Establishing Standards of Care for Amiodarone Monitoring in an Outpatient Setting

,
The American Journal of Pharmacy Benefits, July/August 2017, Volume 9, Issue 4

Implementing a quality-improvement project that allows for education and interdisciplinary collaboration among pharmacists, cardiologists, and primary care providers can improve rates of amiodarone monitoring.

ABSTRACT

Objectives: To assess rates of monitoring of liver, thyroid, and pulmonary function in the 6 months before and after initiation of a quality improvement project, and to evaluate the effect of pharmacist-managed dual warfarin and amiodarone monitoring on maintaining target international normalized ratio (INR).

Study Design: Retrospective electronic chart review.

Methods: Rates of monitoring according to current guidelines for amiodarone monitoring were evaluated. Patients who filled prescriptions for a) amiodarone AND b) warfarin OR any of the following direct oral anticoagulants: apixaban, rivaroxaban, or dabigatran between May 1, 2014, and April 30, 2015, were reviewed. Percent time spent in target therapeutic INR range (% TTR) was used as an outcome parameter to evaluate effect on maintaining target INR.

Results: Seventy-three subjects were in the pre-intervention group and 69 patients were in the post intervention group. All rates of 6-month monitoring increased in the post intervention group as compared with the pre-intervention group. Both the rates of monitoring of alanine aminotransferase (P = .03) and free thyroxine (P <.001) were found to be significantly higher in the post intervention group, as compared with the pre-intervention group. The % TTR was 64% in the pre-intervention group and 58% in the post intervention group.

Conclusions: Collaboration with pharmacists in an outpatient setting leads to improved rates of recommended laboratory tests for amiodarone monitoring. This study demonstrated the effectiveness of an established anticoagulation clinic in maintaining target INR.

Am J Pharm Benefits. 2017;9(4);108-115

Amiodarone is a Vaughan Williams class III antiarrhythmic agent indicated for the management of ventricular fibrillation or unstable ventricular tachycardia.1 Although not currently approved by the FDA for the treatment of atrial fibrillation (AF), it has been used in the past for this condition.2,3 Amiodarone is a life-saving and effective medication, particularly in patients in whom other agents, such as calcium channel blockers and beta blockers, are contraindicated. Moreover, epidemiologists predict that there will be a rise in the prevalence of AF in North America by 2050 to more than 12 million, which is nearly a 2-fold increase from present numbers.4 Consequently, we expect to see an increase in the use of amiodarone in the future.

Due to the drug’s lipophilic properties and large volume of distribution, the drug is stored in muscle and tissue as well as in highly perfused organs, including the liver, lungs and skin. Consequently, the drug has a variable half-life, averaging about 58 days.1 It is primarily eliminated through hepatic metabolism and biliary excretion. Due to amiodarone’s complex pharmacokinetic properties and lipophilicity, it has numerous drug interactions and can cause numerous potential adverse events (AEs).1

The AEs of amiodarone range from mild—such as nausea, photosensitivity, or skin discoloration—to more serious, such as liver, thyroid, and lung toxicities.1 Amiodarone is an iodine-containing compound that is structurally similar to thyroxine. Through multiple mechanisms, amiodarone has the capability of causing hypothyroidism or hyperthyroidism.5

One of the more serious AEs is interstitial lung disease. In a study by Dusman et al, amiodarone pulmonary toxicity was found to be more common with higher doses, advanced age, lower pretreatment diffusion lung capacity for carbon monoxide (DLCO), and high plasma concentration of the active metabolite, desethylamiodarone.6 Patients who receive a total daily dose of 400 mg for 2 months or 200 mg daily for 2 years are considered at greater risk.7 Pulmonary function tests (PFTs), including DLCO, as well as chest x-rays (CXR) can help diagnose amiodarone-induced pulmonary toxicity.8

Another amiodarone-induced toxicity is hepatotoxicity. About 25% of patients will have a transient elevation in alanine aminotransferase (ALT) levels. Symptomatic hepatitis is shown to occur in about 3% of patients on amiodarone. If untreated, more serious complications include cirrhosis, and hepatic failure can occur.9,10 Therefore, monitoring of liver function tests (LFTs) is recommended prior to amiodarone initiation and every 6 months thereafter.11

Numerous drug interactions can occur with amiodarone therapy.12-15 Amiodarone has a significant drug interaction with warfarin. The anticoagulant effect of warfarin is potentiated with the addition of amiodarone. By inhibiting the activity of CYP2C9 and CYP1A2, warfarin cannot be metabolized to inactive metabolites, ultimately leading to an increase in international normalized ratio (INR). Literature suggests a reduction in warfarin dose is recommended with the addition of amiodarone to a regimen.16,17 Within our facility, warfarin weekly dose will be lowered by 30% if amiodarone is added to the patient regimen.

