SMJ 2004 49(2): 70
Barclay M. Goudie, West Gate Health Centre, Dundee
Kate L. Danskin, West Gate Health Centre
Sally S. Al-Agilly, Tayside Centre for General Practice, Dundee
Geraldine Fairfield, Tayside Audit Resource for Primary Care, Dundee
A. David Cunningham, Department of Cardiology, Royal Alexandra Hospital, Paisley
J. Ewan McGregor, Department of Haematology, Ninewells Hospital, Dundee
Philip G. Cachia, Department of Haematology, Ninewells Hospital, Dundee
Correspondence to: bgoudie@westgate.finix.org.uk
BMG receives funding from the Chief Scientist Office, Research Practice Scheme
Abstract
Background and aims: New systems for monitoring anticoagulation in routine clinical practice should be evaluated to inform future service development. The aim of this study was to investigate the workload, patient characteristics and bleeding complications in a nurse run, near patient monitoring service serving a population of 23,368.
Methods: Retrospective review of coagulometer and general practice case records.
Results: Between 1995 and 2000 workload increased from 154 to 269 consultations per month. Three hundred and forty four patients were followed up during a cumulative period of 664.8 years. The 8 week period prevalence of anticoagulation was 8.6 per thousand patients at the close of the study. The median age at enrolment was 71 years. Eighteen per cent of treatment episodes were started when patients were 80 years or older. The most common indication was non rheumatic atrial fibrillation (NRAF) in 51% of treatment episodes. During the study 49.6 % [95% confidence interval (CI) 47.8%-50.6%] of international normalised ratio (INR) tests were within target range and INR was estimated to be within the target range 62% (95% CI 61.8%-62.2%) of the time. INR control varied markedly between individual patients. There were 4 fatal bleeding events (0.6 per hundred patient years) and 14 major bleeding events (2.1 / hundred patient years).
Conclusion: Demand for anticoagulant therapy with warfarin has increased substantially. Service development is required to ensure that appropriately selected patients have access to adequately resourced monitoring which promotes optimum INR control and a minimum of bleeding complications.
Key words: monitoring anticoagulation; warfarin; near patient testing; general practice
Introduction:
Clinical and technological advances have given impetus to the development and redesign of anticoagulant monitoring services. Trial evidence for the efficacy of warfarin in primary1 and secondary2 prevention of stroke has established NRAF as a commonly occurring new indication for anticoagulant therapy. Service options have increased with the introduction of portable analysers which permit near patient testing of the INR3;4, and decision support systems which enable warfarin dosage adjustment to be made by personnel without medical or pharmacy qualifications5;6. These developments fit with the current political drive for more services to be provided in a community setting. There is randomised control trial (RCT) and some observational evidence that nurse led anticoagulation clinics in primary care can achieve the same standard of INR control as hospital clinics7;8. Service planners and clinicians require more information about the uptake, workload and outcome of anticoagulant therapy in actual clinical practice. The aim of this study was to address some of these issues by investigating the prevalence of monitoring, changing workload, patient characteristics, quality of anticoagulant control and incidence of important bleeding complications in a general practice based, nurse run, near patient monitoring scheme.
Methods:
This study was performed in four general medical practices with a total population of 23,368 patients.
Practice based near patient monitoring of anticoagulant therapy, was introduced in September 1995 with funding in the first year provided by the Primary Care Development Fund. This convenient service enabled patients to have an INR test and receive warfarin dosage advice during a single consultation in the surgery or their own home. Each practice was equipped with a Rapidpoint Coag Analyser, (Bayer plc, Bayer House, Newbury, Berkshire) which was operated by trained practice and district nurses to determine INR according to protocol. Warfarin dosage advice was also provided by nurses using a standard protocol. Nurse training, local internal and external quality control, and participation in the NEQAS main users group was administered by haematology laboratory staff 3 in accordance with British Society of Haematology guidelines 9.
