Automated detection of wrong-drug prescribing errors.

BACKGROUND: To assess the specificity of an algorithm designed to detect look-alike/sound-alike (LASA) medication prescribing errors in electronic health record (EHR) data. SETTING: Urban, academic medical centre, comprising a 495-bed hospital and outpatient clinic running on the Cerner EHR. We extracted 8 years of medication orders and diagnostic claims. We licensed a database of medication indications, refined it and merged it with the medication data. We developed an algorithm that triggered for LASA errors based on name similarity, the frequency with which a patient received a medication and whether the medication was justified by a diagnostic claim. We stratified triggers by similarity. Two clinicians reviewed a sample of charts for the presence of a true error, with disagreements resolved by a third reviewer. We computed specificity, positive predictive value (PPV) and yield. RESULTS: The algorithm analysed 488 481 orders and generated 2404 triggers (0.5% rate). Clinicians reviewed 506 cases and confirmed the presence of 61 errors, for an overall PPV of 12.1% (95% CI 10.7% to 13.5%). It was not possible to measure sensitivity or the false-negative rate. The specificity of the algorithm varied as a function of name similarity and whether the intended and dispensed drugs shared the same route of administration. CONCLUSION: Automated detection of LASA medication errors is feasible and can reveal errors not currently detected by other means. Real-time error detection is not possible with the current system, the main barrier being the real-time availability of accurate diagnostic information. Further development should replicate this analysis in other health systems and on a larger set of medications and should decrease clinician time spent reviewing false-positive triggers by increasing specificity.

Automated detection of look-alike/sound-alike medication errors.

PURPOSE: The development and evaluation of an algorithm for detecting potential medication errors due to look-alike/sound-alike (LASA) drug names are described. SUMMARY: A computer algorithm that detects potential LASA errors by analyzing medication orders and diagnostic claims data was developed. The algorithm flags a potential error when (1) a medication order is not justified by a diagnosis documented in the patient's record, (2) another medication whose orthographic similarity to the index drug exceeds a specified threshold exists, and (3) the latter drug has an indication that matches an active documented diagnosis. A review of medication orders and diagnostic claims at a large health system identified cases in which cycloserine was ordered but cyclosporine was the intended treatment. Subsequent review of all cycloserine orders over a 7-year period indicated that 11 of 16 orders were erroneous, prompting placement of an alert regarding the potential for LASA errors involving cycloserine and cyclosporine in the electronic order-entry system. Automated detection and confirmation of LASA errors via chart review can be used retrospectively to identify problematic pairs of drug names and to assess associated error rates within a healthcare system. The same techniques can be used to prevent errors in real time through indication alerts if accurate diagnostic information is available at the time of order entry. CONCLUSION: Automated methods involving the use of medication orders, diagnostic claims, and indications can be used to detect and prevent LASA errors.