Due to the numerous AEs and toxicities associated with amiodarone, patients being started on this drug must be closely monitored. The North American Society of Pacing and Electrophysiology (NASPE) has specific recommendations regarding baseline and routine monitoring of patients being initiated on amiodarone therapy, as outlined in Table 1.11

Despite these recommendations, several studies suggest that the rates of laboratory monitoring are lower than recommended.18-20 The use of pharmacists to improve laboratory monitoring of amiodarone has been evaluated in several studies, and the results demonstrate that utilizing pharmacists for amiodarone monitoring improved adherence to appropriate monitoring.2 To date, only one study has evaluated amiodarone monitoring in a collaborative setting with pharmacists in a Veterans Affairs (VA) Healthcare System. Graham et al demonstrated that implementing a pharmacy-managed protocol significantly improved the rates of maintenance amiodarone monitoring. However, the rates of baseline monitoring were low (<50% for all baseline parameters).21 Due to the collaborative efforts of our cardiology team and pharmacists, we hypothesized that our rates for ongoing monitoring would be higher when compared with those of other facilities. We also anticipated that the addition of a quality improvement project with the involvement of pharmacists would greatly improve our rates of 6-month and annual monitoring.

To date, no studies have evaluated concomitant pharmacist-led monitoring of warfarin and amiodarone in a clinic setting. Little literature is available that provides an accurate mean percentage in which the INR was in therapeutic range. In a meta-analysis conducted by Baker et al, AF patients spent a mean of 55% in their therapeutic range. It was also hypothesized that these patients treated in community setting compared with anticoagulation clinic spent 11% less time in therapeutic INR range.22 Landmark studies involving warfarin use in AF have ranged in percentage of time in therapeutic INR range (TTR) from 55% to 64%.23-25 Our study would greatly add to the data that are already available for the management of warfarin outside of a controlled study environment.

METHODS

Description of Amiodarone Monitoring Quality Improvement Project

The Amiodarone Monitoring Quality Improvement Project was established as a pilot clinic within the anticoagulation clinic at a VA hospital. The anticoagulation clinic primarily manages warfarin and direct oral anticoagulant (DOAC) monitoring of all veterans within the local VA system. Prior to establishing the pilot clinic, pharmacists met with the hospital cardiologists in order to establish an amiodarone monitoring protocol (Table 1). As identified in Table 1, this protocol is more stringent than current amiodarone monitoring guidelines due to local cardiology preference. In the previous process, amiodarone would be initiated on an appropriate patient within the inpatient setting after obtaining baseline laboratory values and tests. At 6 weeks and at 6 months, the patient would return to the electrophysiology clinic for monitoring. After 6 months, patients would be discharged to primary care to be followed by primary care providers for their continued 6-month and annual monitoring thereafter.

In the VA facility, ongoing education and protocols were implemented over the past several years to ensure that amiodarone monitoring was done appropriately during the transition of care from cardiology to primary care. We have continued to see improvement in the rates of monitoring. In order to further assist in continuity of care, a quality improvement project was established on November 1, 2014. Patients who are on warfarin and DOACs are routinely monitored within the medication management clinic.

Patients on warfarin therapy have routine INR follow-up and patients on DOAC therapy are periodically assessed for renal function, drug interactions, and other issues that may preclude them from continuing DOAC therapy. During this routine monitoring, pharmacists identify if these patients are also on amiodarone. If amiodarone was initiated within the past 6 weeks, pharmacists use a computerized template to write a note into the patient’s chart. If there are certain baseline laboratory values or imaging that are incomplete, the pharmacist orders these labs so that baseline values are available for future comparison. While ideally baseline monitoring is done prior to amiodarone initiation, if baseline monitoring was not completed at that time, it is still important to complete baseline labs or imaging as soon as possible after initiation. For patients already on maintenance amiodarone, the pharmacists assess whether 6-month and annual monitoring were completed. If any monitoring points are missing, pharmacists can place orders for the missing laboratory values. CXRs are not ordered directly by a pharmacist, but if a CXR is not completed, the appropriate provider is alerted about overdue required monitoring. Figures 1 and 2 highlight the study intervention on Patient X on warfarin/DOAC with concurrent amiodarone.

