374 - Delays in medically necessary imaging tests and procedures for hospitalized children in Ontario

Publication Number: 4366.374

Peter J. Gill, The Hospital For Sick Children, Toronto, ON, Canada; Farhana Islam, The Hospital for Sick Children, Toronto, ON, Canada; Julia Orkin, The Hospital for Sick Children, Toronto, ON, Canada; Leo Hersi, The Hospital for Sick Children, Toronto, ON, Canada; Lauren Lapointe-Shaw, University of Toronto, Toronto, ON, Canada; Surain B. Roberts, Unity Health Toronto, Toronto, ON, Canada; Amol Verma, University of Toronto, Toronto, ON, Canada; Fahad Razak, University of Toronto Temerty Faculty of Medicine, Toronto, ON, Canada; Sanjay Mahant, The Hospital for Sick Children, Toronto, ON, Canada

Background: Delays in medically necessary imaging can prolong hospital stays, increase costs, and strain limited resources, particularly on weekends, holidays, and during seasonal surges. In children, some specialized imaging requires transfer to specialized centers for pediatric expertise or anesthesia. However, knowledge about the impacts of wait times for imaging procedures on length of stay for hospitalized children is limited.

Objective: We aimed to examine factors contributing to imaging delays and its impact on hospital length of stay for hospitalized children.

Design/Methods: We conducted a multicenter retrospective cohort study across 17 hospitals in Ontario, Canada, between April 2015 and March 2024. We included all pediatric hospitalizations (17 years and under) with at least one CT, MRI, or ultrasound. Our primary outcomes were time-to-test (time elapsed between ordering and completion) and length-of-stay after test ordering (time elapsed from test ordering to discharge). We will conduct binomial mixed-effects models to evaluate the associations between baseline patient- and hospital-level factors and time-to-test, as well as between time-to-test and length-of-stay.

Results: The study included 14,197 pediatric hospitalizations with at least one CT, MRI, or ultrasound across 15 community and 2 children’s hospitals. Overall, 12%, 13%, and 80% of hospitalizations had at least one CT, MRI, and ultrasound orders, respectively. Median time-to-test was shortest for CT at 2.1 hours (IQR: 0.8–6.1), compared with 4.1 hours (IQR: 1.8–24.5) for MRI and 6.3 hours (IQR: 1.7–24.8) for ultrasound. MRI and ultrasound experienced longer delays outside regular business hours (Figure 1): for MRI, median time-to-test rose from 3.5 hours (IQR: 1.7–21.3) on weekdays to 10.0 hours (IQR: 21.7–34.7) on weekends; for Ultrasound, from 4.3 hours (IQR: 1.5–24.0) to 10.5 hours (IQR: 13.3–29.9). Delays were greatest for infants ( < 1 year), with MRI median time-to-test increasing from 4.8 hours (IQR: 2.6–32.8) on weekdays to 19.6 hours (IQR: 2.9–51.9) on weekends, and Ultrasound times rising from 7.0 hours (IQR: 1.9–46.4) to 15.8 hours (IQR: 2.6–37.0).

Conclusion(s): MRI and ultrasound experience substantial off-hour delays, particularly for infants under one year. This is the first health system-wide investigation of determinants associated with imaging delays in hospitalized children. By identifying potentially modifiable factors, these findings will inform targeted improvements in hospital-level processes to optimize length of stay for pediatric inpatients.

Figure 1. Unadjusted Time-to-Test for Imaging Modalities by Order Timing and Age Category.
pas_figure.jpegThis figure illustrates the unadjusted time from order to test (in hours) across different imaging modalities (CT, MRI, Ultrasound), stratified by patient age category and the timing of the order (Weekday Daytime, Night, Weekend). MRIs have the longest unadjusted time-to-test, with significant delays observed for orders placed at night or on weekends, especially for children under 5 years old.

Note: Night is defined as before 8 AM or after 5 PM. Weekend is defined as Friday evening (5 PM or later), Saturday, Sunday, or early Monday morning (before 8 AM). Weekday Daytime is all other times that is not Night or Weekend. Each box plot represents the interquartile range (25th to 75th percentile), with the horizontal line indicating the median. Red dots mark the mean time-to-test for each group.