User interaction metadata for healthcare information systems

Semantically consistent understanding of data/content in information systems is a key issue in collaborative work settings. Even simple data -- such as timestamps -- can be a source of misunderstandings both in data entry and reporting tasks. Healthcare information systems generate timestamps during data entry situations. Timestamped data are used for various purposes, such as efficiency calculations, organizational development, and research. However, timestamps are often generated ambiguously in different ways: they are a source of semantic interpretation errors. In this paper we illustrate the phenomena of timestamp ambiguity in medical settings. We suggest that User Interaction Metadata (UIM) could reduce the problems caused by ambiguous and erroneous timestamps. UIM may reveal hidden heterogeneity in data interpretation. Additionally, it can be used to identify user behavior patterns that are currently unrecognizable from raw transaction data.