This study is a retrospective chart review approved by the Central Arkansas Veterans Healthcare System (CAVHS) Department of Veterans Affairs Institutional Review Board (IRB) and CAVHS Research and Development Committee. The CAVHS Computerized Patient Recording System (CPRS) database was utilized for the study. A patient list was obtained by identifying patients who had filled a prescription for a) amiodarone AND b) warfarin OR any of the following DOACs: apixaban, rivaroxaban, or dabigatran.

This chart review was done in 2 phases: The first phase was six months prior to the pilot clinic and the second phase was six months after the establishment of the pilot clinic for amiodarone monitoring. The following patient information was collected: age; gender; race; amiodarone dose; amiodarone indication, and warfarin or DOAC indication. If amiodarone was newly initiated during the evaluated time period, patient charts were reviewed to see if appropriate laboratory values and imaging were or were not obtained upon amiodarone initiation. These baseline monitoring parameters include ECG, ALT, aspartate transaminase (AST), thyroid-stimulating hormone (TSH), free thyroxine (free T4), PFTs with DLCO, and CXR. If amiodarone was already initiated prior to the evaluated time period, TSH, free T4, AST, and ALT should be done every 6 months, and data were evaluated for completion of these monitoring parameters. CXR should be done annually, and the data were analyzed to see if CXR was obtained in past year. Since all patients were started on amiodarone inpatient by cardiologists and ECG is routinely completed, assessment of completion of baseline ECG was not included. INR values for all patients on concomitant warfarin, who met exclusion and inclusion criteria, were also reviewed.

All eligible veterans being monitored by the VA anticoagulant clinic who initiated on amiodarone therapy or were currently receiving amiodarone therapy between May 1, 2014, and April 30, 2015, were included. Subjects in whom amiodarone was started by an outside provider, had received less than 6 weeks of amiodarone therapy, were pregnant, were under hospice care, or were in home-based primary care (HBPC) were excluded. For the secondary outcome, excluded INR readings were those collected at first initiation of warfarin as well as subtherapeutic INRs at times when warfarin therapy was temporarily halted before or after certain procedures.

The primary outcome was to assess the rates of 6-month monitoring (LFTs, thyroid function tests [TFTs]) and annual monitoring (CXR) in the pre- and post intervention groups. The secondary outcome was to assess the rates of baseline monitoring of liver (AST, ALT), thyroid (TSH, free T4), and pulmonary function (PFTs with DLCO, CXR) in the pre- and post intervention group. Another secondary outcome was to evaluate the effect of pharmacist-managed dual warfarin and amiodarone monitoring on maintaining target INR. This will be measured using the fraction of INRs method of % TTR (number of INRs in target range, divided by total number of INR in selected time interval, multiplied by 100).

Baseline characteristics of race and gender were compared between the pre- and post intervention groups using the Pearson’s c2 test, except for Student’s t test for age and Mann Whitney U test for dosing. Primary outcome assessing rates of monitoring was compared using Pearson’s c2 test. P values <.05 were considered statistically significant.

RESULTS

A total of 202 patients were identified as having concomitant amiodarone and warfarin or DOAC prescriptions filled between the time period of May 1, 2014, and April 30, 2015 (Figure 3): 108 were stratified into the pre-intervention group and 90 were stratified into the post intervention group. After evaluating exclusion criteria, 73 and 69 patients remained in the pre-and post intervention groups, respectively.

The baseline characteristics of the pre- and post intervention group were similar, as shown in Table 2. Overall, the mean ages were 69.8 and 70 years for the pre- and post intervention groups, respectively. The majority of subjects were male and nonblack. Most subjects were receiving vitamin K antagonists for anticoagulation (79.5% in pre-intervention group vs 71% in post intervention group) and most subjects were receiving anticoagulation for the indication of AF. Most subjects were taking a maintenance amiodarone dose of 200 mg/day (83.6% and 76.8% in the pre- and post intervention groups, respectively).

Our primary outcome was to assess the rates of 6-month monitoring of liver (AST, ALT), thyroid (TSH, free T4), and pulmonary function (PFTs with DLCO, CXR) in the pre- and post intervention groups. All rates of 6-month monitoring increased in the post intervention group (Table 3, Figure 4) compared with the pre-intervention group. Differences between the rates of monitoring of ALT (P = .03) and free T4 (P <.001) were found to be statistically significant between the pre- and post intervention groups. No significant differences in the rate of annual CXR monitoring were discovered (Table 4).

