Oral anticoagulants are one of the most common causes of drug-induced adverse events.1 Expert dose management, which keeps the patient within a narrow therapeutic international normalized range (INR), is required to avoid adverse events and optimize therapy. Anticoagulation clinic (ACC) management has been shown to improve outcomes compared with the routine management provided by physicians,2 but estimates suggest that fewer than 50% of patients in the US on oral anticoagulants are managed by such clinics. Point-of-care (POC) prothrombin time monitoring technology, developed 20 years ago, introduced a new model of care by allowing for fingerstick INR monitoring and patient self-testing (PST). Under this model, patients may either monitor their own INR at home and receive dose-adjustment guidelines from their physician, or actually adjust their own anticoagulation dose after proper training.2 Numerous published clinical studies conducted during the past 10 years have validated the benefits and value of POC testing for anticoagulation monitoring compared with venous blood draws and laboratory testing. Despite the clinical evidence, many healthcare professionals have been reluctant to incorporate POC testing into their practices. There are several devices available for self-testing that rely on capillary whole blood and are lightweight, portable, easy-to-use, accurate, and precise (see Table 1). INR results from these portable devices have been shown to equal those from phlebotomy plasma INRs determined in a central laboratory.3
Patient Self-testing/Patient Self-management
Many studies4–28 have reported on the accuracy and precision of POC instruments when used by patients to obtain an INR, and on the suitability of POC instruments for monitoring anticoagulant therapy. Limitations to their accuracy and precision include greater differences compared with a standard plasma-based methodology as INRs increase above the therapeutic range,6,26 incorrect calibration of the international sensitivity index (ISI) of the POC instruments,6 the inability to calculate a mean normal prothrombin time (PT),27 and inaccuracies in INR determination in patients with antiphospholipid antibodies. Multiple studies focusing on clinical outcomes and patient satisfaction have determined the suitability of PST.
It is important to note the comparator group used in controlled trials, since this influences differences. When a usual care (UC) model of anticoagulation management is the comparator group, the PST or patient self-management (PSM) arms will often show marked improvement in time in therapeutic range (TTR, a surrogate measure of quality) or in efficacy and safety outcomes. Alternatively, when PST or PSM is compared with a control arm managed by an ACC, the differences are minimal. Table 2 summarizes randomized trials in which 50 or more patients were studied and clinical outcomes were reported as TTR, adverse events, or both.27–37 Studies were stratified by whether they were PST versus PSM trials and the control groups were UC or ACC care. In a large randomized study, Beyth et al.27 compared PST with UC management. The PST group had a 5.6% incidence of major hemorrhage compared with a 12% incidence in the UC group (p=0.049). Venous thrombosis occurred at a rate of 8.6% in the PST group compared with 13% in the UC group, but the difference was not significant. In a number of trials comparing PSM with UC, mean TTR for the PSM arms was 73% versus 59.9% for the UC arms.28,34–36,38–41 None of these trials showed differences in major hemorrhage or thrombosis, although none of them were powered to do so. When one assesses the outcomes of PST compared with management by an AC, the TTR is only minimally improved (73.3 and 67.1%, respectively).32,33,38 The same is found when PSM is compared with AC care (71.4 and 66.6%, respectively).29,30,37
Only one trial found a significant difference in the rate of major bleeding and thromboembolism.30 These studies and others have been the subject of several detailed meta-analyses.42,43 These meta-analyses are somewhat problematic in that some combine results from PSM and PST, while in others the controls are a combination of UC and ACC management. In the most comprehensive meta-analysis to date, Heneghan et al. pooled estimates from 14 randomized trials of PST showing a significant reduction in thromboembolic events (odds ratio [OR] 0.45, 95% confidence interval [CI] 0.30–0.68), all-cause mortality (OR 0.61, 95% CI 0.38–0.98), and major hemorrhage (OR 0.65, 95% CI 0.42–0.99) versus the comparator.42 For PST and PSM combined, there were significant reductions in thromboembolic events (OR 0.27, 95% CI 0.12–0.59) and death (OR 0.37, 95% CI 0.16–0.85), but not major hemorrhage (OR 0.93, 95% CI 0.42–2.05). Both PST and PSM studies are included.
