SEARCH

SEARCH BY CITATION

Keywords:

  • anticoagulation;
  • coumadin;
  • patient education;
  • warfarin

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Objective

Lack of patient knowledge has been associated with poor anticoagulation control, but the effect of patient education on clinical outcomes is unclear. We systematically reviewed the effect of supplemental patient education vs. usual care on hemorrhage, thromboembolic events (TEEs), time in therapeutic range (TTR) and knowledge test scores for all oral anticoagulants.

Data sources

The data sources were electronic databases, including MEDLINE, EMBASE, CENTRAL, CINAHL and IPA, to February 2012 examining any oral anticoagulant. We reviewed references for additional potentially relevant studies.

Methods

Only randomized controlled trials (RCTs) were considered. Data extraction and quality assessment were conducted with GRADE. Pooled relative risks (RRs) were calculated, and heterogeneity was determined by use of χ2 and I2 statistics.

Results

Seven RCTs (n = 1209) were included in the systematic review, and five RCTs (n = 847) in the meta-analysis. All included studies examined vitamin K antagonists. No significant difference was found for hemorrhage (RR 0.92, 95% confidence interval [CI] 0.04–20.56), TEE (RR 0.66, 95% CI 0.10–4.39), a composite outcome of hemorrhage or TEE (RR 0.48, 95% CI 0.23–1.01), or TTR (mean absolute difference of 2.02%, 95% CI − 2.81 to 6.84). Evidence was conflicting on the impact of supplemental education on test scores. All trials had at least one substantial methodologic limitation.

Conclusion

Current evidence does not support supplemental patient education as a means to improve patient outcomes, but the quality of this evidence is poor. Larger randomized trials are needed with longer follow-up, recruitment of patients initiating anticoagulation in primary care settings, and clearly defined education interventions.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Oral anticoagulants (OATs) are effective for the primary and secondary prevention of thromboembolic events (TEEs) in a variety of conditions, including stroke prophylaxis in patients with atrial fibrillation [1, 2] or prosthetic heart valves [3], and venous thromboembolism [4, 5]. OATs are generally well tolerated, but carry a risk of serious adverse events, primarily bleeding or recurrent thromboembolism. Several new anticoagulants, such as dabigatran, rivaroxaban, and apixaban, are now available and do not require frequent blood testing for monitoring, but still require patient education regarding adherence, dose changes, and drug interactions. Vitamin K antagonists (VKAs) will continue to serve an important role, owing to the higher cost and renal contraindications of the newer medications [6].

Supplemental patient education is frequently proposed as an intervention for improving patient outcomes. These measures, such as dedicated patient education sessions [7-10] or brochures [11], provide more content and detail than the information typically given by a patient's usual healthcare provider. In addition, supplemental patient education has, at times, been combined with other strategies such as focused anticoagulation clinics or patient self-testing with home monitoring devices [7, 12]. Recent guidelines suggest that healthcare providers who manage oral anticoagulation therapy should incorporate patient education to maintain a best practice approach [13].

Observational studies have shown that knowledge test scores after educational programs are associated with improved anticoagulation control and decreased risk of bleeding [9, 14-16]. A recent systematic review identified strategies for patient education and instruments for measuring patient knowledge, but did not quantify the clinical effect of such interventions [17]. Patient education requires additional time and human resources to manage, and thus must be proven to be beneficial before it can be considered to be cost-effective.

Our objective in this study was to systematically review the evidence on supplemental patient education for those starting or continuing OATs, and its effect on clinical outcomes.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Study inclusion

Randomized controlled trials (RCTs) assessing supplemental patient education on the effectiveness and safety of OATs as compared with usual care were considered for inclusion. Studies with patients taking the new OATs dabigatran, rivaroxaban, and apixaban, as well as VKAs, were eligible for inclusion. Included studies enrolled adult patients aged ≥ 18 years, for any indication in an inpatient or outpatient setting, regardless of treatment duration.

Data sources

We searched bibliographic databases and bibliographies of all eligible studies from 1948 to January 2010, then updated to 9 February 2012. Sources included MEDLINE, EMBASE, CINAHL, the Cochrane Central Register of Controlled Trials, and the International Pharmaceutical Association. Search terms included ‘anticoagulation’ and related OATs, combined with ‘education’, limited to RCTs (Appendix 1). Reviews and recent guidelines on the topic were examined to identify other eligible trials, as well as references cited by included studies. Studies were considered irrespective of publication status.

Study selection

Abstracts were screened and excluded if patients were not taking OAT, or if the study was not an RCT. The full text of the remaining potentially relevant studies was examined. A study was included if it involved an analysis of an educational intervention and evaluated at least one of the relevant outcomes. Primary outcomes were hemorrhagic events and TEEs, but we also abstracted the surrogate outcomes of time in therapeutic range (TTR) of International Normalized Ratios (INRs) and knowledge test results. It has been noted that a 10% difference in TTR was described as a clinically significant effect size in a previous study, as it is strongly related to clinical outcomes of hemorrhagic events and TEEs [18]. A second reviewer assessed a sample of 50 articles for both study selection and abstraction. Disagreements were resolved by consensus or arbitration.

