Warfarin remains the most widely prescribed oral anticoagulant drug, with over 6% of Irish patients >80 years now receiving this agent.1 It requires careful monitoring of INR to reduce the risk of both bleeding and thromboembolism. Traditionally, warfarin monitoring has been largely hospital-based; however there has been a move towards anticoagulation clinics in the primary care setting. Access to warfarin monitoring in the community brings patient benefits, including accessibility, reduced waiting times, and improved patient involvement and education, while reducing strain on hospital services. It also brings added responsibility to GPs to manage the prescribing of this potentially dangerous drug.
Our aim was to audit current practice of an anticoagulation service in a general practice in Dublin inner city and compare it with the ICGP Guidelines on Warfarin in General Practice.2 We also wanted to audit the performance of the community anticoagulation service and a nearby hospital anticoagulation service and compare to international guidelines.
A search on the clinic software program, Socrates, identified a total of 65 patients currently being prescribed warfarin. All 65 patient charts were examined, and a data collection sheet recorded whether the clinical indications, duration of therapy, loading regime, dosage, prothrombin time target range and interval for repetition of INR were clearly documented, as per ICGP guidelines. Whether patients were attending the practice anticoagulation clinic, or hospital clinic was also documented.
This audit was complicated by the fact that DAWN, a separate, computer-assisted dosing program, was introduced to the practice in early 2012. This system was also examined to see if it was adherent to ICGP guidelines as previously described.
Patient INR results over a six-month period from December 1, 2011 to May 31, 2012 were collected from the Socrates program in the clinic, the DAWN program and hospital EPR program and recorded on the data collection sheet. Patients who had been an inpatient in the hospital at any time between December 1, 2011 and May 5, 2012, or who had had a day case procedure requiring the cessation of warfarin, eg. colonoscopy, were excluded. The time in therapeutic range (TTR) for each patient was calculated using the Rosendaal method. International guidelines recommend a TTR of ≥60%.3 The TTR and number of visits were recorded on the data collection sheet.
Of the 65 patients prescribed warfarin in the practice, 37 (57%) were female and 28 (43%) were male. Average patient age was 71 +/- 14. Twenty-five (38%) patients attend the community anticoagulation clinic while 40 (62%) patients still attend the nearby hospital clinic. Of these 40 patients, four are engaged in a self-testing programme.
All patients had the clinical indication, duration of therapy, loading regime, dosage, prothrombin time target range and interval for repetition of INR clearly documented in the DAWN program. However, none of the patients had all of the clinical information to satisfy ICGP criteria clearly documented and accessible in their notes on the Socrates program.
Sixteen patients were excluded from this part of the audit. Fifteen of these had either an inpatient stay or procedure which required the cessation of warfarin, and information was not available for one patient. Of the 49 patients studied, 18 patients attend the community anticoagulation clinic, 27 attend the hospital clinic and four patients are self-testing. We found that 78% of patients attending the community anticoagulation clinic (n=18) had a TTR ≥60%, 82% of patients attending the hospital anticoagulation clinic (n=27) had a TTR ≥60%, while 100% (n=4) of patients who were self-testing had a TTR ≥60%.
Patients who attend the community clinic had an average of 7 +/-4 INR checks over a six month period while patients who attend the hospital clinic had their INR checked an average of 8 +/-4 times. Patients who self-test had their INR checked on average 15 +/-4 times, however, this can be partly explained by a different standard of practice for self-testing patients, requiring at lease one INR check a month.
This audit highlighted the inevitable challenges that arise on the introduction of new computer programs/practices to a pre-existing service. While there is no doubt that the introduction of the DAWN computer-assisted dosing program allows more accurate and safer management of patient warfarin levels, currently, only the practice warfarin nurse has access to this program. Therefore, while it is reassuring that 100% of patients on the DAWN program were adherent to ICGP guidelines, there remains a need for this information to also be clearly documented on the clinic Socrates program, which is accessible to all medical and nursing staff in the practice.
Clinic performance results showed a similar percentage of patients reaching a TTR of ≥60% in hospital and community anticoagulation clinics. In addition, there was little difference in the number of INR checks over a six month period between the clinics. Overall, these results are positive, showing a quantitative non-inferiority of the community anticoagulation clinic, which together with the qualitative benefits shown in a 2001 study completed by the RCSI Department of General Practice, supports the overall benefit of the community clinic to patient care and quality of life.4
As a result of this audit, patient charts on Socrates will be updated so as to be in concordance with ICGP guidelines. While clinic performance was satisfactory, there is potential for improvement, and TTR may be increased with the use of the DAWN computer-assisted dosing program. A reaudit is planned to examine the implementation of changes and complete the audit cycle.
- Quirke W, Cahill M, Perera K, Sargent J, Conway J. Warfarin prevalence, indications for use and haemorrhagic events. Irish Medical Journal 2007; 100(3): 402-404
- Kildea-Shine P, O Riordan M. Warfarin in General Practice. 2006; www.icgp.ie
- Baglin TP, Keeling DM, Watson HG. Guidelines on oral anticoagulation (warfarin): Third edition – 2005 update. British Journal of Haematology 2006; 132: 277-285
- Clune B, McGrogan K, O Connor A. GP-led Near Patient Testing Community Anticoagulation Clinics: A Working Model