Key result
Pharmacist-directed EHR alerts show no benefit over prescriber alerts for correcting inappropriate DOAC dosing.
Why the study?
Inappropriate DOAC dosing is common and linked to adverse outcomes, but understanding why errors arise after initial prescribing and how clinicians respond to alerts remains unclear.
Does routing asynchronous EHR notifications about inappropriate DOAC prescriptions to clinical pharmacists improve dose correction rates compared to original prescribers in ambulatory patients?
RCT (n=361)
Cluster-randomized at the prescribing clinician level
No
Does routing asynchronous EHR notifications about inappropriate DOAC prescriptions to clinical pharmacists improve dose correction rates compared to original prescribers in ambulatory patients?
EHR notifications for inappropriate DOAC dosing resulted in dose corrections in about one-third of cases, with pharmacists significantly more likely to document reasons for overriding alerts than prescribers.
Asynchronous EHR alerts improve DOAC dosing corrections without added benefit from pharmacist routing; challenges assumptions favoring pharmacist-directed interventions.
Background: Inappropriate dosing occurs in up to 20% of direct oral anticoagulant (DOAC) prescriptions and is associated with bleeding, hospitalization, and death. Understanding why DOAC dosing errors arise after the initial prescribing period and how clinicians decide to adjust doses may help improve future antithrombotic stewardship efforts. Objectives: This study aimed to describe responses by clinicians to asynchronous electronic health record notifications regarding inappropriate DOAC prescriptions that develop after the initial prescribing period. Methods: As part of a larger randomized trial to improve evidence-based DOAC prescribing, we conducted a manual chart review to assess reasons for inappropriate DOAC prescriptions, clinician response, and reasons for not changing DOAC prescriptions. Results: Most apixaban prescribing errors occurred in patients aged 80 years or older (87.8%). Overall, dose changes occurred after 32.7% of notifications. Notifications to increase the dose of apixaban resulted in fewer dose changes (16.8%) than those to reduce the dose of apixaban (52%), increase the dose of rivaroxaban (43.2%), or decrease the dose of rivaroxaban (43.6%). The original DOAC prescribers were less likely than pharmacists to document a reason for not making a dose change (46.1% vs 90.5%). Renal function, weight, and bleeding-related reasons were commonly cited as reasons not to make a dose change. Conclusion: Various patient factors differentially influenced the likelihood that a clinician would change an inappropriate DOAC prescription. These factors can be used to tailor electronic health record-based notifications to reduce the burden on prescribers while maximizing the impact on patient safety by increasing evidence-based prescribing.
No takes yet. Share an insight, caveat, or question.
Saraswat et al. (2026) conducted an RCT in Inappropriate DOAC prescriptions (n=361). EHR notification routed to clinical pharmacist vs. EHR notification routed to original prescriber was evaluated on Dose change following incorrect DOAC dose notification within 90 days. Asynchronous EHR notifications for inappropriate DOAC prescriptions resulted in dose changes in 32.7% of cases, with similar rates between pharmacist-directed (31%) and prescriber-directed (34.6%) alerts.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: