Assessment of Pediatric Radiation Exposure from Computed Tomography

Project Summary. The objective of this study is to identify pediatric cohorts with the highest levels of radiation exposure from CT scanning by merging EMR data elements with an automated method for calculating cumulative, patient level radiation exposure. The results of this study will serve as preliminary data for a PEDSnet application to PCORI. The goal of which would be to (1) study this process prospectively in additional PCORnet networks that treat both children and adults and; (2) develop and implement multi-institutional efforts to reduce CT scanning in high risk cohorts through partnership with networks such as Image Gently or Solutions for Patient Safety.

Study Design. This is an observational study. It will include all pediatric patients aged 0 to 15 years who receive CT scans of the head, chest, abdomen, and pelvis. 

  • The retrospective data will be pulled from both Digital Imaging and Communications in Medicine (DICOM) files or similar types of files via PACS, and EMR data from January 1, 2012 - December 31, 2014 at all participating PEDSnet institutions. 
  • Information will be automatically abstracted from the imaging file as well as their electronic medical record (EMR) and sent to a secure research workstation.
Engagement. Aim 3 of this study is to develop two groups of stakeholders that can be used to help guide how the preliminary data should be further studied and applied as clinical interventions.

Health care consumers: We plan to engage with the families of patients who had multiple CT scans at one of the PEDSnet hospitals during 2012-2014. We have submitted an IRB application at our own institution that allows us to solicit engagement from a group of “high outliers” at NCH so that we can develop a relationship with this group over the next 6 months leading up to the onset of this study. This group can then be expanded to additional institutions and patient groups as appropriate. Additional non-patient/family stakeholders will include the Chief Medical Officer from Partners for Kids, the Medicaid accountable care organization administered through NCH, “high prescribing physician groups” such as Emergency Medicine physicians, and an external radiologist(s). Projected broad content areas for discussion with this group will include:

·         The layperson’s understanding of risks associated with medical radiation exposure from imaging

·         How to conceptualize a risk-benefit strategy for shared decision-making in medical imaging

·         Development of a patient friendly tool that educates about radiation exposure risk

·         Opinions on the value of developing a cumulative longitudinal tracker

·         Feedback on potential medical homes for this cumulative data

Thought leaders: An additional stakeholder group consisting of medical physicists, ethicists, patient advocacy groups, and potentially other high impact thought leaders will be assembled in order to present the study results as substrate for a discussion related to the above bulleted points as well as development of future interventions for selected populations.

Data Elements. Data elements collected will include clinical information and imaging data.

Clinical information (from the EMR):

·         Age at admission

·         Gestational age

·         Gender

·         Race/Ethnicity

·         Primary Payer

·         Encounter Specific Associated diagnoses (All ICD-9-CM diagnosis codes associated with the admission)

-        All Diagnoses from Prior year dated from 12 months from scan date

·         Procedures Performed (All CPT-4, ICD-9-CM, or HCPCS procedure codes associated with the encounter)

-       All Procedures from prior year dated from 12 months from scan date

·         Dates on which each listed procedure was performed

·         Admission date

·         Discharge date

·         BMI

Imaging data (from PACS and DICOM or equivalent imaging file):

·         Date of acquisition of the image

·         Study description/type

·         Patient’s date of birth

·         Patient’s gender

·         X-ray tube current (in mA)

·         X-ray exposure time (in seconds)

·         X-ray tube potential (in kVp)

·         CT Dose Index (CTDIvol)


Collaborator Institutions