Merging HIS- and RTLS-data to Strengthen Work Organization Analysis in Nursing

Abstract

Nursing work is under pressure due to rising care demands and limited staffing. Hiring more nurses is costly and challenging, which makes it essential to improve their work organization. To identify improvement areas, analyzing how nurses currently work can be very insightful. This study will use process data for work organization analysis that does not put an extra registration burden on nurses.

In most hospitals, the primary source for recording nursing tasks is the Hospital Information System (HIS). However, records in HIS-data can deviate from actual nursing work: several tasks are not recorded, and those that are recorded may include incomplete or inaccurate information. For example, documentation times can differ significantly from the actual execution time. Such discrepancies create gaps and delays that render work organization analysis less reliable when HIS data is used in isolation (Vanthienen et al. [2025]). To address these limitations, alternative sources, such as Real-Time Location System (RTLS) data, have been explored (Fernández-Llatas et al. [2015]). RTLS traces reveal when resources enter or exit specific locations. However, they do not indicate which task was performed. As both HIS- and RTLS-data provide a partial view of nursing work, investigating their integrated use is worthwhile to obtain a more accurate view of task start and end times, locations, and resources (Martin [2018]).

Because integrating HIS- and RTLS-data is non-trivial, this study proposes FINTR, a Footprint-based Integration method for Nursing Task Retrieval. FINTR is a semi-automated approach that merges the two data sources using task footprints. A task footprint encodes domain knowledge and expresses how the execution of a type of nursing task is reflected in HIS- and/or RTLS-data. FINTR uses these footprints to guide and constrain the data integration process.

FINTR consists of three main steps. It first performs candidate filtration to remove combinations of HIS entries and RTLS intervals that cannot represent the same task. For the remaining possibilities, it then scores each candidate based on how well it fits the expected temporal and structural patterns encoded in the task footprint. Finally, FINTR selects the most plausible candidate and merges its HIS and RTLS evidence into a single constructed task instance. Thus, FINTR produces a nursing task log where each entry represents a performed nursing task.

Preliminary evaluations were performed using synthetic data, which has the advantage that the ground truth nursing task log is known, to which the output of FINTR can be compared. Two simulation settings have been considered so far: one in which the documentation times in HIS correspond to the moment of task execution and another in which there are documentation delays of up to ten minutes (i.e., tasks are documented in the HIS after they have been performed). Initial results suggest that FINTR performs well, but further testing is needed.

Overall, the findings show that FINTR offers a viable approach to integrating HIS- and RTLS-data to construct a nursing task log from routinely collected digital traces. The method provides a systematic, reusable framework for creating more reliable logs to support downstream analyses. By offering a more accurate and detailed representation of nursing work, FINTR opens new possibilities for understanding and improving the organization of care.

References C. Fernández-Llatas, A. Lizondo, E. Montón, J. M. Benedi, and V. Traver. Process mining methodology for health process tracking using real-time indoor location systems. Sensors, 15(12):29821–29840, 2015. N. Martin. Using indoor location system data to enhance the quality of healthcare event logs: Opportunities and challenges. In International Conference on Business Process Management, pages 226–238. Springer International Publishing, 2018. A. Vanthienen, N. Martin, and B. Depaire. Investigating the (mis)match between electronic health records and actual nursing work: An observational study. International Journal of Nursing Studies, 151:105309, 2025.