The purpose of this study will be to investigate the relationship between breast imaging and discordant pathology diagnosis during the past five years (i.e. July 1, 2014 - July 1, 2019) for ultrasound-guided biopsies, stereotactic biopsies, and MRI-guided biopsies.
We will obtain this information via a retrospective chart review between July 1, 2014 and July 1, 2019 on mPower, a software database offered by UF Health Jacksonville of radiology studies and reports. Our inclusion criteria includes any ultrasounded, stereotactic, or MRI breast biopsy report that does not report concordance with pathology within the impression or findings of the radiology report for any patient. From these reports we will examine each patient’s imaging from their biopsy, available on either Visage or PACS. We will also need to review the patient’s EPIC chart reviews for their operative reports and pathology reports for a suspected breast lesion during this time frame. The exclusion criteria includes any imaging modality that is not a breast biopsy with ultrasound, stereotactic, or MRI on mPower within the allotted time range or unavailable images. The radiology imaging from this list of patients will be retrospectively reviewed to identify the discrepancy in discordance. This data will be statistically analyzed using SAS statistical software.
The final list of reports that is generated from this mPower search will be exported into an Excel spreadsheet that will be password protected and housed on a UF IT approved server. When we go through the mPower list and search through their medical records on EPIC and review their imaging, the data we need to collect and patient information will be de identified with a unique numeric coding. Each patient will then receive a unique patient deidentifier, and the information gathered from EPIC will be placed in a separate Excel spreadsheet. Data will be stored in an Office of Research Affairs "phile drive", which is only accessible on an UF IT approved server that is password protected and encrypted. The excel sheet will also be password protected, and the unique code will only be distributed to the study staff on this IRB. The code will be destroyed at the end of the chart review.