Challenges to Achieving Optimal Therapeutic Drug Monitoring in the LTC Setting

Therapeutic drug monitoring (TDM) provides valuable data about the safety and efficacy of drug therapy. A lack of TDM has been associated with a significant proportion of preventable adverse drug events among elderly patients. A number of barriers to the performance of optimal monitoring exist in the long-term care (LTC) setting. When faced with a lack of desired TDM, the clinician must attribute this to one of three scenarios: monitoring is feasible but has not been performed; monitoring is only feasible with the use of creative strategies; or monitoring is not feasible, and decisions about whether to continue drug therapy must be considered. This article describes common barriers to optimal monitoring using specific drug therapy examples and discusses options for overcoming these barriers. The pharmacist’s role in TDM in the LTC setting is also highlighted. (Annals of Long-Term Care: Clinical Care and Aging 2010;18[10]:33-38).
____________________________________________________________________________________________________________________

Introduction

Therapeutic drug monitoring (TDM) begins when a medication is first prescribed and involves determining an appropriate initial and maintenance dosage regimen for the patient.¹ There are generally two sources that provide guidance to the clinician about optimal TDM. The first is medication-specific and comes from published product data. The second is environment-specific and derived from long-term care (LTC) regulations and guidelines for drug regimen review. Successful medication monitoring enables team members to: (1) track progress toward therapeutic end points and ensure maintenance once achieved; and (2) detect the emergence or presence of adverse drug events (ADEs). Unfortunately, in the LTC setting, medication monitoring presents a challenge. A variety of considerations either pose barriers to the performance of monitoring activities or affect the interpretation of the findings. This article discusses some of the most commonly encountered challenges in the LTC setting and suggests some strategies for overcoming them.

Consequences Associated with Lack of Monitoring

Consequences of inadequate monitoring can include emergence of ADEs as a result of the failure to detect problems associated with the drug therapy. Prevalence rates of ADEs associated with lack of monitoring vary based on the clinical setting. One study in the LTC setting suggested that 80% of ADEs occurring among LTC residents occurred at the monitoring stage,² and the literature has shown that a high proportion of monitoring-related ADEs result in hospitalization or are found to be preventable.³ Consequences may also include therapeutic failure, a term employed when drug therapy does not achieve the therapeutic goal for which it was prescribed.

Therapeutic failures can result in unnecessary exposure to ineffective drug therapy or exacerbation of the underlying condition. Data describing the prevalence and health outcomes of therapeutic failure are not LTC-specific, but studies suggest that 6.8% to 28% of emergency department or hospital admissions associated with ADEs are due to therapeutic failure.^4-6 Ideally, optimal practice of TDM should be able to detect, or prevent, these scenarios by identifying new symptoms suggestive of adverse effects, identifying under- or overadherence, or evaluating the therapeutic response of the underlying condition. Lack of ability to do this has consequences for the provider as well, because failure to supervise or monitor care has been ranked as the third most common reason for a malpractice claim to be brought against a physician.⁷

Barriers to Appropriate Monitoring

The most common barriers to the performance of monitoring activity can be grouped into three categories: logistical; provider-related; and patient-related. In addition, variables that can affect normative values or clinical interpretation of results include age-related changes in physiology, atypical clinical presentation, and comorbid conditions and associated medications. Examples of logistical barriers include time constraints, access/transportation availability, or financial limitations. For instance, an offsite clinical laboratory may be unable to ensure correct timing of a blood draw to evaluate a trough serum concentration if there is not optimal coordination with the nurse who administers the dose. Provider- and patient-related barriers can be more complex.

