Creation, validation, and assessment of a new measurement tool designed to detect tension in electronic problem lists

A systematic literature review has highlighted many best practices for using the electronic problem list effectively. However, it also indicated that there is a lack of metrics available for describing changes in the way clinicians use the problem list. To address that shortcoming, a new measurement tool was built in collaboration with clinicians from three different local healthcare systems. A reference standard dataset was built and used to validate the tool using training and testing data methodologies. The results of that process led to thresholds and detection criteria to be used in assessing problem lists for changes that skew the data out of alignment with a patient's true health status (Tension). Those criteria and thresholds were used in conjunction with production clinical data to determine if the instrument could measure Tension and whether a new problem list intervention had any effect on Tension when compared to the legacy system. This research has shown that Tension can be accurately measured with this instrument and that the intervention did reduce the overall level of Tension.