Data extraction and quality assessment

A standardized data extraction and quality assessment form was developed from a template [19] and then piloted. The authors of two included studies [10, 20] were contacted for further clarification, but they did not review or approve our data extraction. Hemorrhagic events and TEEs were accepted as per the individual study definition. TTR, the number of days for which the INR was in the therapeutic range relative to the total number of days in the observation period, was calculated in all studies by the method of linear interpolation [21]. The OAT-related knowledge of patients was assessed through post-education knowledge test scores by use of specific tests included in the studies.

Methodologic quality was assessed according to the Cochrane Handbook [22] and Grading of Recommendations Assessment, Development, and Evaluation (GRADE) Profiler (version 3.6) criteria [23, 24]. Specifically, we evaluated the method of randomization, allocation concealment, blinding, analysis by intention to treat, and completeness of follow-up. Inconsistency (heterogeneity) between studies, indirectness, imprecision, and other potential sources of bias, such as publication and reporting bias, were also evaluated with GRADE.

Summary measures

Pooled estimates of relative risk (RR) for hemorrhagic events and TEEs were calculated according to a Mantel–Haenszel method with events per number of participants in a given study arm. Data were pooled by use of a random effects model, except for the composite outcome, for which a fixed effects model was used to moderate any undue influence of small studies [22]. For mean differences in TTR, we employed a random effects inverse variance model. Where standard deviations were not available, we converted from standard error or imputed 95% confidence intervals (CIs) for difference in means [22]. Knowledge scores were calculated as percentage correct responses.

Heterogeneity was assessed by examining clinical characteristics of included studies as well as by formal statistical testing with χ2 and I2. Funnel plots were planned to assess publication bias. Meta-analysis was performed with Review Manager (RevMan, version 5.1) software [25], according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [26], with summary findings imported to GRADEpro. Subgroup analyses based on indication for OATs and methodologic quality parameters were planned.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Study selection

Our search identified 1499 citations, of which 1326 citations with abstracts remained after duplicates were removed (Fig. 1). We excluded 1319 articles for various reasons, as detailed in Fig. 1. None of the studies on newer OATs met our inclusion criteria. We included seven RCTs with patient education arms in our systematic review [7, 8, 10, 11, 20, 27, 28]. Two of these studies only evaluated knowledge test scores, and thus were excluded from meta-analysis [11, 28].

image

Figure 1. Study flow chart. RCT, randomized controoled trial.

Download figure to PowerPoint

Table 1 outlines the characteristics of the seven studies, which randomized a total of 1209 patients to supplemental patient education vs. usual care in hospitals or anticoagulation clinics. The ages of the patients enrolled in the trials ranged from 18 years [8] to 91 years [8, 10] (mean: 59.5 years, based on available data from five studies) [7, 8, 10, 20, 27], and 502 patients (41.4%) were female. Patients received warfarin in five studies [10, 11, 20, 27, 28], and acenocoumarol or phenoprocoumon in two [7, 10]. Venous thromboembolism and atrial fibrillation were the most common indications for VKAs. Patients who could not speak the language of instruction or who had cognitive impairment were explicitly excluded in all but one study [10].

Table 1. Summary characteristics of seven included studies
StudyNo. randomized (n)IndicationOral anticoagulant agentDuration of VKA prior to studyOutcome measureSetting
  1. NR, not reported; TTE, thromboembolic event; TTR, time in therapeutic range; VKA, vitamin K antagonist; VTE, venous thromboembolism.

Pernod et al. [8]302VTENRStart of study (anticoagulant-naïve)1°: Hemorrhagic event or TEEInitially inpatients, discharged to private practice setting
Gadisseur et al. [7]221Any requiring ≥ 3 months of anticoagulationAcenocoumarol, phenprocoumon≥ 3 months

1°: TTR

2°: Hemorrhagic event or TEE

Outpatient anticoagulation clinic service
Machtinger et al. [20]157Any requiring ≥ 3 months of anticoagulationWarfarin≥ 3 months

1°: Time to reach target anticoagulation control

2°: TTR

Outpatient anticoagulation clinic
Laporte et al. [10]86Any requiring ≥ 3 months of anticoagulation; primarily VTEWarfarin, acenocoumarolStart of study (anticoagulant-naïve)

1°: TTR

2°: Hemorrhagic event or TEE

Initially inpatients, discharged to anticoagulation clinic
Khan et al. [27]81Atrial fibrillationWarfarin≥ 12 monthsTTROutpatient anticoagulation clinic
Clark and Bayley [11]45NRWarfarinNRKnowledge (post-test scores)Hospitalized inpatients
Mazor et al. [28]317Any requiring ≥ 3 months of anticoagulationWarfarin≥ 3 monthsKnowledge test scoresOutpatient anticoagulation clinic
Total1209     

The nature of the educational intervention varied, from single teaching sessions via video [28], self-guided instruction booklet [11], one-on-one teaching with written information [8] or group session with written information [27] to multiple interactive sessions [7, 10, 20]. The curricula covered indications for VKAs, benefits and risks, the importance of INR surveillance, drug interactions, and the effect of diet. ‘Usual care’ patients received less education, which was generally unspecified [7, 10, 20, 27, 28] or was mainly standard written information (Table 2) [8, 11].

Table 2. Description of intervention and control characteristics
StudySupplemental education interventionControlOther
TimeEducatorCurriculum
  1. INR, International Normailzed Ratio; VTE, venous thromboembolism.