To keep the pre- and post intervention groups independent, if subjects were initiated on amiodarone during the post intervention time period, the baseline monitoring would only be counted towards the post intervention group if completed on initiation or within 6 weeks of amiodarone intervention during the study period. During the study period, there were only 15 patients newly initiated on amiodarone (Table 5). In the pre-intervention group, all 73 patients were assessed if baseline monitoring was completed on initiation or within 6 weeks of initiation, even if it was done before the study period. This is because prior to the establishment of pilot clinic, there was no change in usual care of amiodarone monitoring. Patients were initiated on amiodarone and monitoring was completed by cardiologists. Although we did note improvements and maintenance in the rates of monitoring, there were no statistical differences between baseline monitoring in the pre- and post intervention groups.

Our secondary outcome was to evaluate the effect of a pharmacist-managed dual warfarin and amiodarone monitoring protocol to maintain target INR as measured by % TTR. After exclusion of INRs taken while warfarin was being held for a procedure or upon warfarin initiation, % TTR was 64% in the pre-intervention group and 58% in the post intervention group (Table 6).

DISCUSSION

In this study, rates of amiodarone monitoring before and after establishing a quality improvement project within an outpatient clinic setting were compared. In terms of baseline monitoring, there was not a significant change between the pre- and post intervention group, likely because baseline monitoring was completed before amiodarone initiation in the inpatient setting. Current rates of baseline amiodarone monitoring were sustained within the inpatient setting, possibly due to the extensive education done by our inpatient pharmacists and cardiologists during the establishment of the amiodarone monitoring protocol.

For the 6-month monitoring, the impact of collaborating with pharmacists could clearly be noted. All rates of monitoring were improved in the post intervention group compared with the pre-intervention group. For annual monitoring, annual monitoring of CXR did not improve, likely because the post intervention time period was six months and CXR are annual monitoring tests. In the pre-intervention group, a year had already lapsed, and as long as CXR had been done within the prior year, it would count towards the rate of monitoring. We anticipate that if the post intervention period was extended to a year, we would see a statistically significant difference. Statistically significant differences were noted in both the rates of ALT and free T4 monitoring. There was no significant difference between the treatment groups for AST monitoring; this may be because clinic pharmacists only order 6-month labs if they are not already done, and AST labs may have been previously ordered, while ALT may have been missing.

Current literature about appropriate amiodarone monitoring is sparse. The rates of monitoring seen in several studies leave much room for improvement. In a cross-sectional retrospective chart review, Stelfox et al noted that PFTs were completed at baseline for 52% of patients, which is much higher than our rate of baseline PFT monitoring. A potential reason for this difference in rates of monitoring is that the study by Stelfox et al did not specify if PFTs with DLCO were completed.23 In our study, if DLCO was not obtained as recommended by NASPE guidelines, then PFTs would not be marked as complete. Another potential reason is that the patient may not be stable enough to complete PFTs prior to amiodarone initiation and the PFTs may have been obtained soon after. Stelfox et al also demonstrated that ongoing monitoring after baseline monitoring occurs in 22% of patients.23 Compared with Stelfox’s results and those of other published studies, our rates of 6-month monitoring appear to be consistently better. Our study had higher rates of 6-month monitoring of ALT (84% vs 88%), free T4 (41% vs 65%), and annual monitoring of CXR (57% vs 70%) and PFTs (52% vs 68%) compared with the pharmacist-managed group in Spence et al.3 Our rates of monitoring is also better compared with those studied by Tjia et al, in which 6-month monitoring of LFTs and TFTs were completed in 35% and 20% of patients, respectively. In addition, only half of the patients had annual CXRs completed.20 Our improved monitoring-completion numbers are likely a result of the education process that was implemented within our facility among pharmacists and primary care providers and the use of a more collaborative approach. Another reason for our improved rates of monitoring could be that our study was conducted in a VA facility, which is generally a closed medical care system. Pharmacists are able to review monitoring parameters conducted in different settings of any VA facility. Nonetheless, compared with the study conducted by Graham et al in a VA facility, our rates of baseline monitoring were still improved.21

This study demonstrated the effectiveness of an anticoagulation clinic in maintaining target INR. We were able to demonstrate % TTR similar to those obtained in randomized control trials.24-26 The % TTR was similar in both the pre- and post intervention groups. This is likely because the pharmacists in the anticoagulation clinic have always been vigilant in monitoring for drug interactions with warfarin and preemptively adjusting warfarin doses. Percent TTR may have been lower in the post intervention group as compared with the pre-intervention group due to various reasons. The post intervention time period was from November to April, and patients may have been less likely to routinely maintain compliance and attend clinic appointments in the wintertime. In addition, the winter months tend to see higher antibiotic use, which can lead to INR variations.27

