research agenda on the safe and appropriate use of medications across all care settings, and Congress should allocate the funds necessary to carry out this agenda. This agenda should encompass research methodologies, incidence rates by type and severity, costs of medication errors, reporting systems, and in particular, further testing of error prevention strategies.
Researchers use a variety of definitions for medication errors, near misses, and adverse drug events (ADEs). There is a clear need to standardize terminology and measures for these phenomena if the field is going to advance.
Much of the literature reviewed by the committee failed to specify in sufficient detail the definition of a medication error used in the study. When detail was given, the definitions varied widely. Studies of medication errors in pediatric populations illustrate the problem: in one study, the definition of a medication dosing error was any dose greater than 10 percent over the recommended dose, while in another study, the definition was any dose greater than 10 times the recommended dose. In another set of studies examining the entire medication delivery system, some research teams did not include as errors events that were detected before they reached the patient, whereas other researchers counted these events as errors. As another example, some studies counted an order that was lacking a prescriber’s signature as a medication error; although this is clearly an error, the potential for harm to patients is substantially different from that resulting from orders with dosage errors. If the definitions of medication errors and their subtypes are not standardized, and outcome data are not collected in a manner that permits assessment of the potential for harm, it is difficult to compare incidence rates across studies and to identify areas of high priority for intervention. The existing variations also result in a broad range of reported frequency and severity of errors.
To address this issue, the committee recommends that an international consensus conference be held to define the terms medication error, near miss, and ADE, as well as the subtypes of these terms, and to identify the practical applications of these definitions in various care settings. This conference could be similar to a session organized by the American College of Chest Physicians/Society of Critical Care Medicine in August 1991 to agree on a set of definitions that could be applied to patients with sepsis and its sequelae (Bone et al., 1992). The definitions introduced as a result of this conference have been widely used in practice and have served as the foundation for inclusion criteria for many clinical trials of therapeutic interventions (Levy et al., 2003). This uniform set of definitions has also made it