Pernod et al. [8]A single 20–30-min sessionOne-on-one with pharmacist, nurse or physician trained in therapeutic education

Pathophysiology of VTE, aims of anticoagulation, benefits and risks, INR surveillance, drug and diet interactions, lifestyle considerations

Picture book and information booklet provided

Unstructured information delivered by physicians, including a standard booklet by the French Heart AssociationCompliance not reported
Gadisseur et al. [7]Three weekly 90–120-min sessionsGroups of 4–5 by physician and paramedic personnel

Blood coagulation system, oral anticoagulant therapy, drug interactions

Written information provided

Usual careCompliance not reported
Machtinger et al. [20]Approximately 2-min session after each routine clinic visitOne-on-one with a trained research assistantStandard medication counseling and non-standardized medication management by pharmacists was provided A visual medication schedule was provided with a ‘teach back’ using a standardized script; additionally, one attempt was made to correct patient understandingUsual care, including standard medication counseling and non-standardized medication management by pharmacists with expertise in anticoagulation lasting 5–10 min

Medication concordance and adherence was assessed at enrollment and two scheduled follow-ups

Financial incentive provided

Laporte et al. [10]Daily educational sessions until discharge (length of each session unknown)One-on-one with inpatient physicians and nursesVisual information provided; necessity for compliance and causes of anticoagulation instability reviewedUsual care, including minimum information consistent with ethical anticoagulation managementCompliance was assessed by use of electronic pill bottle and tablet counts
Khan et al. [27]A single 2-h sessionInteractive group sessions of 2–3 per group; led by a physicianWritten educational material (by Roche) on atrial fibrillation, benefits and risks of anticoagulation, pharmacology of warfarin, compliance, drug interactions, and diet concernsUsual careCompliance not reported
Clark and Bayley [11]Not specified, self-directed by patient (up to 72 h)Self-directed learning by patientProgrammed instruction booklet (written sequential presentation of learning steps, with required questions and immediate feedback) or two-page information handout including indication, mechanism of action, laboratory testing, dosage, drug and diet interactions, and safety factorsNo specific printed or verbal instruction from the investigatorsCompliance not reported
Mazor et al. [28]Not specified, self-directed by patientVideos included a physician–patient encounter; created by clinical members of the research teamVideo depicting a physician–patient encounter about safe warfarin management, including dosing regimen, INR testing, drug and diet interactions, signs of bleeding and compliance. Narrative evidence (patient anecdotes), statistical evidence or both were usedUsual care (no video)Compliance not reported

The length of follow-up in the included studies ranged from 3 to 6 months. The small number of studies meant that publication bias could not be assessed and subgroup analyses based on indications for VKAs or methodologic quality could not be explored.

Quality of evidence and outcomes

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

The quality of evidence from the included seven studies was rated as low for clinical events as well as for TTR, as there was significant risk of bias in all studies (Table 3, Appendix 2, Appendix 3). Limitations of the individual studies included unclear or missing allocation concealment, inconsistency, indirectness, and imprecision.

Table 3. Methodologic quality of included randomized controlled trials
StudyIntervention (no. randomized)Control (no. randomized)Random sequence generation adequateAllocation concealmentDescription of loss to follow-up adequate?Drop-out rate after randomization (%)Outcome assessors blinded?Intention to treat analysisApproximate follow-up time
  1. NR, not reported.

Pernod et al. [8]160142YesNot reportedYes (no loss)NoneYesYes3 months
Gadisseur et al. [7]60161No (Zelen design)No (Zelen design)Yes6NoNo6 months
Machtinger et al. [20]7978YesNoYes10NRNo3 months
Laporte et al. [10]4343YesNoYes6NoYes6 months
Khan et al. [27]4140No (Zelen design)No (Zelen design)Yes5NRNo6 months
Clark and Bayley [11]3015NRNRNoNRNRYes24–72 hours
Mazor et al. [28]22790NRNRYes32NRNo3 weeks

Hemorrhagic events

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Three studies [7, 8, 10] (n = 609 randomized) reported on hemorrhagic events (Fig. 2). Pernod et al. [8] separately defined major bleeding (overt and associated with decrease in hemoglobin level of > 2 g dL−1, needing transfusion of two or more units of packed red blood cells, or if it was intracranial or retroperitoneal) as compared with minor bleeding (all other bleeding). Two other studies did not report these events separately [7, 10]. For the two studies contributing events [7, 8] (n = 523), supplemental education did not result in a reduction in hemorrhagic events (RR 0.92, 95% CI 0.04–20.56), but there was considerable heterogeneity (I2 = 81%, χ2 = 5.40, P = 0.02).

image

Figure 2. Forest plot of comparison by total patients for outcome: hemorrhagic events [7, 8, 10]. CI, confidence interval; d.f., degrees of freedom; M-H, Mantel-Haenszel.

Download figure to PowerPoint

TTEs

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Three studies [7, 8, 10] (n = 609 randomized) reported TEEs (Fig. 3). For the two studies contributing events [8, 10] (n = 388), supplemental education did not result in a significant reduction in TEEs (RR 0.66, 95% CI 0.10–4.39), with moderate heterogeneity (I2 = 28%, χ2 = 1.38, P = 0.24). Adjudication of these events was adequately described in only one study, which identified symptomatic recurrence of TEEs at 90 days of follow-up [8].

image

Figure 3. Forest plot of comparison by total patients for outcome: thromboembolic events [7, 8, 10]. CI, confidence interval; d.f., degrees of freedom; M-H, Mantel-Haenszel.