At the clinician level, there is often disagreement about the appropriateness and optimal frequency of medication monitoring among frail patients. Not all disease states have published guidelines establishing standard goals of therapy and optimal monitoring procedures; among those that do, the literature is often representative of a relatively younger subgroup. It often is not clear how to extrapolate the evidence to a frail population. With respect to frequency, the balance between adequate versus excess monitoring must also be weighed. Excessive monitoring is problematic, and a balance must be met between the frequency that is necessary to ensure safety and efficacy and that which is overzealous, perhaps driven by fear of liability. In evaluations of monitoring tests requested by consultant pharmacists, the resulting data prompted drug therapy adjustment in 41% to 87.5% of cases when requested on the basis of laboratory data older than 6 months or concerns over drug toxicity or drug-drug interactions.^8-10

By contrast, tests performed in excess of the federal indicators (eg, monthly or bimonthly assessments) yielded new information in less than 5% of the repeated assessments and contributed significant excess cost.¹⁰ Although there is controversy about the utility of routine laboratory screening tests, studies suggest that obtaining laboratory data to detect ADEs is useful in the LTC setting.² The consultant pharmacist can assist both the provider and nursing staff by providing advice about which monitoring parameters are prudent to evaluate potential or suspected ADEs. This is especially important following drug therapy changes. Addition of new medication, switches between agents, adjustments in dosage, or withdrawal of medication can result in many unintended effects via drug-drug interaction, drug-disease interaction, or compounding of therapeutic effects. The choice to perform such monitoring in a preemptive or reactive manner may be scenario-specific. For instance, reevaluation of a chemistry panel to assess potassium and hydration status within 30 days of adjustment in the dose of a loop diuretic may be considered more “routine,” whereas reevaluation of the same test to assess sodium might only be ordered after the initiation of a serotonin reuptake inhibitor if there is concern of new-onset dizziness or falls. For this reason, it is important that the consultant pharmacist be a part of the team; however, a barrier to this approach can present if the pharmacist does not have a reliable mechanism to be notified of resident change of status.

In many cases, the consultant pharmacist’s functions are not affiliated with the pharmacy vendor supplying drug product; therefore, the only mechanism to identify recent changes in medication orders is the drug regimen review, which is a retrospective activity. A collaborative approach with the team is needed to involve the pharmacist when problems arise in between these reviews. At the patient level, one of the primary factors to be considered is patient choice. An individual may have established specific directives that preclude certain monitoring activities, and these must be clearly communicated to all members of the healthcare team.

The individual’s consent and ability to participate and cooperate with the monitoring activities is also crucial to performing TDM. Difficulty following directions or participating in an interview due to language barriers; visual, auditory, or cognitive deficits; inability to express distress or articulate symptoms; or agitation associated with monitoring procedures are just a few examples of this. Even obtaining height or periodic body weight measurements can be challenging, and creative strategies may be needed to obtain or approximate needed data. A complete discussion of age-related changes in physiology that affect clinical presentation or alter pharmacokinetics and pharmacodynamics is beyond the scope of this article. However, a few salient examples of how these changes may affect normative data or clinical interpretation of monitoring might include alterations in febrile response, reduced creatinine clearance, thyroid gland responsiveness to thyroid-stimulating hormone, or free fraction of protein-bound drugs (eg, phenytoin) in hypoalbuminemia.

Overcoming Lack of Monitoring

Three basic scenarios should be considered when assessing why proper monitoring is not being accomplished: (1) monitoring is feasible, but is not being performed; (2) monitoring is not feasible without using creative strategies; and (3) monitoring is not feasible, and the risks versus benefits of continued therapy must be assessed. The optimal intervention differs significantly depending upon which of these categories is identified, and a few examples follow.

Scenario 1
When monitoring can be feasibly accomplished to ensure a complete medical chart for the patient, but isn’t being performed, several strategies can be implemented. These involve education, coordination, and recommendation. Education makes providers and facilities aware of the monitoring that is needed to both optimize safety and efficacy of prescribed therapies and ensure that state regulations are being followed.

The consultant pharmacist for the LTC facility can educate the staff on the monitoring parameters recommended for individual medications and on the LTC facility regulations governing medication monitoring. Coordination builds on the concepts of education and incorporates communication, records sharing, and alignment of care activities with the necessary TDM. Communication ensures that the attending physician has conveyed his or her intended plan of care and rationale. Facility staff are in the best position to adhere to the desired monitoring plan, and, likewise, a consultant pharmacist or senior nurse manager is optimally empowered to audit adherence when a clear directive for monitoring is communicated. This is especially important if the decision is made to be more or less aggressive than the published monitoring recommendations suggest. Examples of records sharing include ensuring that the complete records are provided in each discharge summary upon transfer from another care venue, and the sharing of records of encounters outside of the facility.

