Testing a classification model for emergency department errors

This paper presents an evaluation of the Eindhoven Classification Model for categorizing healthcare errors.

The ability to decrease healthcare errors will depend on an understanding of the types and patterns of error that occur in various settings. Research addressing the systematic classification of error is in its infancy.

The sample for this study was two existing sets of healthcare error reports obtained from the emergency department of a regional trauma center. Study data were collected in the summer of 2001 and 2002. The errors in these reports were categorized using a coding tool based on the Eindhoven Classification Model. In addition to testing for inter-rater reliability, the classification model was also evaluated for the number of unclassifiable errors and of categories never selected by the raters (not useful to the error classification model).

Our findings of poor inter-rater reliability, large numbers of unclassifiable errors and categories not selected, suggest that the Eindhoven Classification Model, in its current form, cannot be applied to categorizing healthcare errors in an emergency department setting.

Further study is needed to develop and test error categorization models for use in emergency departments and other healthcare settings.