LIMITATIONS

There were a few limitations to this study, one of which was the time limitation of 6 months in the pre- and post intervention group. We were not able to identity within the electronic system if monitoring was done at another facility. Furthermore, we excluded patients if amiodarone was started by a non-VA provider. However, since the pilot clinic was monitoring patients regardless of whether or not amiodarone was initiated by a non-VA provider, excluding these patients may have decreased the rates of monitoring. Additionally, subjects being followed in the community-based outpatient clinic (CBOC) facilities were included but were not being monitored by the pilot clinic, since CBOCs have their own clinic that routinely monitored warfarin and DOAC patients. Another potential limitation of our study is that laboratory testing, CXR, and PFTs may have been obtained for a reason other than amiodarone monitoring. This may overestimate our identified rates of amiodarone monitoring. Because the IRB approval process for research took an extended time, it was not possible to extensively re-educate those involved in the quality improvement project. However, this will be continued on an annual basis in our facility. Finally, major trials involving warfarin have used the Rosendaal method of % TTR,24-26 which incorporates the frequency of INR measurements and the INR values, and assumes that changes between consecutive INR measures are linear. Our calculation of % TTR using fraction of INRs does not take this into account. One disadvantage of using fraction of INRs is that more frequent testing in patients with labile INRs would result in underestimation of TTR.28 Therefore, it is possible that our % TTR may be even higher than depicted.

It is important to bear in mind that this study was conducted in a VA facility, which is largely a closed medical care system. The pilot clinic was able to be sustainable because outpatient pharmacists are able to review if laboratory values and imaging were done at any point and in any setting within a VA facility. It would be difficult to expect similar results in open care systems in which patients may see varying providers for follow-up care and monitoring. A clinic within a hospital system may not be easily privy to patient data from other facilities. Finally, our results may not be generalizable to female patients since our study sample involved 97% to 98% male patients. Nevertheless, while amiodarone use may sometimes present differences in AEs between males and females, no differences exist between recommended amiodarone monitoring parameters.

CONCLUSIONS

This study demonstrates that incorporating pharmacists into a collaborative management of patients on amiodarone leads to sustained and improved rates of monitoring on recommended laboratory tests. The interdisciplinary involvement of our electrophysiology team, cardiology team, pharmacists, and primary care providers has been pivotal to the improved monitoring for amiodarone seen in our facility. Further research can delve into the impact of improved monitoring on outcome measurements, such as discontinuation of amiodarone due to toxicities or prevention of emergency department visits.

Author Affiliations: Fairleigh Dickinson University (SC), Florham Park, NJ; Central Arkansas Veterans Healthcare System (RA), Little Rock, AR.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (SC, RA); acquisition of data (SC, RA); analysis and interpretation of data (SC, RA); drafting of the manuscript (SC, RA); critical revision of the manuscript for important intellectual content (SC, RA); statistical analysis (SC, RA); administrative, technical, or logistic support (SC, RA).

Address correspondence to: Sibyl Marie Cherian, PharmD, BCPS, BCGP, Fairleigh Dickinson University, 230 Park Ave, Florham Park, NJ 07932. E-mail: sibyl.cherian@gmail.com.

REFERENCES

1. Cordarone [package insert]. Philadelphia, PA: Wyeth Pharmaceuticals, Inc; 2004.

2. Zimetbaum P, Ho KK, Olshansky B, et al; FRACTAL Investigators. Variation in the utilization of antiarrhythmic drugs in patients with new-onset atrial fibrillation. Am J Cardiol. 2003;91(1):81-83.

3. Spence MM, Polzin JK, Weisberger CL, Martin JP, Rho JP, Willick GH. Evaluation of a pharmacist-managed amiodarone monitoring program. J Manag Care Pharm. 2011;17(7):513-522.

4. Ehrlich JR, Nattel S. Novel approaches for pharmacological management of atrial fibrillation. Drugs. 2009;69(7):758-774. doi: 10.2165/00003495-200969070-00001. Review.

5. Basaria S, Cooper DS. Amiodarone and the thyroid. Am J Med. 2005;118(7):706-714. Review.

6. Dusman RE, Stanton SS, Miles WM, et al. Clinical features of amiodarone-induced pulmonary toxicity. Circulation 1990;82(1):51-9.

