- Top of page
Warfarin is the most commonly prescribed oral anticoagulant worldwide for the treatment and prevention of various thromboembolic events (TEs). The anticoagulation effect of warfarin, measured by the international normalized ratio (INR), is subject to wide inter- and intraindividual variability. In addition, warfarin has such a narrow therapeutic window that it could lead to haemorrhagic or thromboembolic events even with careful dosage titration [1, 2].
Management of warfarin therapy in Chinese patients has been challenging because Chinese patients are reported to be more sensitive to the anticoagulation effect of warfarin and require a 40–50% lower maintenance dose of the drug when compared with Europeans [3–5]. The incidence of major bleeding increased sharply as INR rose from 2.4 to 2.9 in a cohort of Chinese patients receiving warfarin therapy with a target INR of 2–3 . Beside the influence of environmental factors such as diet and drug–drug interactions, genetic factors apparently play an important role in affecting the dose requirement of warfarin therapy in the Chinese population [7–9]. The gene for K epoxide reductase complex 1 (VKORC1), which encodes the warfarin target protein, was recently identified. It was reported that over 70% of Hong Kong Chinese patients were homozygous for the VKORC1 H1/H1 genotype and this was associated with a lower warfarin maintenance dose in this cohort of patients .
In Hong Kong, all anticoagulation clinics are managed by physicians. Studies conducted in Western populations had indicated that clinical pharmacist-managed anticoagulation services decreased warfarin-related hospitalization, lowered the incidence of haemorrhagic and thromboembolic events and improved INR control when compared with routine medical care [10–14]. We therefore hypothesized that a new management model (with a pharmacist as the primary care provider) is more effective than the usual practice model (managed solely by physicians) in achieving target INR control in Chinese patients receiving warfarin therapy in Hong Kong. A pharmacist-managed anticoagulation clinic was recently implemented in a teaching hospital in Hong Kong to test this hypothesis and the objective of the present study was to compare the effects of a clinical pharmacist-managed anticoagulation service with those of the physician-managed service on treatment outcomes of warfarin therapy among Chinese patients in Hong Kong.
- Top of page
From November 2002 to June 2004, 141 patients enrolled at the anticoagulation clinic were evaluated for eligibility for the study (Figure 1). Two patients had postpartum deep vein thrombosis and were started on warfarin for 6 weeks and one patient refused to participate. A total of 138 patients provided written consent and were randomized to the pharmacist-managed group (n = 69) and the physician-managed group (n = 69). One patient in the pharmacist-managed group did not return for follow-up after the first visit. The outcome analysis was conducted on the 68 patients in the pharmacist-managed group and the 69 patients in the physician-managed group.
The patients’ demographic data are shown in Table 1. Patients’ characteristics in the two groups were similar, with no significant differences in age, intensity of anticoagulation therapy or type of indication, but the percentage of male patients was significantly lower in the pharmacist-managed group (35%vs. 55%; P = 0.031). One hundred and twenty-five patients (91%) were on warfarin therapy for a target INR 2–3. The most common indications for warfarin therapy were atrial fibrillation (n = 72; 53%), heart valve replacement (n = 24; 18%), deep vein thrombosis (n = 17; 12%) and pulmonary embolism (n = 9; 7%).