Download figure to PowerPoint

Composite endpoint of hemorrhagic events and TEEs

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Three studies [7, 8, 10] (n = 609) contributed events on the composite of hemorrhagic events and TEEs (Fig. 4). There was no significant difference between the supplemental education and usual care groups in the composite endpoint of hemorrhagic events and TEEs, although there was a trend favoring the supplemental education group (RR 0.48, 95% CI 0.23–1.01). Significant heterogeneity was found among the studies (I2 = 70%, χ2 = 6.74, P = 0.03).

image

Figure 4. Forest plot of comparison by total patients for composite outcome: hemorrhagic or thromboembolic events [7, 8, 10]. CI, confidence interval; d.f., degrees of freedom; M-H, Mantel-Haenszel.

Download figure to PowerPoint

TTR

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Four trials [7, 10, 20, 27] that randomized 545 patients reported information on TTR (Fig. 5). Supplemental education did not result in a statistically significant improvement in TTR as compared with usual care (mean absolute difference of 2.02% units; 95% CI − 2.81 to 6.84). There was low heterogeneity among the trials (I2 = 26%, χ2 = 4.08, P = 0.25).

image

Figure 5. Forest plot of comparison by total patients for outcome: time in therapeutic range [7, 10, 20, 27]. CI, confidence interval; d.f., degrees of freedom; IV, inverse variance; SD, standard deviation.

Download figure to PowerPoint

Knowledge testing

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Three studies [8, 11, 28] (n = 664 randomized) reported knowledge testing results, which we did not pool. Two studies reported statistically significant improvement as compared with usual care. The mean absolute difference in post-test scores was 24.5% in Clark and Bayley [11] and a 13.9% improvement in Mazor et al. [28], both at an unspecified follow-up interval. Pernod et al. [8] noted no significant difference in knowledge scores at 3 months of follow-up.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

To our knowledge, this article presents the first systematic review and meta-analysis conducted on the effect of supplemental patient education on outcomes in patients receiving OAT. Strengths of the review include a rigorous systematic approach to searching, data extraction, evaluation of methodologic quality, and combining results. We found no significant benefit associated with supplemental patient education on important clinical events or TTR, and conflicting evidence on knowledge improvement. However, the quality of evidence for each of the outcomes was low.

International anticoagulation guidelines [13] and the United States Joint Commission [29] have suggested that best practice requires patient education on VKAs. Additionally, guidelines from the UK's National Patient Safety Agency advise that healthcare organizations should ensure that patients prescribed VKAs receive appropriate information regarding their safe and effective use [30]. Warfarin has been the most frequently implicated drug leading to emergency hospitalization in the USA, with an estimated 33 000 hospitalizations annually [31]. To prevent morbidity, patients need to learn how to navigate the medication's narrow therapeutic index. Although we found no studies addressing new OATs such as dabigatran or rivaroxaban, patient education will be equally, if not more, important with these medications, given the lack of available monitoring, shorter half-life, challenges in patients with renal impairment, and the lack of an antidote for bleeding. Indeed, dabigatran was the top drug-related cause of serious adverse events, including mortality, reported to the Food and Drug Administration in 2011 [32].

In addition to the lack of evidence of benefit for patient outcomes, supplemental patient education is time-consuming for clinicians, can be overwhelming for patients, and probably has the potential to cause harm if performed poorly. Given the poor description of ‘usual care’ patient education in these studies, a key question now is how much and what type of patient education is optimal.

Limitations of our study relate primarily to the quality and quantity of the available evidence on this topic. Despite the large number of patients taking OATs and the importance of patient education, there are few RCTs that have examined the effects of supplemental patient education. Those people most in need of education and support were often excluded, such as those who could not speak the language or had cognitive impairment. Furthermore, nearly all of our included studies were set in hospitals or anticoagulation clinics, whereas much VKA prescribing takes place in primary care practices, where usual procedures for patient education probably vary widely.

There was considerable heterogeneity in the study methodologies, educational interventions, and populations under evaluation. Given the small sample populations and short follow-up, clinical events were few. In addition, only three of the five trials [7, 8, 20] contributing outcomes for hemorrhagic events, TEEs and TTR calculated an anticipated sample size for their trials, and none was adequately powered for the clinical outcomes. TTR and knowledge test measurements serve only as surrogate markers. Although TTR is a well-recognized surrogate for bleeding or TEE [6], knowledge tests on oral anticoagulation have not been accepted as predicting clinically important outcomes. In addition, length of follow-up ranged from only 3  months to 6 months, and only two studies specifically evaluated anticoagulant-naïve patients [8, 10].

We have found no conclusive evidence that supplemental patient education in patients on OATs improves clinical outcomes or anticoagulation control. Larger RCTs with longer follow-up, with recruitment of patients initiating anticoagulation in primary care settings and using clearly defined education arms are needed.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

We would like to thank librarians D. Giustini of the University of British Columbia and H. Michael of the University of Toronto for their assistance with the literature search, C. Hillis of McMaster University for his assistance in data extraction, and E. Pullenayegum of McMaster University for statistical advice.

Appendix 1: Literature search

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References

Initial search to January 2010

MEDLINE (1948 to January week 4 2010, in-process and other non-indexed citations 5 February 2010, corrections)
  1. exp anticoagulants/(156463)
  2. anticoagula$.mp. (69613)
  3. acenocoumarol.mp. (1088)
  4. dicumarol.mp. (2192)
  5. ethyl biscoumacetate.mp. (269)
  6. phenprocoumon.mp. (13)
  7. warfarin.mp. (15341)
  8. or/1-7 (178231)
  9. exp Education/(504956)
  10. exp Self Care/(30100)
  11. exp Health Knowledge, Attitudes, Practice/(46969)
  12. exp Patient Care Planning/(44071)
  13. exp Patient Care Team/(44180)
  14. education$.mp. (491537)
  15. self-testing$.mp. (261)
  16. self-monitoring$.mp. (4955)
  17. or/9-16 (751072)
  18. and/8,17 (1680)
  19. limit 18 to (English language and humans) (1380)
  20. limit 19 to randomized controlled trial (84)
EMBASE (classic 1947–1979, 1980–2010 week 4)
  1. exp anticoagulant agent/(326241)
  2. anticoagula$.mp. (86873)
  3. acenocoumarol.mp. (3443)
  4. dicumarol.mp. (771)
  5. ethyl biscoumacetate.mp. (746)
  6. phenprocoumon.mp. (9)
  7. warfarin$.mp. (40743)
  8. or/1-7 (341424)
  9. exp patient education/(29676)
  10. exp education/(359127)
  11. exp health education/(89616)
  12. exp attitude to health/(3474)
  13. exp self care/(16586)
  14. exp patient care planning/(816)
  15. education$.mp. (336490)
  16. self-testing$.mp. (259)
  17. exp self evaluation/(10450)
  18. exp self monitoring/(2050)
  19. self monitoring$.mp. (3580)
  20. or/9-19 (443598)
  21. and/8,20 (5857)
  22. limit 21 to (human and English language) (4649)
  23. random$.mp. (494024)
  24. and/22-23 (469)
International pharmaceutical abstracts (1970 to January 2010)
  1. anticoagula$.mp. (8713)
  2. acenocoumarol.mp. (146)
  3. dicumarol.mp. (81)
  4. ethyl biscoumacetate.mp. (19)
  5. phenprocoumon.mp. (0)
  6. warfarin.mp. (3354)
  7. or/1-6 (9077)
  8. education$.mp. (22674)
  9. self monitoring$.mp. (181)
  10. self testing$.mp. (26)
  11. or/8-10 (22809)
  12. and/7,11 (460)
CINAHL

((MH “Anticoagulants+”) or (“anticoagula*”) or (MH “Warfarin”) or (“acenocoumarol”) or (“dicumarol”) or (“ethyl biscoumacetate”) or (“phenprocoumon”)) and ((TX “education*”) or (TX “self-testing*”) or (TX “self-monitoring*”) or (“education*”) or (MH “Education+”) or (MH “Education Research”)) and (“random*”) (88)

Cochrane Central Register of Controlled Trials (CENTRAL)

(anticoagula*) AND (education OR self-testing OR self-monitoring) (117)

Search update to February 2012

MEDLINE (1946 to February week 1 2012, in-process and other non-indexed citations 13 February 2012, corrections)
  1. exp anticoagulants/(170551)
  2. anticoagula$.mp. (77746)
  3. acenocoumarol.mp. (1163)
  4. dicumarol.mp. (2119)
  5. ethyl biscoumacetate.mp. (265)
  6. phenprocoumon.mp. (13)
  7. warfarin.mp. (17619)
  8. or/1-7 (194799)
  9. exp Education/(555390)
  10. exp Self Care/(34221)
  11. exp Health Knowledge, Attitudes, Practice/(58330)
  12. exp Patient Care Planning/(47464)
  13. exp Patient Care Team/(48943)
  14. education$.mp. (553516)
  15. self-testing$.mp. (332)
  16. self-monitoring$.mp. (5946)
  17. or/9-16 (845952)
  18. and/8,17 (2002)
  19. limit 18 to (English language and humans) (1645)
  20. limit 19 to randomized controlled trial (99)
  21. limit 20 to publication date to 2010 to current (15)
EMBASE (classic 1947–1979, 1980–2012 week 6)
  1. exp anticoagulant agent/(440335)
  2. anticoagula$.mp. (129920)
  3. acenocoumarol.mp. (4359)
  4. dicumarol.mp. (1077)
  5. ethyl biscoumacetate.mp. (953)
  6. phenprocoumon.mp. (12)
  7. warfarin$.mp. (54022)
  8. or/1-7 (464253)
  9. exp patient education/(76134)
  10. exp education/(869707)
  11. exp health education/(204827)
  12. exp attitude to health/(66291)
  13. exp self care/(40454)
  14. exp patient care planning/(24728)
  15. education$.mp. (803218)
  16. self-testing$.mp. (445)
  17. exp self evaluation/(16784)
  18. exp self monitoring/(2923)
  19. self monitoring$.mp. (5681)
  20. or/9-19 (1097615)
  21. and/8,20 (9749)
  22. limit 21 to (human and English language) (6739)
  23. random$.mp. (807929)
  24. and/22, 23 (679)
  25. limit 24 to publication date to 2010 to current (173)
International pharmaceutical abstracts (1970 to January 2012)
  1. anticoagula$.mp. (9714)
  2. acenocoumarol.mp.(151)
  3. dicumarol.mp. (81)
  4. ethyl biscoumacetate.mp. (19)
  5. phenprocoumon.mp. (0)
  6. warfarin.mp. (3765)
  7. or/1-6 (10104)
  8. education$.mp. (23678)
  9. self monitoring$.mp. (200)
  10. self testing$.mp. (28)
  11. or/8-10 (23831)
  12. and/7,11 (527)
  13. limit 12 to publication date to 2010 to current (65)
CINAHL

