RESUMEN
OBJECTIVE: To evaluate the use of community health agents (CHAs) to instruct women living in poor rural areas in obtaining self-collected cervical samples and compare the high-risk HPV (hrHPV) hybrid capture (HC) results obtained to those for gynecologist-collected samples. METHODS: After a one-day training, CHAs visited sexually active women, instructing each in the use of collection brush and the Universal Collection Medium tube. One week thereafter, a gynecologist collected cervical samples from, and performed colposcopies on, the same women. A single reference lab performed all HCs. RESULTS: 878 women (Age: 15-69 years) participated. Among self-collected samples, hrHPV prevalence was 33.9% (95% CI: 30.8%-37%), compared with 28.6% (95% CI: 27%-30%) among gynecologist-collected samples. However, 9.3% of the patients were HPV HC II-positive in the self-collected sample and HPV HC II-negative in the gynecologist-collected samples (95% CI: 7.38%-11.22%), whereas 4% tested positive in gynecologist-collected samples and negative in self-collected samples (95% CI: 2.7%-5.3%) (P<0.01; kappa=0.7). Of 9 cases of histologically-confirmed, high-grade squamous intraepithelial lesion, self-collected and provider-collected samples missed one each. CONCLUSION: Self-collected vaginal sampling could be made an additional CHA function under existing program conditions, improving access to cervical cancer screening in poor rural settings.