Comparing the usability and inter-rater reliability of a generic and a domain-specific medical error taxonomy

Medical error taxonomies are used to report or analyse medical errors. Medical error taxonomies have categories that filter and represent information found in incident reports. Different taxonomies with different categories may provide different information about a medical incident. Generic medical error taxonomies, such as the Patient Safety Event Taxonomy, and domain-specific medical error taxonomies, such as the NCC MERP Taxonomy of Medication Errors, have categories with different levels of granularity. It is not known if the granularity of categories in a medical error taxonomy affects the taxonomy&s;s usability and inter-rater reliability. We performed a study to compare the usability and inter-rater reliability of a generic and a domain-specific taxonomy. Ten nurses classified a medication error case using the Patient Safety Event Taxonomy while another ten nurses classified the same case using the NCC MERP Taxonomy of Medication Errors. All nurses then rated on a Likert scale how easy it was to classify the case using their assigned taxonomy. The inter-rater reliability of the nurses’ classifications was measured using raw agreement. It was found that the nurses rated the taxonomies as equivalent in terms of usability. In terms of inter-rater reliability, both taxonomies demonstrated low inter-rater reliability (