Abstract
The monitoring of anticoagulation therapy using vitamin K antagonists demands meticulous attention to various clinical and pharmacologic parameters known to interact with this drug. 1 Pharmacist-managed anticoagulation clinics (ACCs) have consistently been shown to improve the percentage of time within the target therapeutic range (TTR) and reduce complications and mortality rates, 2–4 leading to improved health care utilization and proven cost-effectiveness. 3,5 In 2001, we began a program of introducing clinical pharmacists into our physician-run ACC, with a comprehensive training and certification program and gradual evolvement into a physician–pharmacists-managed ACC. An annual audit of the performance of our ACC was simultaneously conducted. Herein, we report the transitional measures undertaken and the quality outcomes of this hybrid model.
The ACC of our 1400-bed tertiary referral center is run by hematologists but serves all clinical disciplines, with patients on intermediate and long-term anticoagulation therapy for various indications. Prior to 2001, a rotating junior medical officer managed approximately half of clinic attendees with stable maintenance dose warfarin. New or complicated patients were seen by hematologists. A formal anticoagulation management training session was incorporated into our department’s teaching program for junior doctors.
In collaboration with the department of pharmacy, we then started a program of transforming our ACC into a physician/pharmacists-run entity. A core group of pharmacists was recruited and trained under the hematologists (L.H.L. and H.J.N.). A senior pharmacist (M. C. K.) received additional training at an overseas center and supervised the pharmacists in the program. The training module is shown in Table 1 . From a single weekly session, the ACC was expanded to two 3-hour sessions each week, run by 2 hematologists, a junior doctor, and 2 pharmacists. A work flow and governance structure for the clinic is summarized in Figure 1 . Dose adjustments were independently decided by pharmacists but prescriptions were countersigned later by a hematologist, per the Singapore regulatory requirements. Computer-aided warfarin dosing was not available. A standard checklist of triggers for referrals to a doctor is shown in Figure 1.
Training Module for Anticoagulation Clinic Pharmacists
Abbreviations: ACC, anticoagulation clinics; INR, international normalized ratio.
Work flow and governance structure of the anticoagulation clinics (ACCs).
From September 2000 until September 2004, a quality audit of the performance of the ACC was conducted for each 12-month period (henceforth referred as Y0-Y3). Y0 reflected the performance of a clinician-run ACC, while Y1 to Y3 showed the serial performance of the hybrid pharmacist-assisted ACC. Information on international normalized ratio (INR) readings, warfarin dosages, and clinic visits were obtained from computerized patient records. TTR was the primary measure of quality of anticoagulation therapy and was calculated using the Rosendaal method. 6 Therapeutic INR range was defined as readings within ±0.1 of patients’ target ranges. Thrombotic and bleeding complications were secondary outcome measures. Thrombotic events refers to any ischemic or thromboembolic stroke, deep vein thrombosis, pulmonary embolism, or thromboembolism of any other sites. 7 Major bleeding is defined as (1) fatal bleeding, (2) symptomatic bleeding in a critical area or organ, and (3) bleeding causing a drop in hemoglobin of 2 g/dL or more, or transfusion of 2 or more units of whole blood or red cells. 8 These events were recorded during clinic consultations and checked against computerized records of discharges and emergency room visits to this hospital. Statistical analysis of the data was performed using SPSS version 12.0. Poisson regression analysis (using SAS 9.1) test was employed for assessing thrombosis and bleeding. Statistical significance was set at P < .05. This study was approved by our Institutional Review Board.
The Y0 audit included 111 patients with the patient pool doubling in Y1 to Y3. This increment was due to the transfer of patients whose anticoagulation therapy was previously managed by individual physicians. The demographic profile of patients and indications for anticoagulation were similar in all study periods (Table 2 ). Substantial variations in the INR target ranges were due to the tailoring of anticoagulation therapy against individual risk/benefit profile by patients’ primary physicians. The majority of patients were managed within the standard range of 2.0 to 3.0 and 2.5 to 3.5.
Comparison of Patient Demographics, Indications, and Target Intensity of Anticoagulation Therapy
Abbreviations: MHV, mechanical heart valves; DVT, deep vein thrombosis; PE, pulmonary embolism; TE CVA, thromboembolic cerebrovascular accident causes other than AF/MHV.
Serial performance of the ACC is shown in Table 3 . The proportion of patients seen by pharmacists from Y0 to Y3 was 0%, 19%, 46%, and 59%, respectively. Average weekly warfarin dose was closely matched across all periods as were the intervals between clinic visits. Serial TTR showed a progressive improvement with a peak at Y3 (54.44% in Y3 vs 44.78% in Y0; P < .05). Overall, the number of thrombotic and bleeding events reported was low with no statistically significant differences between the various periods.
Comparison of the Performance of the Anticoagulation Clinic During Different Periods
Abbreviations: CI, confidence interval; INR, international normalized ratio.
The increasing number of anticoagulated patients stemming from an ageing population and higher acceptance of this preventive therapy had created a critical strain on manpower resources, which prompted this initiative to involve pharmacists in our ACC. This hybrid model with hematologists providing oversight and training for pharmacists was the logical choice which met the statutory requirements on drug prescription and patient care. We clearly needed a formalized training program to ensure adequate and continuous renewal of manpower. A clear structure of reporting between pharmacists and physicians to cater for complicated cases and provide sufficient patient contact with supervising hematologist was essential. The performance of the ACC following the transition to this hybrid model also had to be audited in order to objectively demonstrate the safety, viability, and utility of this model. The first period provided the data for which the performance of the ACC in the subsequent 3 years could be compared.
The use of well-trained pharmacists evidently introduced greater consistency in the delivery of anticoagulation care in our ACC, with pharmacists taking over a substantial amount of the workload. As the pool of more experienced pharmacists increased, the TTR expectedly increased in tandem as shown by the results of the audit. The highest TTR of 54.65% achieved in Y3 is comparable to the accepted norms in well-managed ACCs. 9
The rates of bleeding and thrombosis that were recorded were lower than what we had expected. A limitation of this exercise was the nature of the audit process which depended on patient recall and retrospective reviews of charts with potential underreporting of events. As patients in Singapore were not geographically assigned to a specific hospital, some bleeding and thrombotic events may have been managed in other hospitals in Singapore.
The model that we have discussed and successfully implemented is by no means unique to our institution. This detailed description may however form a useful platform for institutions seeking to involve pharmacist and other paramedical staff in the running of a successful hybrid ACC. Since completion of this audit, our TTR performance has further improved and now consistently exceed 60%.