7. Wolkove N, Baltzan M. Amiodarone pulmonary toxicity. Can Respir J. 2009;16(2):43-48.

8. Vassallo P, Trohman RG. Prescribing amiodarone: an evidence-based review of clinical indications. JAMA. 2007;298(11):1312-1322. Review.

9. Lewis JH, Ranard RC, Caruso A, et al. Amiodarone hepatotoxicity: prevalence and clinicopathologic correlations among 104 patients. Hepatology. 1989;9(5):679-685.

10. Richer M, Robert S. Fatal hepatotoxicity following oral administration of amiodarone. Ann Pharmacother. 1995;29(6):582-586.

11. Goldschlager N, Epstein AE, Naccarelli GV, et al; Practice Guidelines Sub-committee, North America Society of Pacing and Electrophysiology (HRS). A practical guide for clinicians who treat with amiodarone: 2007 [published correction appears in Heart Rhythm. 2007;4(12):1590]. Heart Rhythm. 2007;4(9):1250-1259.

12. US Food and Drug Administration. Information for healthcare professionals: simvastatin (marketed as Zocor and generics), ezetimibe/simvastatin (marketed as Vytorin), niacin extended-release/Simvastatin (marketed as Simcor), used with amiodarone (Cordarone, Pacerone). FDA website. www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm124362.htm. Updated August 15, 2013. Accessed September 30, 2014.

13. US Food and Drug Administration. Altoprev (Lovastatin extended-release) tablets detailed view: safety labeling changes approved by FDA Center for Drug Evaluation and Research (CDER). FDA website. www.fda.gov/Safety/MedWatch/SafetyInformation/ucm303638.htm. Accessed September 30, 2014.

14. US Food and Drug Administration. Cordarone (amiodarone HCl) tablets detailed view: safety labeling changes approved by FDA Center for Drug Evaluation and Research (CDER). FDA website. www.fda.gov/Safety/MedWatch/SafetyInformation/ucm239971.htm. Accessed September 30, 2014.

15. Siddoway LA. Amiodarone: guidelines for use and monitoring. Am Fam Physician. 2003;68(11):2189-2196.

16. Almog S, Shafran N, Halkin H, et al. Mechanism of warfarin potentiation by amiodarone: dose- and concentration-dependent inhibition of warfarin elimination. Eur J Clin Pharmacol. 1985;28(3):257-261.

17. Sanoski CA, Bauman JL. Clinical observations with the amiodarone/warfarin interaction: dosing relationship with long-term therapy. CHEST. 2002;121(1):19-23.

18. Raebel MA, Carroll NM, Simon SR, et al. Liver and thyroid monitoring in ambulatory care patients prescribed amiodarone in 10 HMOs. J Manag Care Pharm. 2006;12(8):656-664.

19. Bickford CL, Spencer AP. Adherence to the NASPE guideline for amiodarone monitoring at a medical university. J Manag Care Pharm. 2006;12(3):254-259.

20. Tjia J, Field TS, Garber LD, et al. Development and pilot testing of guidelines to monitor high-risk medications in the ambulatory setting. Am J Manag Care. 2010;17(7):489-496.

21. Graham MR, Wright MA, Manley HJ. Effectiveness of an amiodarone protocol and management clinic in improving adherence to amiodarone monitoring guidelines. J Pharm Technol. 2004;20(1):5-10.

22. Baker WL, Cios DA, Sander SD, Coleman CI. Meta-analysis to assess the quality of warfarin control in atrial fibrillation patients in the United States. J Manag Care Pharm. 2009;15(3):244-252.

23. Stelfox HT, Ahmed SB, Fiskio J, Bates DW. Monitoring amiodarone’s toxicities: recommendations, evidence, and clinical practice. Clin Pharmacol Ther. 2004;75(1):110-122.

24. Connolly SJ, Ezekowitz MD, Yusuf S, et al; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139-1151. doi: 10.1056/NEJMoa0905561.

25. Granger CB, Alexander JH, McMurray J, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365 (11):981-992. doi: 10.1056/NEJMoa1107039

26. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883-891.

27. Suda KJ, Hicks LA, Roberts RM. Trends and seasonal variations in outpatient antibiotic prescription rates in the United States, 2006-2010. Antimicrob Agents Chemother. 2014;58(5):2763-2766.

28. Schmitt L, Speckman J. Quality assessment of anticoagulation dose management: comparative evaluation of time-in-therapeutic range. J Thromb Thrombolysis. 2003;15(3):213-216.