Abstract
BACKGROUND: Current consensus recommendations are to not initiate cervical cancer screening for immunocompetent adolescent females before 21 years of age. This is in part because of the very low rate of 0.8 per 100,000 new cervical cancer cases diagnosed among women aged between 20 to 24 years. Timely human papillomavirus vaccination further decreases the incidence of cervical cancer to 4 cases per 100,000 persons by the age of 28 years. Screening before 21 years of age has demonstrated no clear benefit in cancer risk reduction or outcomes. In addition, unindicated screening among adolescents can lead to patient harm and increasing costs to the healthcare system. OBJECTIVE: It is important to assess the rates of overutilization of cervical cancer screening and to identify areas where improvements have occurred and where further opportunities exist. This study aimed to assess the trends over time and the practice and provider factors associated with unindicated cervical cancer screening tests in adolescent females within the largest healthcare system in the state. STUDY DESIGN: Cross-sectional data from patients aged 13 to 20 years who underwent cervical cancer screening between January 1, 2012, and December 31, 2018, across a large multihospital health system were reviewed. All cervical cancer screening results were included. The incidence rate of unindicated screening was analyzed over 6-month intervals using the Poisson regression analysis. RESULTS: The study included data from 118 providers and 794 women. Among the 900 screening results, most (90%) were unindicated: 87% with unindicated cytology testing alone and 14% with unindicated human papillomavirus testing. Screening tests were collected from patients aged 13 to 20 years, many of whom had multiple unindicated cytology tests, with 25 patients having ≥3 tests before the age of 21 years. Most results of cytology testing were negative for intraepithelial lesion or malignancy (77%). Moreover, 52 invasive diagnostic or therapeutic procedures (49 colposcopies and 3 conizations) were performed, of which 45 (87%) followed an unindicated screening test. Between 2012 and 2018, the incidence rate of unindicated cytology decreased by 33% (12.6 to 8.5 unindicated cytology per 1000 encounters). The incidence rate of unindicated screening was lower in the academic setting than in the community setting (incidence rate ratio, 0.43; P<.01). Even with decreases in the overall rates of unindicated screening throughout the study period, there were still 58 unindicated screening tests performed in the final year of this study. CONCLUSION: Despite substantial reductions in unindicated screening for women aged <21 years, there remained areas for improvement. Our data reflected practices of guideline nonadherence up to 7 years after the 2012 guideline. Now, with a new series of changes to the guidelines, which may be even more challenging for patients and providers, it is more important than ever to utilize evidence-based strategies to improve guideline dissemination and adherence.