JMIR Publications

ABSTRACT

Background:

Care teams and collaboration are vital within healthcare institutions, allowing for the effective use of collective health care worker (HCW) expertise and resources. Human-computer interactions (HCI) involving electronic health records (EHRs) have become pervasive and act as an avenue of quantifying these collaborations using statistical and machine learning methods.

Objective:

The aim of this study is to develop an informatics framework measuring HCW collaboration and its characteristics through the concurrent usage of EHRs.

Methods:

Extracting concurrent EHR usage events from audit log data, we define and quantify concurrent EHR sessions. For each HCW, we calculate a novel metric called collaboration intensity, which is derived on a daily basis through the proportion of their EHR activities in concurrent sessions over all their EHR activities and EHR time spent on those activities. Statistical models are used to test the differences in the collaboration intensity between HCWs with different specialties (eg, physicians, residents, nurses). For each patient visit, starting from admission to discharge, we measure temporally concurrent EHR usages, which we call temporal collaboration patterns. Again, we apply statistical models to test the differences in temporal collaboration patterns of the admission, discharge, and intermediate days of hospital stays between weekdays and weekends. Network analysis is leveraged to measure collaborative relationships among HCWs from concurrent sessions. We survey experts to validate the identified collaborative relationships. Principal component analysis (PCA), t-distributed stochastic neighbor embedding (T-SNE), and K-means clustering are used to aggregate collaborative activities to describe concurrent EHR sessions. We gathered 4 months of EHR audit log data from a large academic medical center neonatal intensive care unit (NICU) to validate the effectiveness of our framework.

Results:

There exists a significant difference in the collaboration intensity between the top 13 HCW specialties who have the largest amount of time spent in EHRs (p < .001). The temporal collaboration patterns between weekday and weekend periods are significantly different on admission (p<.001) and discharge days (p<.001), but not during intermediate days of hospital stays. There is more collaboration on admissions and discharges on weekdays, and less so on weekends (p<.001). Neonatal nurses, fellows, and frontline providers, neonatologists, consultants, respiratory therapists, and ancillary and support staff often collaborate using EHRs. NICU professionals confirmed the identified collaborative relationships (p < .001). We identified 50 clusters of collaborative activities. Over 87% of concurrent sessions can be described by a single cluster, with the remaining 13% of sessions comprising multiple clusters.

Conclusions:

Leveraging concurrent EHR usage through audit logs to analyze HCW collaboration may improve our understanding of collaborative patient care, given the collaboration characteristics we describe. HCWs collaborating using EHRs can potentially affect the quality of patient care, discharge timeliness, and clinician workload, stress, or burnout.