Using name overlap analysis to understand medication name search safety.

This article examines the extent of left-to-right character overlap in medication names within real-world computer systems and its impact on safely searching for medications by name. Using automated analysis of 20,020 randomly generated medication name lists drawn from a database of 2,249 unique names, the study found that the number of keystrokes required to uniquely identify a medication varies widely, ranging from 3 to 29 characters depending on name overlap. For high-alert or "never" medications—those that must never be administered in error—the overlap required 3 to 10 keystrokes, often more than the fixed minimum of five characters currently recommended by the Institute for Safe Medication Practices (ISMP). The authors recommend upgrading medication search software to support dynamic, incremental searching that adapts to the specific list being searched, standardizing medication names to minimize overlap and special characters, and requiring additional user input (such as indication or dosage form) when selecting high-risk medications to improve safety.