This not only ensures that the facility maintains the appropriate evidence of ongoing monitoring, but also avoids unnecessary duplication of monitoring activity at the facility level. Examples of alignment of care activities with TDM include considering the timing of vitals assessment in relation to the timing of medication administration, performing assessment scales when a resident is awake and alert, and coordinating medication administration around the timing of a laboratory draw. The monthly records review performed by the consultant pharmacist is an important mechanism for identifying this type of lack of monitoring. This records review can also evaluate whether the problem is simply a lack of evidence that monitoring has been performed, or if there is a need for clarification regarding what the intended directives for monitoring are supposed to be. If patterns of lack of monitoring are identified facility-wide, overarching interventions can include standard order sets put in place to make desired monitoring automatic, such as scheduled vitals with each prescription of a cardiovascular medication.

Pre-set orders communicate a clear expectation, making it easier to prospectively adhere and retrospectively evaluate the monitoring activity. Electronic reminder systems have been a subject of interest to proactively enhance medication monitoring in light of evidence suggesting that lack of monitoring may be a key area to decrease preventable ADEs. Some of these have been targeted at the physician or midlevel practitioner and include computer order entry with clinical decision making. Most of the studies evaluating these systems have not been conducted in LTC settings and have included small numbers of patients or noncommercially available systems.^11,12 However, a significant reduction in ADEs has been reported in the literature within both intensive care units and general hospital settings using these tools.¹¹

The few studies within the LTC setting have produced mixed results. Judge and colleagues found that prescribers with medication monitoring systems in place were slightly more likely to take an appropriate action when they received an alert compared with providers who did not receive alerts (relative risk, 1.1; 95% confidence interval, 1.0-1.22).¹³ Warnings that generated the greatest response among providers were those related to warfarin or medications that increase the risk of central nervous system effects. Gurwitz and colleagues, who evaluated the use of computer order entry with clinical decision making in more than 1100 patients residing within two LTC facilities, did not find a significant reduction in ADEs.¹⁴ The risk for nonsignificant alerts and the potential for this to decrease prescriber sensitivity to clinically significant alerts have also been noted as a challenge with these systems.^13-15

The use of electronic systems may have an important role in the future, because a lack of monitoring has been demonstrated to contribute to preventable ADEs. While most data describing outcomes with such systems are from other settings,¹⁶ they can be applied to the LTC setting; however, there is a need to tailor existing computer order entry and clinical decision support systems to reflect the unique needs associated with prescribing in the LTC setting, as well as methods to ensure that implementation is cost-effective.^14,17

Scenario 2
Some barriers to proper monitoring require creative strategies to overcome. Sometimes, when a drug’s required monitoring activities impede optimal use, the easiest option is to switch therapy, but this is not always optimal. By contrast, there may be a decision to “loosen” acceptable therapeutic goals, but if this is not consistent with a patient’s desires for healthcare outcomes, this may be a poor tradeoff for convenience. There also may be temptation to simply discontinue monitoring orders, but continuation of unmonitored drug therapy is not ideal.

This is especially true for monitoring intensive disease states or medications where a lack of TDM is potentially dangerous. Diabetes mellitus, especially when insulin is the primary therapy, and oral anticoagulation with warfarin are two good examples where out-of-the-box thinking can have a significant impact on proper monitoring. Diabetes mellitus can be monitoring-intensive, with fingerstick monitoring required up to four times daily. This can help ensure that patients are not becoming hypoglycemic throughout the day and facilitates the adjustment of insulin doses should there be consistent excursions of glycemic control. However, the intensity of this monitoring regimen can be reevaluated, particularly for individuals with a stable history of control. One solution could involve decreasing the total number of fingersticks over the period of a week, yet alternating days and times, with the understanding that over the course of a few days, all necessary time periods would have glucose values represented. Other options may involve beginning with more intensive four-times-daily monitoring and gradually reducing the frequency for those who are able to maintain fairly stable control.¹⁸

The optimal monitoring frequency for patients with type 2 diabetes must be sufficient to facilitate reaching glucose goals.¹⁹ Many times, the biggest barrier to monitoring blood glucose can occur at the patient level, either from the pain associated with fingerstick or agitation or refusal to cooperate secondary to cognitive impairment. A topical lidocaine product can be applied to the finger before each fingerstick. Comparison of blood glucose data with and without the application of lidocaine can be maintained in the facility’s glucometer control log if there is concern about interference with a particular brand of test strip or machine. Alternatively, a different body site for testing may also decrease pain.