All patients receiving warfarin on 1st September 1995 were invited to transfer to near patient monitoring. Five (4%) of these declined. Thereafter the vast majority of patients (approximately 98%) were included until early 2000 when one practice stopped accepting patients.
Data regarding all INR tests were downloaded from practice coagulometers and transferred onto practice based Microsoft Access databases. These data include patient identifier, operator identifier, time, date and result of all INR tests. It was not possible to identify patients in whom coagulometer data entry were inaccurate. Clinical data were obtained from case records held by the practice or in the case of deceased patients recalled from Primary Care Services in Aberdeen. This was not possible in temporary residents, and patients who had moved practice or died more than 3 years prior to case record review. Data were then anonymised and transferred to a central database for analysis. This process was approved by the Caldicott guardian.
Quality of anticoagulant control was assessed using (1) proportion of INR results within target range, and (2) proportion of time INR estimated to be within the target range. A minimum of 6 tests was required for time in range calculations which were made on the assumption of linear change between consecutive INR results. Intervals between tests of more than 8 weeks were not subject to calculation.
We classified bleeding complications as (1) fatal, or (2) major when intracranial (documented by imaging), surgery or angiographic intervention was required to stop bleeding, bleeding resulted in a reduction in haemoglobin of 2g/dl or more or necessitated transfusion of 2 or more units of blood or when bleeding resulted in permanent loss of organ function e.g. intraocular bleeding causing blindness. Other bleeding events were classified as minor.
Results:
Twelve thousand five hundred and seventy two INR results were retrieved from the coagulometer memories for the period 18th September 1995 to 1st August 2000. The actual number of tests performed was higher than this figure as all INR data for one practice was lost during a 7 month period in 1997-98. Workload increased from 154 tests per month in year one to 269 tests per month in year five (Figure 1). Each INR test represents one near patient testing consultation. Patient numbers increased from 123 in year one to 236 in year five.
Eleven thousand four hundred and forty four tests (91%) were performed in 344 patients for whom detailed clinical information was obtained. Three hundred and eighty episodes of warfarin treatment were monitored in these patients during a cumulative follow up period of 664.8 years. Multiple episodes of anticoagulation were required in patients with recurrent venous thrombo-embolism and NRAF which recurred following cardioversion. In addition patients receiving continuous treatment with warfarin were considered to have had more than one episode of monitoring when the indication changed e.g. when valvular disease was treated surgically, and in any situation where the INR target range was altered. The average retest interval was 21 days. The 8 week period prevalence of anticoagulation was 8.6 per thousand patients at the close of the study. Fifty six per cent of patients were female. Age at the time of enrolment ranged from 16 to 95 years with a median of 71 years (Figure 2). Sixty nine episodes of monitoring (18%) were started when patients were 80 years or older.
Eighty four episodes (22%) were monitored by district nurses in the community as the patients were unable to attend the surgery.
The most common indications for anticoagulation were NRAF in 194/380 (51%) and venous thrombo-embolism in 70/380 (18%) of episodes of monitoring.
The INR target range was 2 -3 in 322/380 episodes of monitoring and was wider than 1.0 in 36 (9.5%)
In the course of the study 49.6 % [95% confidence interval (CI) 47.8%-50.6%] of tests were within target range and INR was estimated to be within the target range 62% (95% CI 61.8%-62.2%) of the time.
INR control varied markedly between individual patients (Figure 3).
Four patients had fatal bleeding events (0.6 per hundred patient years) and 14 major bleeding events (2.1 / hundred patient years) occurred in 12 other patients (Table I). A further 43 bleeding episodes, in 29 patients, resulted in hospital attendance for assessment or treatment. Forty one were minor episodes and 2 could not be classified because of inadequate data. In addition 4 patients with no history of overt bleeding or demonstrable bleeding lesion received blood transfusion for anaemia.
Discussion:
These results demonstrate a rapid increase in demand for monitoring of warfarin therapy which has occurred without a comparable increase in service capacity. Increasing workload is attributable to treatment of NRAF which is a new indication for anticoagulation. Many patients are very elderly and a substantial number cannot travel independently. These findings have important clinical and resource implications.
