Abstract
IMPORTANCE: Cervical screening guidelines in the US recommend that most females can exit routine screening at age 65 years following 2 recent consecutive negative cotest results (concurrent human papillomavirus and cytology tests). However, empirical data on the subsequent risks of cancer and cancer death in this subgroup of females are limited. OBJECTIVE: To estimate the risks of cervical cancer and cervical cancer death among females who meet the cotesting criteria to exit screening. DESIGN, SETTING, AND PARTICIPANTS: In this decision analytical comparative modeling study, 4 decision analytical models from the Cancer Intervention and Surveillance Modeling Network-Cervical modeling consortium that fit common US epidemiological data targets were validated against published estimates of 3- and 5-year risks of cervical intraepithelial neoplasia grade 3 (CIN3) among females meeting exit criteria at Kaiser Permanente Northern California (KPNC). MAIN OUTCOMES AND MEASURES: Age-conditional and cumulative risks of cervical cancer and cervical cancer death at ages 65, 70, 75, 80, and 85 years were estimated by performing a comparative modeling analysis of the 4 models to estimate the risks of cervical cancer and cervical cancer death after exiting screening. RESULTS: All models estimated a 5-year risk of CIN3 that was within the range of empirical data from KPNC. Projections of the cumulative and age-conditional risks of cervical cancer and cervical cancer death increased with time since exiting screening. The cumulative risks of cervical cancer and cervical cancer death by age 70 years were estimated to range from 0.001% to 0.003% and from 0% to 0.001%, respectively. The cumulative risks of cervical cancer and cervical cancer death by age 85 years ranged from 0.026% to 0.081% and from 0.005% to 0.038%, respectively, across models. Results were sensitive to assumptions about screening test sensitivity and incidence of high-risk human papillomavirus. CONCLUSIONS AND RELEVANCE: In this decision analytical comparative modeling study, a low risk of cervical cancer and cervical cancer death was estimated among females who fulfilled the US criteria to exit screening with cotesting; however, the risks increased with age and/or time since screening exit. The findings suggest that future guidelines should consider acceptable risk levels when defining screening modality and exit age requirements.