Table 1. Patient demographics and indications for warfarin therapy
| ||Number (%) Pharmacist-managed group||Physician-managed group||P-value|
|No. of patients||68||69||–|
|Male||24 (35)||38 (55)||0.031|
|Age (mean ± SD)||58 ± 14.0||60 ± 14.0||0.484|
|Intensity of anticoagulation therapy|
| Low (therapeutic INR: 2–3)||63 (93)||62 (90)||0.783|
| High (therapeutic INR: 2.5–3.5)|| 5 (7)|| 7 (10)|| |
| Atrial fibrillation||37 (54)||35 (51)||0.794|
| Heart valve replacement|
| Mitral valve replacement only|| 6 (9)|| 4 (6)||0.725|
| Aortic valve replacement only|| 3 (4)|| 6 (8)||0.505|
| Both mitral and aortic valve replacement|| 1 (2)|| 4 (6)||0.371|
| Deep vein thrombosis|| 8 (12)|| 9 (13)||1.00|
| Pulmonary embolism|| 5 (7)|| 4 (6)||0.745|
| Cerebrovascular accident|| 2 (3)|| 2 (3)||1.00|
| Valvular heart diseases|| 3 (4)|| 2 (3)||0.681|
| Cardiomyopathy|| 1 (2)|| 1 (1)||1.00|
| Miscellaneous|| 2 (3)*|| 2 (3)†||–|
The distribution of patient-years of the two study groups among various INR categories for low-intensity and high-intensity anticoagulation therapy is shown in Figures 2 and 3, respectively. Table 2 shows that the patients in the pharmacist-managed group spent more patient time (64%) in therapeutic INR range than those in the physician-managed group (59%) (P < 0.001). Similarly, for both low-intensity and high-intensity anticoagulation therapy, patients in the pharmacist-managed group consistently spent more patient time in the therapeutic range than those in the physician-managed group (P < 0.001). When the INR results from both groups were pooled, the 137 patients spent 61% of time in the therapeutic INR range and 77% of time in the expanded range.
Figure 2. Distribution of patient-years among the international normalized ratio (INR) categories in the low-intensity group (target INR = 2–3, extended target INR = 1.8–3.2). Pharmacist group ( ), Physician group ( )
Download figure to PowerPoint
Figure 3. Distribution of patient-years among the international normalized ratio (INR) categories in the high-intensity group (target INR = 2.5–3.5, extended target INR = 2.3–3.7). Pharmacist group ( ), Physician group ( )
Download figure to PowerPoint
Table 2. Control of INR and incidence of complications
| ||Pharmacist-managed group||Physician-managed group||P-value|
|Patient time (%) in therapeutic INR range||64||59||< 0.001|
|Patient time (%) in therapeutic range for the group with target INR of 2–3||64||60||< 0.001|
|Patient time (%) in therapeutic range for the group with target INR of 2.5–3.5||66||44||< 0.001|
|Patient time (%) in expanded therapeutic range*||78||76||< 0.001|
|Numbers with bleeding (number per 100 patient-years)|
| Major|| 1 (1.6)|| 2 (3.1)|| 1.00|
| Fatal|| 0|| 0|| |
|Number of thromboembolic events (number per 100 patient-years)|
| Major|| 1 (1.6)|| 1 (1.6)|| 1.00|
| Fatal|| 0|| 0|| |
No patient had a fatal event in the present study. One patient from each group experienced a transient ischaemic attack (1.6 events per 100 patient-years; P = 1.00). Both patients were hospitalized and the symptoms resolved spontaneously. The INR at the time of the event was not documented in either case. One patient in the pharmacist-managed group experienced menorrhagia with anaemic symptoms at an INR of 2.39 during admission (1.6 events per 100 patient-years) and two patients in the physician-managed group experience bleeding (one had extensive bruising with gross haemoptysis with an INR of > 5 and one had a peptic ulcer with iron deficiency anaemia; 3.1 events per 100 patient-years) (P = 1.00).