((MH “Anticoagulants+”) or (“anticoagula*”) or (MH “Warfarin”) or (“acenocoumarol”) or (“dicumarol”) or (“ethyl biscoumacetate”) or (“phenprocoumon”)) and ((TX “education*”) or (TX “self-testing*”) or (TX “self-monitoring*”) or (“education*”) or (MH “Education+”) or (MH “Education Research”)) and (“random*”) AND limit publication date to 2010 to current. (18)

Cochrane Central Register of Controlled Trials (CENTRAL)

(anticoagula*) AND (education OR self-testing OR self-monitoring) AND limit publication date to 2010 to current. (10)

New anticoagulants MEDLINE (1946 to February week 1 2012, in-process and other non-indexed citations 13 February 2012, corrections)
  1. dabigatran.mp (721)
  2. rivaroxaban.mp (552)
  3. apixaban.mp (286)
  4. or/1-3 (1105)
  5. exp Education/(568780)
  6. exp Self Care/(35399)
  7. exp Health Knowledge, Attitudes, Practice/(60866)
  8. exp Patient Care Planning/(48333)
  9. exp Patient Care Team/(49990)
  10. education$.mp. (568735)
  11. self-testing$.mp. (343)
  12. self-monitoring$.mp. (6211)
  13. or/5-12 (869413)
  14. and/4,13 (14)
  15. limit 14 to (English language and humans) (7)
  16. limit 15 to randomized controlled trial (0)
New anticoagulants EMBASE (classic 1947–1979, 1980–2012 week 6)
  1. dabigatran.mp (2314)
  2. rivaroxaban.mp (1893)
  3. apixaban.mp (1067)
  4. or/1-3 (3223)
  5. exp patient education/(79076)
  6. exp education/(899574)
  7. exp health education/(212096)
  8. exp attitude to health/(71163)
  9. exp self care/(42273)
  10. exp patient care planning/(25428)
  11. education$.mp. (834537)
  12. self-testing$.mp. (479)
  13. exp self evaluation/(17511)
  14. exp self monitoring/(3054)
  15. self monitoring$.mp. (6084)
  16. or/5-15 (1144206)
  17. and/4,16 (130)
  18. limit 17 to (human and English language) (94)
  19. random$.mp. (868727)
  20. and/18,19 (8)
New anticoagulants international pharmaceutical abstracts (1970 to January 2012)
  1. dabigatran.mp (95)
  2. rivaroxaban.mp (98)
  3. apixaban.mp (51)
  4. or/1-3 (181)
  5. education$.mp. (23905)
  6. self monitoring$.mp. (209)
  7. self testing$.mp. (29)
  8. or/5-7 (24066)
  9. and/4,8 (3)
New anticoagulants CINAHL

((“dabigatran”) or (“rivaroxaban”) or (“apixaban”)) and ((TX “education*”) or (TX “self-testing*”) or (TX “self-monitoring*”) or (“education*”) or (MH “Education+”) or (MH “Education Research”)) and (“random*”). (0)

New anticoagulants Cochrane Central Register of Controlled Trials (CENTRAL)

(dabigatran OR rivaroxaban OR apixaban) AND (education OR self-testing OR self-monitoring) (0)

Appendix 2: GRADE evidence profile table

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References
Quality assessmentNo. of patientsEffectQualityImportance
No. of studiesDesignRisk of biasInconsistencyIndirectnessImprecisionOther considerationsComparisons by total patientsControlRelative (95% CI)Absolute
  1. CI, confidence interval; MD, mean difference; RR, relative risk. *Patients could not be blinded to intervention. In some cases, follow-up physicians were not blinded to allocation. †Settings varied between studies, including inpatient wards, outpatient anticoagulation clinics staffed by specialists, and primary care offices. Also, educational interventions varied in the number of patients per group, length of education, and content of education. ‘Usual care’ varied among studies. ‡Imprecision may be attributable to very few events for hemorrhagic and thromboembolic outcomes. In addition, these were often secondary outcomes. §Major and minor bleeding as reported in studies were combined for the meta-analysis. ¶Time in therapeutic range is a surrogate for complications, such as bleeding, thromboembolic events, or death.

