Randomized Controlled Trial of Supervised Patient Self-Testing of Warfarin Therapy Using An Internet Based Expert System

Improved anticoagulation control results in improved outcomes, with a decrease in the incidence of bleeding and thromboembolic events. The outcomes of oral anticoagulation therapy (OAT) are dependent on the model of care used to manage warfarin therapy, with better outcomes associated with the use of a specialised anticoagulation management service (AMS), computerized decision support and increased testing frequency (facilitated by patient self-testing (PST) of the international normalised ratio (INR)) (Ansell and Hughes 1996). This study uses a novel strategy to combine the advantages of these three approaches to warfarin management. A prospective, randomized controlled cross-over study was carried out at the anticoagulation clinic of Cork University Hospital, Ireland, to test the hypothesis that supervised PST using an internet based expert system could provide comparable anticoagulation control to that provided by traditional outpatient AMS management. This was a cross-over study; patients served in both AMS and supervised PST arms, with random assignment as to the order of management. Both arms were six months in duration. Patients on established long term OAT who had internet access were eligible for enrolment. During the supervised PST arm of the study, patients measured their INR at home using the CoaguChek XS^® point of care meter, (Roche Diagnostics, UK) either twice weekly, weekly or every two weeks depending on their anticoagulation control. They entered this result along with other information relating to their warfarin therapy (e.g. signs or symptoms of thrombosis or hemorrhage, missed doses, concurrent illnesses, dietary or medication changes etc) onto the CoagCare^® (Zycare Inc, Chapel Hill, NC) web page. Patients with a therapeutic INR and no other issues were automatically provided with algorithm-derived dosing and repeat testing instructions. Patients with non-therapeutic results, or symptoms suggestive of thromboembolic or hemorrhagic complications, were prioritized for caregiver review according to problem severity. During the AMS arm, patients were required to attend the clinic every four to six weeks, or more frequently, if clinically indicated. Dosage adjustments were performed by the anticoagulant clinic staff using the laboratory INR and the APEX^® (iSOFT, UK) computer decision support software. The primary outcome variable to assess anticoagulation control was the difference in time in therapeutic range (TTR) during the AMS management and supervised PST management period. In addition, the number of INR measurements indicative of serious under- or over-anticoagulation and the number of serious hemorrhagic or thrombotic adverse events were also compared. One hundred and sixty two patients were enrolled over a nine-month period (July 2006–April 2007). The majority of study patients were male (61.4%) and the mean age of the study population was 59.6 +/−14.3 years. One hundred and thirty two (81.5%) patients completed both arms of the study. The mean TTR during the AMS arm was 60.2% +/−19.5%, which increased to 71.4% +/− 13.6% during the PST period (p<0.001). Eighty-seven patients (65.91%) achieved better anticoagulation control during the supervised PST period. Patients measured their INR almost four times more frequently while home-testing giving a mean frequency of INR testing of 4.5 days compared with 17.4 days for the AMS period. Extreme INR values (< 1.5 and > 5.0) occurred more frequently during the AMS arm of the study (6% vs 2.4%, p<0.001). There was no significant difference in the adverse event rate between the two study periods. Table 1.

Table 1. Adverse events

Daily time to manage 70 patients ranged from 10 to 45 minutes (mean 23.19+/−9.48 mins/day). This novel system of supervised PST using an internet based expert system improves the clinical effectiveness of OAT when compared with management by a specialized anticoagulant management service.