LePendu P et al, Clinical Pharmacology & Therapeutics, 2013
With increasing adoption of electronic health records (EHRs), there is an opportunity to use the free-text portion of EHRs for pharmacovigilance. We present novel methods that annotate the unstructured clinical notes and transform them into a deidentified patient–feature matrix encoded using medical terminologies. We demonstrate the use of the resulting high-throughput data for detecting drug–adverse event associations and adverse events associated with drug–drug interactions.
[ More ]
Published:
10 April 2013 |
Keyword(s): Adverse Drug Events, Data Mining, Drug-drug interactions, Electronic Health Records, Free text, Pharmacology, United States
Koplan KE et al, Journal of General Internal Medicine, 2012
BACKGROUND
As electronic health records (EHRs) become widely adopted, alerts and reminders can improve medication safety, but excessive alerts may irritate or overwhelm clinicians, thereby reducing their effectiveness. We developed a novel “stealth” alert in an EHR to improve anticoagulation monitoring for patients prescribed a medication that could interact with warfarin. Instead of alerting the prescribing provider, the system notified a multidisciplinary anticoagulation management service, so that the prescribing clinicians never saw the alerts. We aimed to determine whether these “stealth” alerts increased the frequency of anticoagulation monitoring following the co-prescription of warfarin and a potentially interacting medication.
[ More ]
Published:
30 July 2012 |
Keyword(s): Alert fatigue, Clinical decision support, Drug-drug interactions, Electronic Health Records, Medication Safety, United States
Yu DT et al, Pharmacoepidemiology and Drug Safety, 20(2)
Background
The Food and Drug Administration issues black-box warnings (BBWs) regarding medications with serious risks, yet physician adherence to the warnings is low.
Methods
We evaluated the impact of delivering BBW-based alerts about drug–drug, drug–disease, and drug–laboratory interactions for prescription medications in outpatients in an electronic health record with clinical decision support. We compared the frequency of non-adherence to all BBWs about drug–drug, drug–disease, and drug–laboratory interactions for 30 drugs/drug classes, and by individual drugs/drug groups with BBWs between the pre- and post-intervention periods. We used multivariate analysis to identify independent risk factors for non-adherence to BBWs.
[ More ]
Published:
28 December 2010 |
Keyword(s): Adherence, Alerts, Clinical decision support, Drug-drug interactions, Electronic Health Records, Electronic Prescribing, United States