Major and minor bleeding (follow-up mean 4.2 months)
3Randomized trialsSerious*SeriousNo serious indirectnessSeriousNone5/220 (2.3%)13/303 (4.3%)RR 0.92 (0.04–20.56)3 fewer per 1000 (from 41 fewer to 839 more)⊕ very lowCritical§
Thromboembolic event (follow-up mean 4.2 months)
3Randomized trialsSerious*SeriousNo serious indirectnessSeriousNone3/203 (1.5%)5/185 (2.7%)RR 0.66 (0.10–4.39)9 fewer per 1000 (from 24 fewer to 92 more)⊕⊕ very lowCritical
Hemorrhagic or thromboembolic events (follow-up median 4.2 months)
3Randomized trialsSerious*SeriousNo serious indirectnessSeriousNone8/263 (3%)18/346 (5.2%)RR 0.48 (0.23–1.01)27 fewer per 1000 (from 40 fewer to 1 more)⊕⊕ very lowCritical
Time in therapeutic range (follow-up mean 4.2 months; better indicated by lower values)
4Randomized trialsSerious*SeriousSeriousNo serious imprecisionNone216317Not applicableMD 2.02 higher (2.81 lower to 6.84 higher)⊕ very lowImportant

Appendix 3: GRADE summary of findings table

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References
Comparisons by total patients for patients on anticoagulation

Patient or population: patients with patients on anticoagulation

Settings:

Intervention: comparisons by total patients

OutcomesIllustrative comparative risks* (95% CI)Relative effect (95% CI)No of participants (studies)Quality of the evidence (GRADE)
Assumed risk Control Corresponding risk Comparisons by total patients
  1. CI, confidence interval; RR, risk ratio. GRADE Working Group grades of evidence: high quality – further research is very unlikely to change our confidence in the estimate of effect; moderate quality – further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate; low quality – further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate; very low quality – we are very uncertain about the estimate. *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). †Patients could not be blinded to intervention. In some cases, follow-up physicians were not blinded to allocation. ‡Settings varied between studies, including inpatient wards, outpatient anticoagulation clinics staffed by specialists, and primary care offices. Also, educational interventions varied in the number of patients per group, length of education, and content of education. ‘Usual care’ varied among studies. §Imprecision may be attributable to very few events for hemorrhagic and thromboembolic outcomes. In addition, these were often secondary outcomes. ¶TTR is a surrogate for complications, such as bleeding, thromboembolic events, or death.

Major and minor bleeding

Follow-up: mean 4.2 months

43 per 100039 per 1000 (2–882)RR 0.92 (0.04–20.56)523 (3 studies)⊕⊝⊝⊝ very low,,§

Thromboembolic event

Follow-up: mean 4.2 months

27 per 100018 per 1000 (3–119)RR 0.66 (0.10–4.39)388 (3 studies)⊕⊝⊝⊝ very low,,§

Hemorrhagic or thromboembolic events

Follow-up: median 4.2 months

52 per 100025 per 1000 (12–53)RR 0.48 (0.23–1.01)609 (3 studies)⊕⊝⊝⊝ very low,,§

Time in therapeutic range

Follow-up: mean 4.2 months

 The mean time in therapeutic range in the intervention groups was 2.02-fold higher (2.81-fold lower to 6.84-fold higher) 533 (4 studies)⊕⊝⊝⊝ very low,,