Attempts by different staff members or deviating from the typical monitoring schedule to maximize times of the day when the resident is most receptive may be necessary. Further, while not ideal for monitoring blood glucose levels, using a urine dipstick test can at least monitor patients to ensure avoidance of overt hyperglycemia. Periodic assessment of glycated hemoglobin A1c (HbA1c) levels can provide general information about glycemic control, but this test alone is not an optimal surrogate for blood glucose assessment when monitoring safety or efficacy of therapy. There is disagreement over what the optimal target HbA1c value should be among frail patients, and the accuracy of the clinical interpretation of HbA1c may be compromised among individuals with concomitant renal dysfunction or anemia.^20-23

With respect to anticoagulation, the frequency of monitoring required with warfarin use and the susceptibility of the elderly patient to ADEs make anticoagulant therapy particularly challenging among frail patients. The first step in assessing the optimal monitoring strategy entails a records review to determine history of control. This is suggested with an understanding that lab data, current and past medication orders, and clinical notations about course of therapy are all traditionally maintained in separate places in the clinical record. By converging these data into a single “anticoagulation history,” patterns may become more apparent, enabling decisions about frequency of follow-up to be tailored to an individual resident. There may be a temptation among clinicians to employ a lower-intensity international normalized ratio (INR) range; however, the evidence for such recommendations is weak.²⁴

Once the basic plan for frequency of follow-up in both maintenance and adjustment phases has been determined, the next step is to consider the venue and method of monitoring (eg, facility vs clinic-based, venipuncture vs fingerstick). In one study of anticoagulant therapy managed by telephone versus office visit, the investigators found no statistically significant difference in the number of INRs within goal range (68.3% via telephone vs 70.6% via office visit) or in rates of thromboembolism or hemorrhage between the two groups (1.42% vs 2.81% and 5.67% vs 5.62% per patient-year, respectively).²⁵

Telephone consultations also are not reimbursable by Medicare, which may prevent some providers from using this method. One alternative that is possible in many states is to establish a pharmacist-managed anticoagulation service via a collaborative drug therapy management agreement with a supervising physician. Evaluations of pharmacist-managed anticoagulation in a variety of settings have demonstrated equivalent or superior therapy outcomes as compared with standard care in a number of clinical settings.^25-39 Implementation in a LTC environment requires modification of traditional pharmacy and consultant service models, and additional demonstration projects evaluating feasibility, satisfaction, costs, and outcomes are specifically needed in the LTC setting. Given a recent estimate that up to half of all anticoagulation ADEs could be avoided with better INR control, such initiatives are compelling.⁴⁰

Another alternative involves point of care (POC) INR monitoring devices, which may be preferable to venipuncture for some residents because they provide immediate results, allowing timely adjustments of warfarin therapy.⁴¹ With some instruments there is concern of positive bias with higher INR values.⁴¹ However, a newer version of the CoaguChek® system, CoaguChek XS, which uses an electrochemical clot detection method with an international sensitivity index (ISI) of 1.01, was found to have only an average of 0.22 INR units above the venous INR result.⁴² Franke and associates found that while the use of POC devices in a family practice setting did not improve the percentage of INRs within goal range, combination with a standardized dosing protocol did improve the proportion of INR values within the goal range from 30.8% to 45.9% (P < .5).⁴³

The use of POC devices in the LTC setting is confounded by cost (approximately $1200 for the device and 48 test strips), which is currently not Medicare-reimbursable for those professionals who are not recognized Medicare providers.⁴⁴ In addition, a recent FDA advisory warns that careful cleaning of the machine is necessary when POC devices are used for more than one individual. Availability of trained professional staff to perform the test, interpret the results, and make the necessary changes in drug therapy may also be a limiting factor. Diabetes and anticoagulation management are but two scenarios where alternate monitoring strategies may be needed.