It is not possible to correct for differences in case mix and as a consequence we cannot compare our results directly with those of RCTs and other observational studies. INR control varies between individuals and patient selection is therefore an important determinant of the quality of control achieved by a monitoring service. Populations in which there is a high prevalence of warfarin treatment are likely to include a larger subgroup in whom good control is difficult to achieve. In our study warfarin dose depended on a written protocol and new evidence suggests that computer decision support would have been more effective5;6. Ongoing staff development and more effective audit might also have promoted better INR control. There is a need for clearer policies regarding the suitability of patients for treatment and further research to promote a better understanding of the determinants of INR control in individual patients.
The benefits of anticoagulation will always have to be balanced against the risk of bleeding. Estimating this risk from the literature is difficult for many reasons including differences in definition of bleeding severity, intensity of therapy and case mix. There is some evidence that oral anticoagulation has become safer in recent years10 however our results, and those of most other observational studies 11 demonstrate that the risk in actual clinical practice is higher than in RCTs1. Evidence based assessment of thromboembolic risk has improved the selection of patients with NRAF who will benefit most from treatment12. Further development of similar tools to identify patients at high risk of bleeding13 is now required to improve safety.
At a time of increasing demand for warfarin therapy health care planners need to understand the dangers and costs associated with this treatment. Service development is required to ensure that carefully selected patients have access to high quality monitoring systems where performance review ensures optimum INR control and a minimum of bleeding complications.
Acknowledgments
We wish to thank all those colleagues in the Ancrum Medical Centre, Princes Street and West Gate Health Centre practices who contributed to the delivery and evaluation of this service. All participating practices are members of the Tayside Primary Care Research and Development Network (TayRen).
References
1. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch.Intern.Med. 1994;154:1449-57.
2. Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. EAFT (European Atrial Fibrillation Trial) Study Group. Lancet 1993;342:1255-62.
3. Cachia PG, McGregor E, Adlakha S, Davey P, Goudie BM. Accuracy and precision of the TAS analyser for near-patient INR testing by non-pathology staff in the community. J.Clin.Pathol. 1998;51:68-72.
4. Murray ET, Fitzmaurice DA, Allan TF, Hobbs FD. A primary care evaluation of three near patient coagulometers. J.Clin.Pathol. 1999;52:842-5.
5. Chatellier G, Colombet I, Degoulet P. Computer-adjusted dosage of anticoagulant therapy improves the quality of anticoagulation. Medinfo. 1998;9 Pt 2:819-23.
6. Manotti C, Moia M, Palareti G, Pengo V, Ria L, Dettori AG. Effect of computer-aided management on the quality of treatment in anticoagulated patients: a prospective, randomized, multicenter trial of APROAT (Automated PRogram for Oral Anticoagulant Treatment). Haematologica 2001;86:1060-70.
7. Fitzmaurice DA, Hobbs FD, Murray ET, Holder RL, Allan TF, Rose PE. Oral anticoagulation management in primary care with the use of computerized decision support and near-patient testing: a randomized, controlled trial. Arch.Intern.Med. 2000;160:2343-8.
8. Fitzmaurice DA, Murray ET, Gee KM, Allan TF. Does the Birmingham model of oral anticoagulation management in primary care work outside trial conditions? Br J Gen Pract 2001;51:828-9.
9. Guidelines on oral anticoagulation: third edition. Br J Haematol 1998;101:374-87.
10. Palareti G, Leali N, Coccheri S, Poggi M, Manotti C, D'Angelo A et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet 1996;348:423-8.
11. Landefeld CS,.Beyth RJ. Anticoagulant-related bleeding: clinical epidemiology, prediction, and prevention. Am.J.Med. 1993;95:315-28.
12. Lip GY,.Lowe GD. ABC of atrial fibrillation. Antithrombotic treatment for atrial fibrillation. BMJ 1996;312:45-9.
13. Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am.J.Med. 1998;105:91-9.