Dr David Fitzmaurice and Ellen Murray describe their primary care nurse-led anticoagulation clinic which offers an effective and popular alternative to hospital clinics


Over the past 10 years or so there has been a continuous exponential growth in the number of patients receiving oral anticoagulant therapy worldwide (warfarin in the UK).

The main driver to this increase has been the consistent evidence for a 2/3 relative risk reduction in sustaining a stroke for patients with atrial fibrillation (AF). However, there remains a large pool of patients with AF who do not receive warfarin therapy and the rise in patient numbers shows no sign of abating.

Traditionally, patients receiving warfarin in the UK have attended hospital-based outpatient clinics for measurement of the international normalised ratio (INR) and interpretation of the result by an expert, usually a haematologist. Although this model is still widely used, there is increasing strain on the hospital service because of the sheer numbers of patients.

It is estimated that up to 2% of the population would potentially benefit from warfarin, making the scale of the problem similar to that of diabetes.

As academic practitioners we became interested in how primary care could address this increasing problem and help to alleviate the pressures on secondary care while maintaining clinical standards.

Currently, around 500 000 patients in the UK take warfarin – 1% of a general practice population. The numbers continue to increase by approximately 10% per annum.

The context

Bellevue Medical Centre is an inner-city practice with high levels of deprivation and a list size of approximately 7000. As part of a small pilot research project, we set up a nurse-led oral anticoagulation clinic using computerised decision support software (CDSS) to facilitate warfarin dosing in 1994.1

The clinic was intended to function in a similar fashion to other chronic disease clinics, such as asthma and diabetes clinics. This clinic model was developed to accommodate the increased number of patients requiring warfarin monitoring.

This small-scale study demonstrated that anticoagulation control within the practice-based clinic was superior to that obtained within the hospital clinic. Within this model, however, the INR was still measured at the hospital, with results faxed back to the practice. This led to a small delay in transmitting results and dosage advice back to patients.

At this point, in 1995, we introduced near-patient testing (NPT) for INR measurement. The model described here using NPT and CDSS is now widely known as 'The Birmingham model' (see Figure 1).

Figure 1: The Birmingham model of oral anticoagulation
  • The near-patient test (NPT) currently used within the clinic is Coaguchek-S (Roche Diagnostics) which retails at around £400. Individual test strips for international normalised ratio (INR) measurement cost around £2.30 per test. We also evaluate new NPTs as they come on the market.
  • The computerised decision support software (CDSS) currently used is BAP-PC (The Department of Primary Care and General Practice, University of Birmingham) which currently retails at around £400. We were involved in the development of this software which is based on the Coventry equation for warfarin dosing. Again, we also evaluate new software systems as they become available.
  • The overall costs of the Birmingham model are around £120 per patient per year compared with around £90 per patient per year for patients attending hospital outpatients.

The anticoagulant clinic at Bellevue is currently managed by a G-grade practice nurse, who has overall responsibility for administration, audit and managing the clinic. We run two clinics per week, with one of the clinics run by a second G-grade practice nurse. This ensures that there is cover for illness and holidays. It also allows for some flexibility for patients and for early repeat testing should it be necessary.

Aims and objectives

The aim of the clinic is to offer standardised and clinically effective anticoagulation management to primary care patients receiving warfarin therapy.

The objectives are as follows:

  • To manage patients receiving warfarin therapy following stabilisation in hospital clinics, i.e. patients with at least two INR results within 0.5 of the target INR.
  • To initiate and manage warfarin therapy for patients with AF in the general practice clinic.
  • To provide optimum care in terms of INR control.
  • To provide optimum care in terms of patient satisfaction, e.g. accessibility, waiting time, and continuity of care.

Which patients will it cover?

All patients receiving warfarin therapy are offered anticoagulation management in the practice nurse-led clinic. Those excluded include:

  • Patients with poor control, i.e. consistently unstable INR results.
  • Domiciliary patients (practices will need to make a decision regarding patients requiring home visiting).
  • Patients who prefer to continue with hospital care.

Indications for warfarin

Table 1 (below) lists the indications for warfarin use and target INRs.

Table 1: Indications for warfarin use and target INR*
Target INR 2.5
  • Pulmonary embolus
  • Proximal deep vein thrombosis
  • Calf vein thrombosis
  • Recurrence of venous thromboembolism
  • Non-rheumatic AF
  • AF other causes
  • Mural thrombus
  • Cardiomyopathy
  • Cardioversion
  • Symptomatic inherited thrombophilia
Target INR 3.5
  • Recurrence of venous thromboembolism while on warfarin therapy
  • Antiphospholipid syndrome
  • Mechanical prosthetic valve
Not indicated
  • Bioprosthetic valve
  • Ischaemic stroke without AF
  • Retinal vein occlusion
  • Peripheral arterial thrombosis and grafts
  • Coronary artery thrombosis
  • Coronary artery graft thrombosis
  • Coronary angioplasty and stents
*These indications and targets are taken from British Society of Haematology guidelines, as published in British Journal of Haematology 1998; 101: 374-87.


