Assessing safety of inpatient direct oral anticoagulant prescribing utilising an electronic prescribing system

Abstract

Abstract Introduction Direct oral anticoagulants (DOACs) are favoured over oral vitamin K antagonists (VKA) due to their fixed-dose regimen and reduced thromboembolic and bleeding risk. Despite clear dose adjustments based on patient characteristics, several observational studies have demonstrated 15-20% of patients being overdosed or underdosed when compared to licensing.1,2 Inappropriate dosing can lead to harm e.g., bleeding or thrombosis which would prolong/escalate hospital stay. Aim To assess clinical appropriateness of inpatient DOAC prescriptions across multiple sites of a large London based NHS Trust. Methods This study was conducted retrospectively over a seven month-period across 4 sites from a large London based NHS Trust from November 2021. Electronic prescribing system was used to generate daily reports that enabled a specialist haematology pharmacist to assess all adult inpatient DOAC prescriptions for appropriateness as per product licensing for indication. The report included patient characteristics (age, gender, body weight), prescribed DOAC (agent, dose, indication), serum creatinine, hospital site. Queries were escalated for review by the pharmacy or medical teams. Type of intervention (dose, agent, or no change) was recorded, and allocated a clinical severity rating using IMPACCTS.3 This scale runs from a score of 1 indicating a good practice intervention to a score of 5 that prevents serious or major harm including death. This study was approved by the University College London (UCL) School of Pharmacy research ethics committee. Results A total of 1,761 inpatient DOAC prescriptions were reviewed, of which 10.1% had a clinical query requiring escalation. Results demonstrate that 77.0% (n=137) of all queries were dose-related, 33.2% overdosed and 43.8% underdosed compared to licensing, the most common agent requiring adjustment being apixaban. Renal-related queries were most frequently observed (50.6%), with 12.8% of patients on edoxaban having CrCl >110ml/min. We found that 2.3% of prescription queries raised required review and stopping antiplatelet agents co prescribed with anticoagulation. Most interventions made had a severity rating of 4 (74.2%), followed by 15.2%, and 10.7% had a score of 1 and 3, respectively. Following pharmacist recommendations, changes made were either dose (38.7 %) or agent (14.6%). Alternatively, it was not applicable due to a change in clinical status, i.e., renal function improved (1.1%) or patient re-weighed (6.2%). Changes to prescriptions did not apply to some patients as documentation of clinical reasoning were recorded (19.7%), or due to patients being discharged (13.5%). Discussion/Conclusion This study has demonstrated a significant number of DOAC prescriptions are inappropriate at the point of prescribing when compared to product summary of characteristics for the indications listed. The use of a centralised report to assess appropriateness of DOAC prescribing across several sites has facilitated a centralised mechanism led by a haematolgy specialist to improve safer DOAC prescribing. The most frequent interventions being made included dose amendments based on calculated creatinine clearance and amending (or adding) indications entered on the system. Further work is required to demonstrate effective change management strategies that have been implemented to further improve the safety of DOAC prescribing within the Trust.