Association between Social Vulnerability Index (SVI) and Cervical Cancer Screening Modality in Community Based Primary Care Clinics Offering In-clinic HPV Self-Sampling

Abstract

Background: Cervical cancer is preventable through routine screening and early detection of precancerous changes. U.S. Food and Drug Administration (FDA) approval of HPV self-collection as a cervical cancer screening option has allowed for another modality of screening. HPV self-collection may help address barriers to screening, especially among Somali women in the U.S. We examined whether neighborhood-level social vulnerability is associated with screening modality choice between HPV self-sampling versus clinician-collected screening among all women and Somali American women. Methods: We conducted a secondary data analysis from the Isbaar hybrid type 2 effectiveness-implementation study, which evaluated the integration of HPV self-sampling as a routine cervical cancer screening option across three primary care clinics in Minneapolis, Minnesota. This secondary analysis included all eligible women between the ages of 30-65 who were due for cervical cancer screening and who initiated screening during the 12-month study period (2023-2024). Separate analyses were conducted among all women and also restricted to Somali American women. The primary exposure was neighborhood-level social vulnerability, which was measured using the Social Vulnerability Index (SVI). Primary analysis evaluated SVI categorized into quartiles (low, low-medium, high-medium, and high). Secondary analyses evaluated SVI subthemes (socioeconomic status, household characteristics, racial and ethnic minority status, and housing type and transportation). The outcome was screening modality, defined as HPV self-sampling or clinician-collected cervical cancer screening. Multivariable logistic regression models were used to estimate the association between SVI or SVI subthemes and screening modality, adjusting for age, clinic site, interpreter needed, time overdue for screening, and (among all women only) Somali ethnicity. Results: Among all women, 282 of 984 (28.7%) completed screening by HPV self-sampling. Of the 303 (30.8%) Somali American women, 110 of 302 (36.4%) completed screening by self-sampling. Among all women, overall, SVI was not associated with self-sampling uptake. Among Somali American women, those in the low-medium SVI category had higher odds of self-sampling than those in the low category (adjusted odds ratio (aOR) = 6.14; 95% CI: 1.26, 21.73; p = 0.004), but associations with the medium and medium-high categories were not significant. In the subtheme analyses, the odds of self-sampling decreased with increasing socioeconomic status vulnerability among both all women and Somali American women, though associations did not reach statistical significance. Among Somali American women, the odds of self-sampling increased with higher racial and ethnic minority status vulnerability (aORs 5.36 to 10.44) and higher housing type and transportation vulnerability (aORs 10.75 to 11.60), with borderline statistically significant associations. Discussion: In this secondary analysis of the Isbaar study, SVI was not associated with screening modality among all women, suggesting that HPV self-sampling was taken up similarly across the levels of vulnerability. Among Somali American women, self-sampling was more common in non-low SVI groups, though not all associations were statistically significant. SVI subthemes patterns were mixed, with some subthemes showing patterns of increased or decreased self-sampling across vulnerability levels. These findings suggest that offering HPV self-sampling as a screening modality in community clinics could be a practical step towards equitable cervical cancer screening.