The clinic should only be managed by trained staff 'authorised users'. Personnel responsible for the clinic should be aware of professional accountability and undertake the clinic management only if they feel competent to do so.

Short courses are provided regularly at the Department of Primary Care, Birmingham University, and an accredited course with formal assessment is being developed. Other methods of training may be available through the local haematology department.

Training should include:

  • An introduction to oral anticoagulation therapy
  • An understanding of the test to be performed
  • The INR and how it is derived
  • An understanding of the specific NPT method for deriving INR
  • Setting up the instrument
  • The target INR: how it relates to diagnosis and action to take if results are outside limits
  • All results must be recorded fully with patient ID, operator ID, and batch number of reagents and quality control material.
  • Faults, maintenance and repair must be recorded in an NPT log book.
  • Health and safety issues – disposal of sharps, Control of Substances Hazardous to Health (COSHH) regulations.

Management of the clinic

There must be appropriate storage facilities for equipment and reagents, e.g. access to constant refrigerated storage space in an appropriate area.

When deciding which NPT to purchase, the factors listed in Table 2 (below) need to be considered.

Table 2: Factors to consider when purchasing a near-patient INR test


  • Internal: How consistent are the results of a test on a day-to-day basis?
  • External: How do the results compare with an external standard?
  • Some coagulometers need regular calibration using standardised samples, whereas in others the calibration is built into the technology or accomplished by the input of a predetermined code for a new set of reagents. The complexity of the calibration method is a major influence on the degree of staff training.

Potential operator variables

  • NPTs are particularly prone to operator-dependent errors as they may be performed in many different situations with the potential for use in uncontrolled circumstances. The more steps involved the greater the potential variability in results

Acceptability to patients

  • The degree of invasiveness of a test needs to be acceptable to the patient. Finger prick sampling may seem less invasive than venous blood sampling


  • Time taken to train staff and perform each test
  • Cost of quality control material and external assurance scheme
  • Cost of technology and reagents


  • Does the manufacturer supply a service contract and warranty?

Health and safety

  • Cleaning and disinfection/cross infection between patients
  • Is the equipment for multipatient use with minimal risk of cross-infection?

Portability of equipment

  • Is the equipment appropriate for home visiting?

Supporting quality assurance schemes

  • Does the manufacturer provide internal quality control; is it simple and quick to perform?
  • External quality assurance is provided by collaboration with a hospital laboratory and should be performed regularly. The practice will need to join a quality assurance scheme to maintain this

Liaison with the local laboratory will be required:

  • For support with training of staff
  • For external quality assurance
  • For referral of patients back to hospital care
  • For referral of patients who are difficult to control to specialist care
  • If there are any concerns regarding NPT performance or results.

Identifying patients

The practice computer database will identify most patients. Patients can also be identified from warfarin prescriptions.

Hospital anticoagulant clinics will need to be informed of all patients attending the practice clinic. A letter should be sent to the hospital clinic to inform them if the practice clinic is going to take over the anticoagulation monitoring.

Similarly the hospital clinic must be informed if patients need to be referred back to hospital. The referral letter should include clinical indication, recommended period on warfarin and target INR.

Patients can attend the practice clinic once their INR has been within the therapeutic range on two occasions. The hospital clinic can then discharge the patient into GP care if appropriate arrangements have been made.

Setting up the clinic

Allocate protected practice nurse/ GP time. A practice nurse trained in anticoagulation management can manage the clinic with support from an interested GP.

Send written invitations to new patients to attend the anticoagulant clinic, and inform hospital clinics currently managing the patients of the practice clinic. Inform the primary healthcare team about the clinic and ask them to identify any new patients on warfarin.

The INR will be estimated using NPT equipment, and the INR result will be interpreted using CDSS.

Perform internal quality control at the start of each clinic using material supplied by the manufacturer. Perform external quality control every 3 months using samples from an external quality assessment scheme, e.g. National External Quality Assessment Scheme (UK) (NEQAS).

Organise support for the clinic from the local haematology department laboratory staff and consultant haematologist.

Adhere to health and safety procedures recommended by laboratory staff to protect both patients and clinical staff at all times.

Ordering equipment and documenting procedures

  • Record the name of the NPT and the serial number used within the clinic and for home visits.
  • Record the name and lot number of reagent strips and internal quality ordered directly from the manufacturer and maintain under refrigerated conditions.
  • Record the name and telephone number of manufacturer.
  • Obtain membership to external quality control scheme .
  • Obtain documentation log book for recording control data.
  • Record CDSS and laptop computer serial numbers.
  • Obtain finger-pricking device, lancets, disposable gloves and sterilising fluid for cleaning machine and mopping up spilled hazardous waste.
  • Obtain patient hand-held record books.

Clinic procedure

Prepare NPT and complete internal quality control procedure. Document control result, batch numbers etc.

Counsel the patient regarding the clinic process and check for:

  • Bleeding or thrombotic incidences
  • Tablet compliance and change of medication
  • Lifestyle changes, e.g. alcohol binges.

