Co-Reporting of HPV and Pap Cytology Results: Informatics Experience at a Large Academic Women's Hospital
Christopher Gilbert, Karen Lassige, Jonee Matsko, Nancy Mauser, R Marshall Austin, Anil Parwani, Chengquan Zhao, Anthony Piccoli, Liron Pantanowitz. UPMC, Pittsburgh, PA
Background: Human papillomavirus (HPV) DNA tests have been FDA-approved for co-testing in women 30 and older and as a reflex test for women with ASC abnormalities. Reporting Pap cytology diagnoses along with HPV test results in these circumstances has posed significant informatics challenges for cytology laboratories. The aim of this study is to describe our experience adding new functionality to our anatomic pathology laboratory information system (LIS) to provide this capability.
Design: Changes to our LIS (CoPath ver. 3.2, Cerner) and workflow were implemented in May 2005. Cytotechnologists select the HPV request type in the LIS during accessioning. Cases with pending or reflex HPV tests are tracked by the LIS, providing visual alerts to pathologists prior to sign-out. The LIS generates barcoded labels and a log of specimens designated for testing (Digene HC2). In the case of reflex testing, this occurs after the assignment of an ASC diagnosis by the pathologist. HPV test results are entered into the LIS by cytotechnologists using barcoded diagnostic quick-texts and combined with the cytology interpretation into a single report prior to sign-out. To evaluate the effect of the LIS changes we compared various parameters before and after implementation (2003-2010). Clinician satisfaction was solicited in an informal manner.
Results: Integration of HPV testing and reporting into the LIS has enabled cytotechnologists to order and perform testing under the supervision of microbiology staff, aided the interpretation of cases, and provided a shift in revenue to cytology. It has been particularly well received by our clinicians who desired a single consolidated report. HPV requests increased 60% over the time period. The LIS and related workflow changes had no significant effect on Pap turn-around time, but were associated with a 3% absolute increase in the number of ASC-US results. There was no significant change in pap volume, and no significant change in ASC-H or other abnormal results.
Conclusions: Modifications to our LIS functionality were required to handle new demands for HPV testing. LIS-generated barcodes, logs, and alerts helped automate the composition of combined reports. After implementation, there was an increase in ASC-US diagnoses, but this may have been due to a new laboratory policy of targeted educational re-screening of negative Pap tests preceding all histopathologic CIN2+ diagnoses also begun in 2005. There was no delay in turn-around-time despite a significant increase in HPV test volume.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 228, Tuesday Afternoon