The mean cPPPM and its components are shown in Table 3. There were no significant differences in the mean cPPPM regarding medication use, emergency room utilization and hospitalization. The mean cPPPM for the pharmacist-managed clinic (US$46 ± 35) (£26 ± 20) was significantly lower than that for the physician-managed clinic (US$67 ± 51) (£38 ± 29) (P < 0.01). The mean total cPPPM of the pharmacist-managed group (US$76 ± 95) (£43 ± 53) was also significantly lower than that of the physician-managed group (US$98 ± 158) (£55 ± 89) (P < 0.01). The cost of the pharmacist-managed service was assumed to be 50% of the physician-managed service in the base-case analysis. A sensitivity analysis was therefore conducted of the cost of the pharmacist-managed service over a range of 10–100% of the cost of the physician-managed service. It showed that the total cPPPM of the pharmacist-managed group was significantly lower than that of the physician-managed group when the cost per pharmacist-managed clinic was 10–60% of the cost of the physician-managed clinic. There was no significant difference in the total cPPPM between the two groups when the cost of the pharmacist-managed clinic was 70–90% of that of the physician-managed clinic. The pharmacist-managed group became significantly more costly when the cost of the pharmacist-managed clinic was the same as that of the physician-managed clinic.
Table 3. Mean cost per patient per month
| ||Pharmacist-managedgroup, US$ (£)||Physician-managedgroup, US$ (£)||P-value|
|Clinic visit|| 45.7 ± 34.7 (25.6 ± 19.4)|| 67.4 ± 51.1 (37.7 ± 28.6)||< 0.01|
|Medication|| 2.7 ± 1.6 (1.5 ± 0.8)|| 2.6 ± 1.1 (1.4 ± 0.6)|| 0.269|
|Emergency room|| 1.0 ± 2.7 (0.5 ± 1.5)|| 1.0 ± 4.0 (0.5 ± 2.2)|| 0.195|
|Hospitalization|| 26.1 ± 90.7 (14.6 ± 50.8)|| 26.9 ± 145.3 (15.0 ± 81.4)|| 0.102|
|Total|| 76 ± 95 (43 ± 53)|| 98 ± 158 (55 ± 88)||< 0.01|
The results of PSQ-18 survey are shown in Table 4. The scores for technical quality, interpersonal manner, communication, time spent with clinician and accessibility were significantly higher in the pharmacist-managed group. There was no significant difference in the scores for general satisfaction and financial aspect. The overall mean score of the pharmacist-managed group (3.8 ± 0.2) were significantly higher than that of the physician-managed group (3.6 ± 0.3) (P < 0.001).
Table 4. Results of PSQ-18 assessment
| ||Mean score (SD)Pharmacist-managedgroup||Physician-managedgroup||P-value|
|General satisfaction||4.0 (0.5)||3.8 (0.5)|| 0.134|
|Technical quality||4.0 (0.3)||3.8 (0.4)|| 0.014|
|Interpersonal manner||4.0 (0.4)||3.7 (0.6)|| 0.003|
|Communication||4.0 (0.4)||3.7 (0.6)|| 0.003|
|Financial aspect||3.6 (0.6)||3.6 (0.8)|| 0.901|
|Time spent||3.9 (0.4)||3.3 (0.6)||< 0.001|
|Accessibility||3.6 (0.4)||3.3 (0.4)||< 0.001|
- Top of page
In the present study, the Chinese patients in the pharmacist-managed group spent more time in the therapeutic and the expanded therapeutic INR ranges when compared with those in the physician-managed group and the difference achieved statistical significance. The mean cPPPM was significantly reduced by 22%[US$98 (£55) in the physician-managed group vs. US$76 (£43) in the pharmacist-manage group].
Coordinated anticoagulation care has been endorsed by the ACCP to be the primary approach in the management of warfarin therapy and it has also been accepted as the model of care in the UK and in the Netherlands . Physicians, pharmacists and nurses, if properly trained, can all be providers of an anticoagulation service. In our study, both the pharmacist-managed and the physician-managed clinic were operated as a coordinated anticoagulation service and the 137 patients managed by the two groups achieved high-quality anticoagulation control of 61% and 77% of patient time spent in the therapeutic and the expanded therapeutic INR ranges, respectively. These results are consistent with the findings in European ancestries that the therapeutic and the expanded therapeutic INR ranges were achieved 40–64% and > 75% of the time, correspondingly, in patients managed by an anticoagulation clinic. Nonetheless, the patients in the pharmacist-managed group had a slightly but significantly better INR control when compared with those in the physician-managed group. The clinical pharmacist in the pharmacist-managed group provided intense education to the patients and their caregivers, checked patients’ adherence more thoroughly and, as a drug expert, provided extra attention to the potential warfarin–drug and warfarin–herb interactions at each clinic visit. Moreover, the pharmacist conducted telephone follow-up of those with poor adherence or difficult INR control and a hotline telephone was available to the patients to inquire about warfarin-related issues. All these may contribute to the improved INR control in the patients managed by the clinical pharmacist.
