The Impact of the Human Papillomavirus Vaccine on High-Grade Cervical Lesions in Urban and Rural Areas: An Age–Period–Cohort Analysis

Abstract

Disparities in human papillomavirus (HPV) vaccination exist between urban (metropolitan statistical areas (MSAs)) and rural (non-MSAs) regions. To address whether the HPV vaccine’s impact differs by urbanicity, we examined trends in cervical intraepithelial neoplasia grades 2 or 3 and adenocarcinoma in situ (collectively, CIN2+) incidence in MSAs and non-MSAs among Tennessee Medicaid (TennCare)-enrolled women aged 18–39 years and among the subset screened for cervical cancer in Tennessee, United States. Using TennCare claims data, we identified annual age-group-specific (18–20, 21–24, 25–29, 30–34, and 35–39 years) CIN2+ incidence (2008–2018). Joinpoint regression was used to identify trends over time. Age–period–cohort Poisson regression models were used to evaluate age, period, and cohort effects. All analyses were stratified by urbanicity (MSA versus non-MSA). From 2008–2018, 11,243 incident CIN2+ events (7956 in MSAs; 3287 in non-MSAs) were identified among TennCare-enrolled women aged 18–39 years. CIN2+ incident trends (2008–2018) were similar between women in MSAs and non-MSAs, with largest declines among ages 18–20 (MSA average annual percent change (AAPC): −30.4, 95% confidence interval (95%CI): −35.4, −25.0; non-MSA AAPC: −30.9, 95%CI: −36.8, −24.5) and 21–24 years (MSA AAPC: −14.8, 95%CI: −18.1, −11.3; non-MSA AAPC: −15.1, 95%CI: −17.9, −12.2). Significant declines for ages 18–20 years began in 2008 in MSAs compared to 2010 in non-MSAs. Trends were largely driven by age and cohort effects. These patterns were consistent among screened women. Despite evidence of HPV vaccine impact on reducing CIN2+ incidence regardless of urbanicity, significant declines in CIN2+ incidence were delayed in non-MSAs versus MSAs.