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Quality of evidence and outcomes
  7. Hemorrhagic events
  8. TTEs
  9. Composite endpoint of hemorrhagic events and TEEs
  10. TTR
  11. Knowledge testing
  12. Discussion
  13. Acknowledgements
  14. Disclosure of Conflict of Interests
  15. Appendix 1: Literature search
  16. Appendix 2: GRADE evidence profile table
  17. Appendix 3: GRADE summary of findings table
  18. References
  • 1
    You JJ, Singer DE, Howard PA, Lane DA, Eckman MH, Fang MC, Hylek EM, Schulman S, Go AS, Hughes M, Spencer FA, Manning WJ, Halperin JL, Lip GYH. Antithrombotic therapy for atrial fibrillation: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141 (2 Suppl): e531Se575S.
  • 2
    Wells G, Coyle D, Cameron C, Steiner S, Coyle K, Kelly S, Tang A, Healey J, Hsieh S-C, Berkom JV. Safety, effectiveness, and cost-effectiveness of new oral anticoagulants compared with warfarin in preventing stroke and other cardiovascular events in patients with atrial fibrillation. Canadian Agency for Drugs and Technologies in Health (CADTH): Therapeutic Review 2012; Available: http://cadth.ca/media/pdf/NOAC_Therapeutic_Review_final_report.pdf . Accessed 28 January 2013.
  • 3
    Whitlock RP, Sun JC, Fremes SE, Rubens FD, Teoh KH. Antithrombotic and thrombolytic therapy for valvular disease: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141 (2 Suppl): e576S600S.
  • 4
    Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, Nelson ME, Wells PS, Gould MK, Dentali F, Crowther M, Kahn SR. Antithrombotic therapy for VTE disease: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141 (2 Suppl): e419S94S.
  • 5
    Fox BD, Kahn SR, Langleben D, Eisenberg MJ, Shimony A. Efficacy and safety of novel oral anticoagulants for treatment of acute venous thromboembolism: direct and adjusted indirect meta-analysis of randomised controlled trials. BMJ 2012 Nov 13; 345: e7498.
  • 6
    Ageno W, Gallus AS, Wittkowsky A, Crowther M, Hylek EM, Palareti G. Oral anticoagulant therapy: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141: e44S88S.
  • 7
    Gadisseur APA, Breukink-Engbers WGM, Van Der Meer FJM, Van Den Besselaar AMH, Sturk A, Rosendaal FR. Comparison of the quality of oral anticoagulant therapy through patient self-management and management by specialized anticoagulation clinics in the Netherlands: a randomized clinical trial. Arch Intern Med 2003; 163: 263946.
  • 8
    Pernod G, Labarère J, Yver J, Satger B, Allenet B, Berremili T, Fontaine M, Franco G, Bosson JL. EDUC'AVK: reduction of oral anticoagulant-related adverse events after patient education: a prospective multicenter open randomized study. J Gen Intern Med 2008; 23 (9): 14416.
  • 9
    Léger S, Allenet B, Pichot O, Figari G, Calop J, Carpentier P, Bosson JL. Impact d'un programme d'éducation thérapeutique sur les attitudes de prévention vis-à-vis du risque iatrogène: étude pilote contrôlée visant les patients sous anticoagulants oraux pour maladie thromboembolique veineuse. J Mal Vasc 2004; 29: 1528.
  • 10
    Laporte S, Quenet S, Buchmuller-Cordier A, Reynaud J, Tardy-Poncet B, Thirion C, Decousus H, Mismetti P. Compliance and stability of INR of two oral anticoagulants with different half-lives: a randomised trial. Thromb Haemost 2003; 89: 45867.
  • 11
    Clark CM, Bayley EW. Evaluation of the use of programmed instruction for patients maintained on warfarin therapy. Am J Public Health 1972; 62: 11359.
  • 12
    Gadisseur APA, Kaptein AA, Breukink-Engbers WGM, Van Der Meer FJM, Rosendaal FR. Patient self-management of oral anticoagulant care vs. management by specialized anticoagulation clinics: positive effects on quality of life. J Thromb Haemost 2004; 2: 58491.
  • 13
    Holbrook A, Schulman S, Witt DM, Vandvik PO, Fish J, Kovacs MJ, Svensson PJ, Veenstra DL, Crowther M, Guyatt GH. Evidence-based management of anticoagulant therapy: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141: e152S84S.
  • 14
    Palareti G, Legnani C, Guazzaloca G, Lelia V, Cosmi B, Lunghi B, Marchetti G, Poli D, Pengo V, the ad hoc study group of the Italian Federation of Anticoagulation C. Risks factors for highly unstable response to oral anticoagulation: a case–control study. Br J Haematol 2005; 129: 728.
  • 15
    Tang EOYL, Lai CSM, Lee KKC, Wong RSM, Cheng G, Chan TYK. Relationship between patients' warfarin knowledge and anticoagulation control. Ann Pharmacother 2003; 37: 349.
  • 16
    Kagansky N, Knobler H, Rimon E, Ozer Z, Levy S. Safety of anticoagulation therapy in well-informed older patients. Arch Intern Med 2004; 164: 204450.
  • 17
    Wofford J, Wells M, Singh S. Best strategies for patient education about anticoagulation with warfarin: a systematic review. BMC Health Serv Res 2008; 8: 407.
  • 18
    Samsa GP, Matchar DB. Relationship between test frequency and outcomes of anticoagulation: a literature review and commentary with implications for the design of randomized trials of patient self-management. J Thromb Thrombolysis 2000; 9: 28392.
  • 19
    Group CCaCR. Data extraction template for Cochrane reviews. Cochrane Collaboration, 2009; 1.3.0.
  • 20
    Machtinger EL, Wang F, Chen L-L, Rodriguez M, Wu S, Schillinger D. A visual medication schedule to improve anticoagulation control: a randomized, controlled trial. Jt Comm J Qual Patient Saf 2007; 33: 62535.
  • 21
    Rosendaal FR, Cannegieter SC, van der Meer FJ, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993; 69: 2369.
  • 22
    Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2. The Cochrane Collaboration, 2009; www.cochrane-handbook.org. Accessed 5 June 2009.
  • 23
    Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ. What is ‘quality of evidence’ and why is it important to clinicians? BMJ 2008; 336: 9958.
  • 24
    Brozek J, Oxman A, Schünemann H. GRADEpro [Computer program]. Version 3.2 for Windows. 2008.
  • 25
    Review Manager (RevMan) [Computer Program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
  • 26
    Moher D, Liberati A, Tetzlaff J, Altman D. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097.
  • 27
    Khan TI, Kamali F, Kesteven P, Avery P, Wynne H. The value of education and self-monitoring in the management of warfarin therapy in older patients with unstable control of anticoagulation. Br J Haematol 2004; 126: 55764.
  • 28
    Mazor KM, Baril J, Dugan E, Spencer F, Burgwinkle P, Gurwitz JH. Patient education about anticoagulant medication: is narrative evidence or statistical evidence more effective? Patient Educ Couns 2007; 69: 14557.
  • 29
    The Joint Commission US. National patient safety goals. 2012, http://www.jointcommission.org/assets/1/6/NPSG_Chapter_Jan2012_HAP.pdf. Accessed 20 February 2012.
  • 30
    Cousins D. Risk assessment of anticoagulation therapy. National Patient Safety Agency, 2006, http://www.npsa.nhs.uk/patientsafety/alerts-and-directives/alerts/anticoagulant. Accessed 20 February 2012.
  • 31
    Budnitz DS, Lovegrove MC, Shehab N, Richards CL. Emergency hospitalizations for adverse drug events in older Americans. N Engl J Med 2011; 365: 200212.
  • 32
    Anticoagulants the Leading Reported Drug Risk in 2011. Institute for Safe Medication Practices, 2011 Quarters 3-4. QuarterWatch: Monitoring FDA MedWatch Reports. Available: www.ismp.org/quarterwatch/pdfs/2011Q4.pdf. Accessed 28 January 2013.