The Effects of Race, Primary Language, Insurance and Other Factors on Time to Pediatric Outpatient MRI Completion:...

Disparities in healthcare access in the United States have been associated with race and ethnicity, as well as socioeconomic factors. Because delays in imaging may result in delayed diagnosis or clinical management, we are evaluating practices within our radiology department in hopes of decreasing disparities in access to imaging. The objective of this study is to determine the disparities in time to outpatient MRI scheduling and completion by race, ethnicity, primary language, socioeconomic status, insurance and other factors at a tertiary children's hospital.

After Institutional Review Board approval, we retrospectively extracted data from all outpatient MRI exams completed at our center between 10/5/2020 and 8/31/2022. Collected data included sex, age, race/ethnicity, primary language, medical complexity, insurance type, address, need for anesthesia, ordering specialty, and order acuity. We determined times to MRI scheduling or completion using mixed effects Cox regression models and determined associations between unadjusted and fully adjusted models.

We analyzed 14,002 completed outpatient MRI orders from 9714 unique patients. 56.2% were White, 19.2% Hispanic, 8.4% Asian, 4.5% Black/African-American, 1.4% American Indian/Alaska Native, 0.7% Native Hawaiian/Pacific Islander, 5.7% two or more races/ethnicities, and 3.8% "Other." In fully adjusted models, there was no significant association between race/ethnicity and time to MRI scheduling and completion. In fully adjusted models, time to completion of MRI was slower among those with Medicaid (adjusted hazard ratio [95% confidence interval] of 0.92 [0.87, 0.98]), a primary language other than English (0.90 [0.82, 0.99]), non-complex chronic illness (0.72 [0.67, 0.79]), complex chronic illness (0.72 [0.67, 0.78]) and need for anesthesia (0.75 [0.71, 0.79]).

At our tertiary children's hospital, time to completion of outpatient MRI was not associated with race, but was greater among those with Medicaid insurance, whose primary language was not English, and needing anesthesia. Advocating for faster prior authorization by Medicaid, utilizing our hospital's live interpreter phone number for scheduling, and incorporating greater child life support to decrease anesthesia use are considerations for decreasing these disparities, although surveying patients and families most impacted by these discrepancies will be important to identify the most promising interventions.

Data are not publicly available to preserve individuals' privacy due to IRB restrictions. Data may be available upon reasonable request by contacting the corresponding author.