Quantity Matters: Association of Advanced Imaging Utilization With the Deployment of a Provider-in-Triage Model

The Provider in Triage (PIT) model is increasingly common in Emergency Department (ED) practice. It has demonstrated positive effects on ED efficiency, though the impact on advanced imaging utilization is less defined. The objective of the current study was to investigate the association of PIT deployment at our academic ED with advanced imaging utilization.

This was a retrospective study at a single tertiary care ED encompassing all patients seen from July 2021 through December 2024, with the PIT process deployed on July 1, 2023. The incidence of advanced imaging for adult ED patients (age 18 or greater) was assessed before and after PIT implementation. This included computed tomography (CT), magnetic resonance imaging (MRI), radiology ultrasound (RUS), and point of care ultrasound (POCUS) studies. Two separate investigators independently retrieved imaging utilization data from the Picture Archiving and Communication Software (PACS) program via a standardized data abstraction form. A multivariate logistic regression analysis was then performed to account for the covariates of admission rate and emergency severity index (ESI) scores between the pre-PIT and PIT cohorts. Adjusted regression coefficients (β) were used to assess differences in imaging utilization.

There were 92,243 visits in the pre-PIT time period and 89,105 in the PIT time period. Overall CT incidence increased from 0.36 studies per ED visit (95%CI 0.36-0.37) in the pre-PIT period to 0.49 in the PIT period (95%CI 0.49-0.50). In multivariable weighted regression adjusting for admit proportion and mean ESI, PIT was associated with an absolute increase of 0.129 CT studies per ED visit (β=0.129, 95% CI 0.096–0.163, p < 0.001), corresponding to 12.9 additional CT studies per 100 ED visits. Significant increases were seen for the subcohorts of abdominal CT, chest CT, and head CT. Increases were also observed for MRI (β=0.003, 95% CI 0.001–0.006, p=0.01) and RUS (β=0.01, 95% CI 0.004–0.015, p=0.002). POCUS demonstrated a negative trend in utilization that did not reach statistical significance.

In our single-center analysis, deployment of PIT was associated with increased use of advanced imaging, most notably CT, and this relationship remained significant in multivariable analysis. Further study is needed to determine if there is a causal role of the PIT process in driving imaging utilization rates.