Informatics methods for laboratory evaluation of HPV ordering patterns with an example from a nationwide sample in the United States, 2003-2009

BACKGROUND: Laboratory data is a rich source of information that can be used to estimate adherence to physician guidelines and motivate improvement in clinical practice. Human papillomavirus (HPV) testing is an important component of cervical cancer screening programs with established screening guidelines. The purpose of this study was to develop methods to estimate concordance with published guidelines for HPV testing in order to provide clinicians and payors specific feedback about overscreening. METHODS: This retrospective analysis of laboratory test ordering patterns evaluated 454,532 HPV tests ordered from September 2003 to October 2009 from 110 facilities and performed at ARUP laboratories. We used laboratory data including patient demographics, ordering frequency, timestamps and results to examine the proportion of HPV tests ordered on women under 21 years, ordered on women between 21 and 29 years apparently before cytological examination, repeated less than 1 year after a positive HPV result in women over 30 years, and repeated less than 3 years after a negative HPV result in women over 30 years. RESULTS: The absolute number and proportion of HPV tests performed on women under 21 years declined from 20% in 2005 to 5% in October 2009. The proportion of HPV tests performed women between 21 and 29 years also declined during this period. Approximately one-third of HPV tests performed on women between 21 and 29 years arrived for HPV testing before cervical screening had presumably been completed. The most common follow-up intervals for HPV testing on women over 30 years were 6 months following a positive HPV result and 12 months following a negative HPV result. Only 6% of repeat HPV testing in women over 30 years followed a negative HPV result by 3 years or more. Approximately one-fourth of HPV tests ordered the year ending October 2009 were unnecessary based on the American Society for Colposcopy and Cervical Pathology guideline. CONCLUSIONS: We demonstrate simple methods to evaluate appropriate utilization of HPV testing using laboratory data. Our data illustrates that some aspects of HPV test ordering have become more consistent with guidelines over time. However, a large portion of HPV testing in the United States is unnecessary. This highlights opportunities for optimization of a rational cancer prevention strategy to reduce unnecessary screening, colposcopy and biopsies.