In a study conducted by Fihn et al., the relative risk for a first serious bleeding event was 1.9 (95% confidence inteval 1.3, 3.0) times greater in women than in men even after adjustment for intensity of treatment. In our study, though the pharmacist-managed group had significantly more female patients than the physician-managed group (65%vs. 45%, P = 0.031), there was no significance difference in the incidence of major bleeding complications between the study and control groups. Nevertheless, the planned sample size of our study was not powered to assess the difference in incidence of complications between the two groups. The bleeding rates detected in our study (1.6–3.1%) were similar to those reported in the anticoagulation service (0–2.4%) but lower than that reported in usual medical care (3.9–17.8%) [13, 17, 18, 20, 21]. Similarly, the rates of TE in both groups were the same (1.6%) and they were also consistent with the reported TE rates for anticoagulation service (0–3.5%) but lower than that reported in usual medical care (6.2–11.8%) [13, 17, 18, 20, 22].
The mean total cPPPM was significantly reduced when patients were managed in the pharmacist-managed clinic. The result was similar to overseas cost-effectiveness analyses of pharmacist-managed anticoagulation service – it was less costly and more effective than the usual medical service provided by physicians [13, 17, 21]. The cost saving was due mainly to the decreased running cost of the clinic visits in the pharmacist-managed group. Sensitivity analysis showed that the pharmacist-managed service became more costly if the cost of the clinic visit was the same as that of the physician-managed clinic.
The patients in the pharmacist-managed group showed a significantly higher overall patient satisfaction score compared with that of the physician-managed group, attributed to the improved satisfaction scores in technical quality, interpersonal manner, communication, time spent and accessibility. In addition, the pharmacist conducted telephone follow-up for selected patients with difficult INR control or adherence issues between their clinic visits and a phone line was available for accessing pharmacist consultation. Our finding of higher patient satisfaction in the pharmacist-managed anticoagulation clinic was consistent with the results of patient satisfaction of a trial conducted by Wilson et al.
One limitation of the study is that the present sample size did not provide enough power to detect the difference in incidence of complications between the study and control groups. Nevertheless, the sample size provided adequate power to detect the difference in the percentage of patient time in target INR ranges, total direct medical cost as well as patient satisfaction. Another limitation was that the administrator of PSQ-18 was not blinded to the patient allocation despite the survey administrator not being an investigator of the present study. The actual ‘time spent’ by the pharmacist or physician for clinic visits in the two groups was not captured in our study. A sensitivity analysis was therefore included to examine the variation of cost of clinic visit that might arise due to different ‘time spent’ in the two groups. Other potential limitations applicable to both groups included self-reporting of complications, change in brand/generic of warfarin and adherence to treatment recommendations by the patients.
In conclusion, a management model for anticoagulation therapy including a pharmacist as the primary care provider with physician consultation for predetermined conditions was more effective and less costly than a model of care managed solely by physicians in achieving target INR control for Chinese patients on warfarin therapy. Currently, this is the only pharmacist-managed anticoagulation clinic in Hong Kong. The results of the present study support the implementation of this model of care for anticoagulation management in other hospital settings in Hong Kong. Future research should also investigate the cost-effectiveness of a pharmacist service to manage anticoagulation therapy in other settings such as nursing home patients and patients engaged in self-monitoring of anticoagulation therapy.