Regardless of the disease state or drug therapy involved, timing and frequency are, in most cases, the two easiest variables to modify when facing a barrier to monitoring. Some monitoring is preferable to no monitoring, and when the concession for reduced monitoring must be made, this should be carefully documented. Ideally, this is done prospectively to demonstrate that the plan of care has been purposefully negotiated and all variables considered, thereby avoiding the interpretation that lack of optimal monitoring is due to negligent supervision of care.

Scenario 3
In some cases, despite caregivers’ best efforts to employ creative strategies, the desired monitoring simply cannot be performed. In these situations, the decision must be made whether to continue drug therapy without monitoring, or to discontinue drug therapy. There are obvious risks associated with continuing drug therapy without performing the recommended monitoring; however, this could be justified in certain circumstances, particularly if the rationale is carefully documented. This may be especially true for those medications or disease states for which there is more than one type of monitoring activity that can be performed. Consider, for instance, a scenario in which a LTC resident or his or her responsible party elects to discontinue all monitoring involving venipuncture. The resident’s medications include levothyroxine at a dose that has been stable for many years.

While it is understood that history of stability may not necessarily predict changes in metabolic needs if the resident’s status declines in the future, this is a scenario where it may be justified to forgo laboratory monitoring in favor of other clinical monitoring that can provide surrogate indicators of thyroid function (eg, heart rate, appetite, sleep, behavior). The record should clearly document the rationale behind the lack of laboratory assessment. By contrast, consider the same resident for which laboratory monitoring is suspended, but the drug in question is an HMG CoA reductase inhibitor such as pravastatin or atorvastatin.

Aside from major efficacy or safety outcomes, such as a cardiovascular event or rhabdomyolisis, few routine surrogate monitoring activities can be substituted for laboratory assessment of lipid parameters or hepatic enzymes. Without such it is difficult to assess what outcome drug therapy is achieving. Clinical judgment about whether therapy should continue requires two considerations: safety and efficacy. For drugs that have either a narrow therapeutic index or limited parameters through which to assess side effects, safety concerns may prompt the choice to discontinue the drug. For drugs that are generally better tolerated, clinical monitoring for common adverse reactions can be documented, and if there is an established history that the drug is well tolerated by the patient, acute safety concerns may be minimal.

From an efficacy standpoint, one must consider prognosis. Drugs for which efficacy cannot be evaluated without monitoring—especially when the therapeutic goal is incompatible with prognosis—are of dubious utility. However, drugs that are intended to provide symptomatic relief, or for which the efficacy can be indirectly assessed via surrogate parameters, might justifiably continue even if the optimal monitoring parameters cannot be performed. In either case, careful documentation is needed to demonstrate that prudent consideration of variables went into determining the plan of care.

Conclusions

Many challenges to performing optimal TDM in the LTC setting exist, but the prevalence of adverse consequences associated with lack of monitoring should compel the clinician to identify creative strategies to ensure the safety and efficacy of drug therapy. When faced with a lack of TDM, the first step is to categorize the problem into one of three scenarios: (1) monitoring is feasible, but is not being done or documented; (2) monitoring is only feasible with creative strategies; or (3) monitoring is not feasible and considerations about whether to continue drug therapy must be weighed. All members of the healthcare team play a role in the direction and performance of TDM. Pharmacists play an important role in the evaluation of TDM, and novel practice models of collaborative drug therapy management may further enhance this in the future.

The authors report no relevant financial relationships.

Dr. Sleeper is Associate Professor and Division Head, Lubbock Campus; Dr. Blaszczyk is Assistant Professor, Dallas Campus; Drs. Capper and Fox are Assistant Professors, Abilene Campus; and Dr. Treadway is Associate Professor, Dallas Campus, Texas Tech University Health Sciences Center, School of Pharmacy, Geriatrics Division.

References

1. Kang JS, Lee MH. Overview of therapeutic drug monitoring. Korean J Intern Med 2009;24(1):1-10.

2. Gurwitz JH, Field TS, Judge J, et al. The incidence of adverse drug events in two large academic long-term care facilities. Am J Med 2005;118(3):251-258.