Assess the patient's condition requiring warfarin, i.e. risk vs benefit: does the patient require annual specialist or GP follow-up? Check the recommended period of time on warfarin.

Assess whether warfarin therapy is still appropriate, e.g. presence of dementia, multiple adverse drug reactions.

Perform a blood test using capillary blood. Venous samples can be taken at the patient's request or for capillary results greater than NPT upper limits. Perform the INR test using NPT and enter results into the CDSS.

Follow suggestions given by the computer for dosing and recall dates unless clinically inappropriate, e.g. patient known to be non-compliant with therapy. Complete patient record card and give to patient with verbal instruction regarding dosage and recall.

Record the INR, warfarin dosage and recall date in the patient's notes and practice computer. Back up the computer software taken at the end of each clinic.

Clinic performance

ýellevue was awarded Beacon status in 1999 for its work in developing anticoagulation services. The clinic is run according to a protocol (see Figure 2, below) developed from a series of anticoagulation study days which we run to train practice nurses and GPs.

Figure 2: Protocol for the Bellevue Medical Centre nurse-led primary care anticoagulation clinic
protocol for the nurse-led primary care anticoagulation clinic

We have run six open days as part of the Beacon service and have also presented the clinic model at a regional open day. The model's effectiveness compared with routine hospital outpatient clinics was established as part of an MRC-funded randomised controlled trial in 12 Birmingham general practices.2

Routine data from regular audit of the clinic has demonstrated that patients spend around 70% of time within the therapeutic range with serious adverse events running at approximately 0.1% per annum.3

The Bellevue clinic caters only for Bellevue patients. We currently manage around 45 patients; this is slightly fewer than usual as the list size has decreased because of local redevelopments. Approximately 60% of these patients have AF as their primafy indication; the second largest group comprises patients with mechanical heart valve replacements.

Generally all adult patients are managed through the clinic, with paediatric patients being monitored through the local children's hospital. Occasionally it becomes necessary to refer patients back to hospital clinics, in a similar way that patients with brittle diabetes are referred for specialist intervention.

Over a 3-year period to the end of 2000, two patients were referred back to the hospital clinic, the first because of poor compliance and the second because he had a mitral valve replacement and was highly unstable.

The future

There has been some discussion around providing the service for surrounding practices, particularly within the local primary care group (PCG). We have mixed feelings about this 'hub and spoke' model, which has the potential to reintroduce all the problXms of hospital outpatient clinics, such as long waiting times and impersonal service.

One of the major advantages of a primary care model is the personal continuity of care given by the practice nurse, with all the benefits of a holistic approach to the patient. While this is not impossible to provide within a hub and spoke system, it is much less likely. The real problem is how to fund the service when starting from scratch. Currently, PCGs seem to offer the main source of new funding.

The Birmingham model of oral anticoagulation therefore offers a safe and effective alternative to hospital outpatient clinics. It is, however, more expensive from the NHS perspective. Patient satisfaction with the model is high and if patient costs are taken into account the Birmingham model becomes cost neutral.

While this model is becoming increasingly widespread, it is not suitable for all situations. Small practices will not find it an attractive proposition, and practices with easy access to alternative testing sites would also not find it attractive. It does, however, offer a solution for those regions where secondary care services are already swamped.

The next evolution is likely to be patient self-management where patients measure their own INRs on an NPT and adjust their dose according to a simple algorithm. This model is already widespread in Germany and is currently being evaluated in the UK.

  • This article is based on the document A standard operating procedure (SOP) for a primary care anticoagulation clinic using near-patient testing (NPT) and computerised decision-support software (CDDS), compiled by Ellen Murray, David Fitzmaurice and Patrick Kesteven (consultant haematologist, Freeman Hospital, Newcastle upon Tyne). Copies are available from Dr David Fitzmaurice or Ellen Murray, Department of Primary Care and General Practice, Division of Primary Care, The Medical School, University of Birmingham, Edgbaston, Birmingham B15 2TT.


  1. DA Fitzmaurice, FDR Hobbs, ET Murray et al. Evaluation of computerized decision support for oral anticoagulation management based in primary care. Br J Gen Pract 1996; 46: 533-55.
  2. DA Fitzmaurice, FDR Hobbs, ET Murray et al. Oral anticoagulation management in primary care with the use of computerized decision support and near-patient testing. Randomized, controlled trial. Arch Intern Med 2000; 160: 2343-8.
  3. DA Fitzmaurice, FDR Hobbs, ET Murray. Primary care anticoagulant clinic management using computerized decision support and near patient International Normalized Ratio (INR) testing: routine data from a practice nurse-led clinic. Fam Pract 1998; 15: 144-6.

NHS Beacon Awards
Beacon status is awarded to practices, trusts and other healthcare organisations within the NHS that have demonstrated good practice. The NHS Beacon programme aims to spread best practice across the health service. For further information visit the website: www.nhs.uk/beacons


Guidelines in Practice, August 2001, Volume 4(8)
© 2001 MGP Ltd
further information | subscribe