Understanding and Addressing Barriers to Cervical Cancer Screening: Patient Perspectives and the Impact of HPV Self‑Sampling Interventions

Abstract

In 2025, there were an estimated 13,360 cases of cervical cancer and 4,320 deaths from the disease in the United States. Most cervical cancers are diagnosed in individuals who are unscreened or have never been screened. Cervical cancer screening (CCS) adherence has been declining, with disparities across some sociodemographic groups, including individuals with limited English proficiency (LEP). Common barriers to CCS include lack of a regular provider, high out-of-pocket costs or lack of insurance, low awareness of CCS, and emotional factors including fear and embarrassment; additional barriers amongst those with LEP include lack of familiarity with preventative medicine, complicated scheduling systems, and limited physician time to discuss preventive care due to the use of interpreters. Human papillomavirus self-sampling (HPV-SS) is an emerging strategy that can address common barriers to CCS across subpopulations. Although offering mailed HPV-SS kits is effective for increasing CCS overall, understanding whether specific populations may need targeted outreach is important for supporting equitable implementation. In addition, as HPV-SS expands, a key question is whether non-clinic-based screening may encourage or discourage people from participating in other recommended preventative care by removing the in-person visit. This dissertation aims to understand and address barriers to CCS by exploring patient perspectives and evaluating the impact of HPV-SS interventions. The first chapter explores the barriers and facilitators to CCS and perspectives on HPV-SS among Mandarin-speaking women with LEP, a population with low CCS rates. Twenty interviews were conducted in Seattle, Washington. Using the Socio-Context Framework and Andersen’s Behavioral Model of Health Services Use, several barriers and facilitators to CCS were identified across sociocultural and individual domains. Cultural factors included stigma surrounding sexually transmitted infections, concerns about privacy, and fear of receiving a cancer diagnosis. Social factors included limited discussion of CCS within social networks and experiences of cancer among family members or friends. Individual-level barriers included high medical costs, uncertainty regarding insurance coverage, and difficulties navigating the healthcare system. Overall, participants reported high acceptability of HPV-SS. However, concerns around possible incorrect use and potential sample contamination were raised. The second chapter identified patient characteristics that modify the effectiveness of mailed HPV-SS interventions using data from the STEP trial (Self-Testing options in the Era for Primary HPV screening for cervical cancer), a 2020-2022 pragmatic trial conducted within Kaiser Permanente Washington. Analyses included overdue (n = 2,823) and screening-adherent (n = 9,398) participants randomized to direct-mail HPV-SS, opt-in HPV-SS, or education only. Electronic health records were used to identify baseline patient characteristics and CCS within 6 months after randomization. Poisson regression models with an identity link were used to estimate absolute risk differences (RDs) for CCS completion comparing either the direct-mail or opt-in group with the education group. We conducted a descriptive evaluation of effect modification by estimating intervention effects within levels of each patient characteristic. We found that direct-mail HPV-SS was effective across most subgroups, with some differences observed in the magnitude of effect. These findings suggest that tailoring implementation strategies to patient characteristics could advance equity of CCS. The third chapter used data from STEP to evaluate (1) whether offering mailed HPV-SS kits to individuals overdue for CCS impacts influenza vaccination uptake and (2) whether CCS uptake behavior impacts uptake of influenza vaccination among individuals who are mailed HPV-SS kits. We found that influenza vaccination uptake was similar between the direct-mail and education groups. Within the direct-mail arm, compared to non-screeners, those who screened in-clinic or completed a kit had significantly higher influenza vaccination uptake. Our results suggest that offering home-based HPV-SS does not reduce participation in influenza vaccination uptake among individuals overdue for CCS.