Predictors of routine HPV-based cervical cancer screening adoption in federally qualified health centers within a multi-state practice research network 2012-2022: retrospective observational study.

Abstract

Cervical cancer screening guidelines in the United States have evolved with the introduction of human papillomavirus (HPV)-based modalities, including co-testing in 2012 and primary HPV testing in 2018, in addition to Pap-only screening. However, adoption of these modalities in federally qualified health centers (FQHCs) has been underexplored. We examined longitudinal patterns of HPV-based cervical cancer screening adoption (e.g., proportion screened with co- and primary HPV testing) across three FQHC systems in Washington and Idaho and explored factors associated with receipt of HPV-based modalities. Using electronic health record data, we identified the first routine cervical cancer screening among 11,316 individuals from 2012-2022 across 25 clinics within three FQHCs. Screening modality at the screening was classified as Pap-only, co-testing, or primary HPV testing. We quantified the proportion of each screening modality by year and FQHC from 2012-2022 and used multivariable logistic regression to estimate associations of factors (age, race, ethnicity, rurality, insurance status, primary care visits, and FQHC) with HPV-based screening for 2012-2018 and 2018-2022 guideline periods. The proportion of up-to-date screening in 2022 was 34.0%, 22.7%, and 37.6% for FQHCs A, B, and C, respectively. Among individuals receiving their first routine screening from 2012-2022, 59.1% underwent Pap-only, 38.8% co-testing, and 2.0% primary HPV testing, respectively. Pap-only was nearly universal in 2012 but declined to 57.0% by 2022, while co-testing increased to 41.0%. Primary HPV testing remained < 7.0% across all years and FQHCs. During the 2012-2018 guideline period, odds of co-testing (vs. Pap-only) were higher among Hispanic (aOR = 1.56; 95% CI, 1.35-1.80) and Black (aOR = 1.64; 95% CI, 1.18-2.28) individuals and screened at FQHC C (aOR = 4.21; 95% CI, 3.41-5.22). Odds were lower among rural (aOR = 0.20; 95% CI, 0.17-0.24) and uninsured patients (aOR = 0.51; 95% CI, 0.44-0.58). From 2018-2022, odds of co-testing were higher among individuals aged 40-59 and Black, lower at FQHC B (aOR = 0.48; 95% CI, 0.37-0.63), and higher at FQHC C (aOR = 1.34; 95% CI, 1.06-1.68). Adoption of HPV-based screening modalities increased over time and varied by patients' rurality, race, and ethnicity. Lower adoption in some organizations may reflect limited capacity/resources for adopting newer screening modalities, which could impact equitable access to screening.