3. Thomsen LA, Winterstein AG, Søndergaard B, et al. Systematic review of the incidence and characteristics of preventable adverse drug events in ambulatory care. Ann Pharmacother 2007;41(9):1411-1426.

4. Yee JL, Hasson NK, Schreiber DH. Drug-related emergency department visits in an elderly veteran population. Ann Pharmacother 2005;39(12):1990-1995.

5. Kaiser RM, Schmader KE, Pieper CF, et al. Therapeutic failure-related hospitalisations in the frail elderly. Drugs Aging 2006;23(7):579-586.

6. Franceschi A, Tuccori M, Bocci G, et al. Drug therapeutic failures in emergency department patients. A university hospital experience. Pharmacol Res 2004;49(1):85-91.

7. Luce JM. Medical malpractice and the chest physician. Chest 2008;134(5):1044-1050.

8. Handler SM, Shirts BH, Perera S, et al. Frequency of laboratory monitoring of chronic medications administered to nursing facility residents: Results of a national Internet-based study. Consult Pharm 2008;23(5):387-395.

9. Longe RL. Laboratory tests recommended by pharmacists in a skilled-nursing facility. Am J Hosp Pharm 1989;46(5):970-972.

10. Cooper JW. Routine laboratory tests: Helpful or just costly? Nurs Homes Sr Citiz Care 1988;37(6):8-10.

11. Wolfstadt JI, Gurwitz JH, Field TS, et al. The effect of computerized physician order entry with clinical decision support on the rates of adverse drug events: A systematic review. J Gen Intern Med 2008;23(4):451-458.

12. Kaushal R, Shojania KG, Bates DW. Effects of computerized physician order entry and clinical decision support systems on medication safety: A systematic review. Arch Intern Med 2003;163(12):1409-1416.

13. Judge J, Field TS, DeFlorio M, et al. Prescribers’ responses to alerts during medication ordering in the long term care setting. J Am Med Inform Assoc 2006;13(4):385-390.

14. Gurwitz JH, Field TS, Rochon P, et al. Effect of computerized provider order entry with clinical decision support on adverse drug events in the long-term care setting. J Am Geriatr Soc 2008;56(12):2225-2233.

15. Rochon PA, Field TS, Bates DW, et al. Computerized physician order entry with clinical decision support in the long-term care setting: Insights from the Baycrest Centre for Geriatric Care. J Am Geriatr Soc 2005;53(10):1780-1789.

16. Feldstein AC, Smith DH, Perrin N, et al. Improved therapeutic monitoring with several interventions: A randomized trial. Arch Intern Med 2006;166(17):1848-1854.

17. Subramanian S, Hoover S, Gilman B, et al. Computerized physician order entry with clinical decision support in long-term care facilities: Costs and benefits to stakeholders. J Am Geriatr Soc 2007;55(9):1451-1457.

18. Tommasulo B, Cal CE, Hom S, et al. Optimizing fingerstick glucose monitoring in the nursing home. J Am Med Dir Assoc 2008;9(3):B20.

19. Rizvi AA. Management of diabetes in older adults. Am J Med Sci 2007;333(1):35-47.

20. Meyers RM, Broton JC, Woo-Rippe KW, et al. Variability in glycosylated hemoglobin values in diabetic patients living in long-term care facilities. J Am Med Dir Assoc 2007;8(8):511-514.

21. Debard A, Charmion S, Ben Ameur S, et al. Inappropriate low glycated hemoglobin and hemolysis [in French]. Rev Med Interne 2009;30(6):525-527.

22. Camargo JL, Gross JL. Conditions associated with very low values of glycohaemoglobin measured by an HPLC method. J Clin Pathol 2004;57(4):346-349.

23. Coban E, Ozdogan M, Timuragaoglu A. Effect of iron deficiency anemia on the levels of hemoglobin A1c in nondiabetic patients. Acta Haematol 2004;112(3):126-128.

24. Kearon C, Ginsberg JS, Kovacs M, et al; Extended Low-Intensity Anticoagulation for Thrombo-Embolism Investigators. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med 2003;349(7):631-639.

