From Mr G. N. Pimblett, MRPharmS
SIR,-The Journal recently ran a report of a study comparing clinic management and self-management of anticoagulant monitoring (PJ, July 22, p122). Since 1997, a number of pharmacist-led primary care clinics have been operating in Sefton and St Helens and Knowsley health authority areas on Merseyside. In total these have been providing anticoagulant monitoring to approximately 700 patients during 1999 and 2000 and have audited the data on INR (international normalised ratio) control over this period. Furthermore, a local trust has a pharmacist-led monitoring service that includes a satellite clinic in the community. This clinic probably deals with a further 200-300 patients. INR control from that clinic is similarly audited.
The conclusion of the reported study was that self-management was as effective as clinic management and better accepted by patients. I do not have the study to evaluate it in detail but feel that there are several points worth raising on the strength of The Journal report alone.
Patients in the reported study were assessed every one or two weeks. It is not clear whether dosage adjustments were made on the basis of these INR measurements. However, would it be ethical to allow a patient to continue on a dose that was likely to increase the risk of a serious haemorrhage or major thrombo-embolic event? Probably not. I assume, therefore, that dosage adjustment was made where appropriate. Furthermore, most health care professionals involved in anticoagulant monitoring would probably feel that the numbers of patients in range would increase with frequency of monitoring. If pharmacists operating local clinics saw each of their patients every two weeks I am sure they would expect to achieve significantly better INR control than is possible with extended intervals between visits.
The study length and design of the study are dubious. Patients on anticoagulants frequently have co-morbidity that is subject to seasonal change. This means that patients studied during the winter are more likely to exhibit poorer control of INR than those studied during the summer. The effect of co-morbidity is exacerbated by the possibility of drug interactions - antibiotics, many of which have the potential to interfere with INR control, are more frequently prescribed in winter. A cross-over study designed to run over six months (three months in each arm) may therefore be affected by the time of year the data in each arm is being collected. Audit data from our clinics shows that INR control is less stable over the winter/Christmas months.
The self-management arm of the reported study achieved acceptable INR control of 52 per cent compared with 49 per cent in the clinic management group. Within South Sefton the INR control is in the range 60-65 per cent. The Kirkby clinics and Formby clinics control is of a similar standard.
This level of INR control has been maintained over the past 12 months against a background of increasing caseload. This compares favourably with the results from this study.
The numbers of patients enrolled in this study was very low. With only 50 patients it is difficult to assign any great significance to any the outcomes mentioned. The four South Sefton sites manage a caseload of about 400-450 patients. When the Kirkby clinics and the Formby site are included this must rise to approximately 1,000 patients. Thus there exists audited data for about 1,000 patient years of monitoring in 1999 showing a consistent level of INR control above that demonstrated for either arm of this study. With data from 1998 and 2000 this figure must now be approaching 2,500 patient years of monitoring.
The duration of the study was only three months. Current national recommendations allow for an interval of up to 12 weeks between appointments for stable patients (in ‘Guidelines on oral anticoagulation', third edition, by the Thrombosis and Haemostasis Task Force). This is substantially more than the one- to two-week interval allowed in this study. The average interval between visits at the South Sefton clinics is approximately six weeks.
The information on patient satisfaction is inevitable. Within South Sefton pharmacists see approximately 30 patients per session. This compares favourably with the 250 patients who have to attend the local trust service. It is unsurprising that patients prefer the local or home-based service. Logistical problems with the enormous numbers of patients involved make this inevitable. Patients experience long delays minimised only by the hard work and dedication of the staff. Given this environment, few patients would prefer it to a model with less stress and waiting. In the primary care clinics, however, patients are seen and advised on appropriate doses in a relatively short time. Waiting times are likely to feature in a forthcoming audit.
There is no information about the educational standards of the patients enrolled in the study. Well-educated, relatively young patients might well be able to self-manage anticoagulation. Patients in areas of high deprivation, the elderly or patients of low educational achievement may have greater difficulties coping with the complexity of anticoagulant pharmacokinetics. Many would find funding the near patient testing machines impossible.
I feel that self-management of anticoagulant monitoring is a viable option in a relatively small number of patients. Given the strain on UK trust based services there is little in the way of professional support available for those embarking on it.
I have seen a figure of 40,000 quoted for self-managing patients in Germany. Given that the UK has an approximately similar population to Germany and that figures for anticoagulated patients in the UK are 500,000 to 1,000,000 this represents only 5-10 per cent of the "market". The other 90 per cent of the market presumably require "traditional" anticoagulant monitoring services.
Graham Pimblett
Chairman, Local Anticoagulant Pharmacists' Group, Liverpool