT Awareness Month, which is held each March.
Studies suggest that computer-assisted dosing programs perform well when used by experienced clinicians, but use by less experienced clinicians may be problematic
Henry I. Bussey, Pharm.D., FCCP, FAHA
Various computer programs are available to assist with managing patients on vitamin K antagonist (VKA) therapy. Two large recent trials suggest VKA management with 4 such systems is equivalent to experienced clinician management if an experienced clinician if overseeing the use of the programs. The fact that the systems failed to recommend a dose in 10% to 40% of instances, that the clinician had to change the recommended dose in 10% to 20% of instances, and that the suggested appointment interval was changed 34% of the time raise the question of how the anticoagulation control might have been affected if a less-experienced clinician had been overseeing the computer-generated recommendations.
Study #1: Investigators in the Netherlands compared their existing computer algorithm (TRODIS) to a new and more complex system algorithm (ICAD) in 712 patients.1 The ICAD system generated a dose recommendation in more instances than did the TRODIS (97.5% vs. 60.8%). Physician acceptance of the recommended dose, however, was lower with the ICAD system (79.3% vs 90.9% with TRODIS). Therefore, an acceptable recommendation was provided 77.4% and 55.3% of the time with the ICAD and TRODIS systems, respectively. The mean time in range was 79.8% and 80.2% with the ICAD and the TRODIS systems, respectively. It should be noted that these high in-range percentages are likely due in part to the fact that the target INR ranges used encompassed a 1.5 INR unit range rather than a 1 INR unit range as used in the U.S.
Study #2: Poller and colleagues reported the results of a randomized study that was conducted in 32 centers in 13 countries. This study compared two software programs (the PARMA 5 and DAWN AC systems) with "manual" dosing by experienced clinicians.2 Approximately 13,219 patients were randomized to either computer-assisted management or "manual" dosing. During 18,617 patient years of follow-up, and after excluding the first 3 weeks of dosing, the overall unadjusted INR time in range was 65.7% and 67% with manual and computer-assisted dosing, respectively. The medical staff provided the dosage 21.7% of the time in the computer-assisted group because the computer failed to do so (10.8%) or the dose was changed by the medical staff (10.9%). The computer-advised appointment intervals were changed 34.3% of the time.
Conclusion: Although these two new studies suggest that computer programs perform reasonably well in guiding anticoagulation management when their use is supervised by an experienced clinician, there appears to be a substantial frequency with which the programs do not provide a recommendation at all or do not provide an acceptable recommendation. One must question, therefore, how these systems would perform when used by less-experienced clinicians.
I should point out that I may have a bias in that I do not believe that available computer dosing programs are likely to achieve optimal warfarin management. Further, I have been involved in the development of another computer management system that provides clinical cues but does NOT provide a dosage recommendation. Although studies show that an acceptable dose can be generated in many instances, I am more concerned about the cases in which an unacceptable dose may be suggested by the computer and accepted by a clinician. While I encourage and support the use of software programs that assist in managing patient care, I feel that optimal anticoagulation management calls for dosing decisions to be made by a knowledgeable, experienced anticoagulation clinician. I realize that others may feel differently and I invite those with different views to submit a letter to the editor at firstname.lastname@example.org.
van Leeuwen Y, Rombouts EK, Kruithof CJ, van der Meer FJ, Rosendaal FR. Improved control of oral anticoagulant dosing: a randomized controlled trial comparing two computer algorithms. J Thromb Haemost. 2007 Aug;5(8):1644-9. Epub 2007 May 7.
Poller L, Keown M, Ibrahim S, et al. An international multicenter randomized study of computer-assisted oral anticoagulant dosage vs. medical staff dosage. J Thromb Haemost. 2008 Jun;6(6):935-43.
ClotCare is a 501(c)(3) non-profit organization generously supported by your tax-deductible donations and grants from our industry supporters.
Click here to view full list of new postings