Based on these data, portable PT monitors offer the potential to lower the risk–benefit ratio of anticoagulant therapy, improve patient satisfaction and possibly patient compliance, and, by reducing the labor intensity of physician management, encourage the use of warfarin when indicated. Although larger randomized, prospective trials are needed, current data suggest that future management of anticoagulation should include PST and PSM therapy. This fact was recognized by the most recent American College of Chest Physicians (ACCP) Consensus Conference on Antithrombotic Therapy, where the following recommendation was made: “PSM is a choice made by patients and healthcare providers that depends on many factors. In patients who are suitably selected and trained, PST or PSM is an effective alternative treatment model. We suggest that such therapeutic management be implemented where suitable (Grade 2B).“2
While generalized training guidelines for PSM of any disease state apply to self-anticoagulation, the high risk–benefit ratio of warfarin and the potentially serious consequences of non-therapeutic levels of anticoagulation warrant special considerations in patient education. The majority of patients taking oral anticoagulants are older with multiple co-morbidities that require special attention. Patient education must begin with candidate identification. Certainly, not all patients are able or willing to perform the test and keep the necessary records to ensure safe and accurate monitoring. The most important consideration in choosing good candidates is incentive. Incentives differ among patients. Less frequent clinic visits, particularly for patients who have difficulty reaching the clinic, may be sufficient for some patients, whereas patients who have an aversion to phlebotomy or poor venous access might appreciate the ease of obtaining results from a fingerstick. Some patients are excited to learn about their conditions and to actively participate in their care. Patients at an increased risk for hemorrhage, or individuals whose PT fluctuates widely, might welcome the opportunity for more frequent and less invasive monitoring. For some patients, a trusted physician’s recommendation will suffice. Other important considerations in the identification of good candidates include the adequacy of motor skills to perform the test, sufficient mentation and memory, sufficient literacy, and adequate eyesight, with or without glasses, to see the device screen.
Patients who have the potential to benefit the most include those who have limited access to testing due to distance or disability and those who require increased testing frequency (i.e. those patients with heart failure, liver cirrhosis, or marked variation in PT values). For children or patients with disabilities that would preclude their ability to competently perform their own testing, the availability of a parent, spouse, or qualified care provider allows these patients to be considered for PST.
There are two aspects of patient education: patients must be educated about anticoagulation so that they understand what, where, and why they are performing the test, and they must also be trained in the use of the monitor.
The challenge for the healthcare provider is to convey the necessary information to the patient and train the patient in the use of the monitor without conveying excessive information. Ideally, training would be provided by a multidisciplinary team including nursing, pharmacy, laboratory, and dietary staff. Topics to be covered include: mechanism of anticoagulant action; importance of maintaining a therapeutic range; interpretation of test results; factors that affect the PT—diet, drugs, and liver function; bleeding risks and monitoring; testing frequency; and indication and duration of therapy. Trainers should simplify the process as much as possible and provide written instructions in short, clear, simply phrased sentences. The tasks allocated to the patient should be clearly listed in simple, straightforward, consistent terms, and should be located in an easily accessible place. The trainer should refer to the list throughout the training session. When instructing patients in the actual use of the instrument, the trainer must remember that the majority of patients are not comfortable with electronics and thus may be intimidated by the apparent complexity of the monitoring device and its accompanying materials. The trainer should remember that older patients often benefit from consistency, and their routine should be disrupted as little as possible. For a patient who is accustomed to laboratory phlebotomy and a subsequent telephone call regarding dosage changes, the instrument can be overwhelming. All efforts should be made to facilitate this transition and to inspire confidence in the method. Once patients have performed the test, they must record their results. A calendar may function as both a testing schedule and a place for recording results. The trainer must verify that the patient understands the following before leaving the clinic: how to use the instrument, the days and times to perform the test, where to write down results, and what to do with the information (i.e. PSM versus doctor or anticoagulation-based management).
The trainer should try to anticipate any problems and develop a plan to solve them. The trainer should not assume that patients remember everything they have been taught, even though they may seem to understand at the time. Patients should have a point person to call if they experience any problems.
Finally, the trainer must give patients the opportunity to express concerns and ask questions. It is also important for the trainer to reinforce the incentive identified by each patient and the trainer at the beginning of the session. A transitional period is appropriate, during which patients perform PST measurements once a week with comparison testing carried out by the physician. This allows for the evaluation of the results of patients compared with the results of professionals and verification of proper functioning of the device, and addresses patient questions. Once patients are placed on the PST program, INR determinations should be supplemented with clinic visits every six to 12 months. The implementation of PST may seem daunting; however, multiple companies assist patients in obtaining these devices. Table 3 lists these companies and contact information, as well as the devices the companies support.