25. Wittkowsky AK, Nutescu EA, Blackburn J, et al. Outcomes of oral anticoagulant therapy managed by telephone vs. in-office visits in an anticoagulation clinic setting. Chest 2006;130(5):1385-1389.

26. Bond CA, Raehl CL. Pharmacist-provided anticoagulation management in United States hospitals: Death rates, length of stay, Medicare charges, bleeding complications, and transfusions. Pharmacotherapy 2004;24(8):953-963.

27. Dager WE, Branch JM, King JH, et al. Optimization of inpatient warfarin therapy: Impact of daily consultation by a pharmacist-managed anticoagulation service. Ann Pharmacother 2000;34(5):567-572.

28. Bungard TJ, Gardner L, Archer SL, et al. Evaluation of a pharmacist-managed anticoagulation clinic: Improving patient care. Open Med 2009;3(1):e16-e21.

29. Jones C, Lacombe G. Enhancing patient care via a pharmacist-managed rural anticoagulation clinic. Healthc Q 2009;13 Spec No:69-74.

30. Burkiewicz JS. Effect of access to anticoagulation management services on warfarin use in patients with atrial fibrillation. Pharmacotherapy 2005;25(8):1062-1067.

31. Poon IO, Lal L, Brown EN, Braun UK. The impact of pharmacist-managed oral anticoagulation therapy in older veterans. J Clin Pharm Ther 2007;32(1):21-29.

32. Garwood CL, Dumo P, Baringhaus SN, Laban KM. Quality of anticoagulation care in patients discharged from a pharmacist-managed anticoagulation clinic after stabilization of warfarin therapy. Pharmacotherapy 2008;28(1):20-26.

33. Menzin J, Boulanger L, Hauch O, et al. Quality of anticoagulation control and costs of monitoring warfarin therapy among patients with atrial fibrillation in clinic settings: A multi-site managed-care study. Ann Pharmacother 2005;39(3):446-451.

34. Amruso NA. Ability of clinical pharmacists in a community pharmacy setting to manage anticoagulation therapy. J Am Pharm Assoc 2004;44(4):467-471.

35. Jackson SL, Peterson GM, Vial JH, Jupe DM. Improving the outcomes of anticoagulation: An evaluation of home follow-up of warfarin initiation. J Intern Med 2004;256(2):137-144.

36. Ernst ME, Brandt KB. Evaluation of 4 years of clinical pharmacist anticoagulation case management in a rural, private physician office. J Am Pharm Assoc 2003;43(5):630-636.J Thromb Thrombolysis 2003;15(2):113-118.

38. Holden J, Holden K. Comparative effectiveness of general practitioner versus pharmacist dosing of patients requiring anticoagulation in the community. J Clin Pharm Ther 2000;25(1):49-54.

39. Locke C, Ravnan SL, Patel R, Uchizono JA. Reduction in warfarin adverse events requiring patient hospitalization after implementation of a pharmacist-managed anticoagulation service. Pharmacotherapy 2005;25(5):685-689.

40. Oake N, Fergusson DA, Forster AJ, van Walraven C. Frequency of adverse events in patients with poor anticoagulation: A meta-analysis [published correction appears in CMAJ 2007;177(3):271]. CMAJ 2007;176(11):1589-1594.

41. Dorfman DM, Goonan EM, Boutilier MK, et al. Point-of-Care (POC) versus central laboratory instrumentation for monitoring oral anticoagulation. Vasc Med 2005;10(1):23-27.

42. Greenway A, Ignjatovic V, Summerhayes R, et al. Point-of-care monitoring of oral anticoagulation therapy in children. Comparison of the CoaguCheck XS system with venous INR and venous INR using and International Reference Thromboplastin preparation (rTF/95). Thromb Haemost 2009;102(1):159-165.

43. Franke CA, Dickerson LM, Carek PJ. Improving anticoagulation therapy using point-of-care testing and a standardized protocol. Ann Fam Med 2008;6(suppl 1):S28-S32.

44. BIOUSA Inc: Medical Products. https://www.biousa.com. Accessed May 10, 2010.