Banner img
Banner img

Potential Value of Online Anticoagulation Management

The feasibility and potential value of automated online anticoagulation monitoring of warfarin-treated patients

Brandi N. Thoma, Pharm.D. Candidate,1 Laura Barron, Pharm.D.,2 Megan Anderson, Pharm.D.,3 Marie B. Walker, BBA,4 Henry I. Bussey Pharm.D., FCCP,5

1Temple University School of Pharmacy, Philadelphia, Pennsylvania, 2College of Pharmacy, The University of Texas at Austin, El Paso, Texas, 3Drake College of Pharmacy and Health Sciences, Des Moines, Iowa, 4Clot Care Online Resource, San Antonio, Texas, 5 College of Pharmacy, The University of Texas at Austin, San Antonio, Texas.

Adapted from original poster presented at the American College of Clinical Pharmacy 2007 Annual Meeting, Denver, Colorado, October 13-17, 2007.


Purpose: Veeger, et al recently reported that 25% of warfarin-treated patients account for most out of range INR values and the majority of adverse events1. Previously, Rospond, et al reported that the majority of warfarin-treated patients could achieve prolonged periods of anticoagulation stability2. Current technology* can provide automated online anticoagulation monitoring (AOAM) which could substantially reduce the time commitment of patients and clinicians. The feasibility and value of AOAM is dependent on the extent and duration of anticoagulation stability in individual patients.

Methods: Warfarin doses and INR values of 52 consecutive patients seen in an anticoagulation clinic were analyzed to determine the degree of INR control for the group, the number of patients with stable INRs (2 months of therapy without a dosage change), and the duration of periods of INR stability. The initial 3 month dose titration phases were excluded as were peri-procedural INRs.

Results: 52 patients provided 194 patient-years of data. INR time in range was 69.63% (87.77% if the range was expanded by +/- 0.3 INR units). Fifty-two (100%) of 52 patients achieved INR stability. These 52 patients were stable for 149.6 patient-years of their 194.2 total patient-years (77%) of monitoring. The mean duration of stability was approximately 156 days with a range of 56 to 1014 days.

Conclusion: These data suggest that 100% patients can maintain INR stability for 77% of the time. Use of AOAM, therefore, could be employed for 77% of the time and reduce clinic visits by a similar amount to an average of once every 5 months.

*The ClotCare Online Management System


•Determine the time in therapeutic range for a cohort of patients on oral anticoagulation (OAC) managed by an outpatient anticoagulation clinic.
•Determine how many patients are able to achieve at least 56 days of INR stability without requiring a dosage change.
•Determine how many periods of stability account for the total patient years of the cohort.
•Determine the average duration of stability.


A recent study by Veeger et al. examined individual time in the therapeutic range (ITTR) for patients on oral anticoagulation and found that variations in anticoagulation control are not randomly distributed among all patients.1 Rather, the quartile of patients with the poorest INR control accounted for most of the out-of-range values and the majority of bleeding and clotting complications. Similarly, Rospond et al. previously reported that the majority of patients were able to achieve long periods of stability, while a minority of patients never achieved a stable INR.2 Their data suggest that the majority of patients may achieve good INR control and have a significantly lower rate of complications than the entire group. The long periods of INR stability in the majority of patients brings in to question the need for monthly monitoring of these patients. Monthly monitoring creates an inconvenience for patients that is expensive and may require lost time from work. These frequent visits are also costly and require clinician time which often may not be necessary. If long periods of INR stability are common, then it may be feasible to utilize INR self-testing and automated online anticoagulation monitoring (AOMA) to reduce the demands and costs of frequent clinic monitoring.


Sixty three patients seen over a 3 day period at the Anticoagulation Clinics of North America (ACNA) located in San Antonio, Texas were identified and data were extracted from their INR log sheets. Patients had to have at least 9 months of data to be included in the analysis and the initial 3 months of data were excluded as a dose titration period. INR log sheets contained the goal therapeutic INR range for each patient, date of visit, warfarin dosing regimen, INR/PT, and instructions on new dosing regimen, if a change was necessary. Eleven patients were excluded because they had less than 9 months of data. The data for the remaining 52 patients were entered into The ClotCare System as the following: date, INR value, and subsequent weekly warfarin dose. If a patient had their warfarin discontinued and later restarted, the first 3 months of data for the second treatment period was excluded in order to allow the INR to re-stabilize. If the INR target range was recorded as less than one INR unit (e.g.. 2 to 2.5), the expanded range of 2 to 3 was used to evaluate whether or not the patient data was within therapeutic range. Lastly, INRs that were intentionally lowered for a procedure were also excluded as were INRs for 2 weeks following such peri-procedural management.

Data were analyzed by the Rosendaal method for time in therapeutic range (TTR) as well as the TTR +/- 0.3 INR units. The number of patients who achieved INR stability (defined as at least 56 days without a change in warfarin dose), the time spent within a stability period, and the number of stability periods were also calculated.


Fifty-two patients yielded 194.22 years of data. TTR for the entire cohort was 69.63% and TTR +/- 0.3 INR units was 87.77% . INRs were less than 1.5 and greater than 5 1.26% and 0.45% of the time, respectively. Fifty-two out of 52 patients (100%) achieved stability. These patients achieved 351 periods of stability and 688 periods of instability. The 351 periods of stability accounted for 149.58 of the 194.22 total patient years of data (77.02%).

The mean of all stability periods was 155.5 days (approximately 22.24 weeks, or 5.17 months @4.3 weeks per month). The minimum period of stability, by definition was 56 days, while the maximum period of stability was 1014 days (approximately 144.86 weeks or 33.69 months @ 4.3 weeks per month). The standard deviation was 127.28 days.

The mean duration of time on a specific dose, including stable and unstable periods was 68.28 days (approximately 9.75 weeks or 2.27 months @4.3 weeks per month) with a range of 1 to 1014 days and a standard deviation of 97.35 days.

Table 1: INR Stability - 194.2 patient-years

Number (%) Stable Pats.

Pat-yrs(% time) Stable)

Mean Duration of Stability

Range of Stability Duration

52 (100%)

149.6 (77%)

5.17 mo.

56 d – 33.7 mo


The analysis of INR stability data from this cohort of patients is consistent with that previously reported in 19892 and in 20061 from two other anticoagulation services. 100% of 52 patients achieved at least one period of INR stability (8 weeks without requiring a dosage change) and the entire cohort spent 77% of the time in a stable period. The mean duration of stability was 5.17 months. Collectively, these data would suggest that it may be reasonable to reduce clinical evaluation of anticoagulated patients by approximately 77% with patients being evaluated, on average, approximately once every 5 to 6 months. Such infrequent monitoring, however, would require some system of automated surveillance and frequent INR testing. Frequent INR testing can be achieved with self-testing devices and automated surveillance can be achieved with online or telephonic systems. Self-testing and AOAM has the potential to reduce the cost of anticoagulation management for both clinicians and patients; and the resources exist to implement such a system. The financial feasibility of establishing such a system, however, will require a change in the practices of many payers that do not reimburse online or telephonic management.


1. Veeger NJGM, Piersma-Wichers M, Hillege HL et al. Early detection of patients with a poor response to vitamin K antagonists: the clinical impact of individual within target range in patients with heart disease. Journal of Thrombosis and Hemostasis 2006;4: 1625-7.

2. Rospond RM, Quandt CM, Clark GM, Bussey HI. Evaluation of factors associated with stability of anticoagulation therapy. Pharmacotherapy 1989; 9(4):207-213.


Dr. Bussey and Ms. Walker are co-developers and Ms. Walker is owner of The ClotCare System.