2023 HSR&D/QUERI National Conference

1051 — Abnormal cervical cancer screening results among Veteran and Non-Veteran participants in the NHIS, 2010-2018

Lead/Presenter: Elisheva Danan,  COIN - Minneapolis
All Authors: Danan ER (Minneapolis VA, CCDOR), Than CT (VA Greater Los Angeles) Chawla N (VA Greater Los Angeles) Hoggatt KJ (San Francisco VA) Yano EM (VA Greater Los Angeles)

Objectives:
Effective cervical cancer prevention programs require timely, evidence-based management of abnormal Pap and human papillomavirus (HPV) screening results. In early 2020, VA’s Office of the Inspector General cited 48% of inspected VA facilities for inadequate follow-up of abnormal cervical cancer screening results. Previous estimates suggested that lifetime prevalence of an abnormal cervical cancer screening result may be up to three times as high among Veterans compared to non-Veterans, but disparate data sources limit direct comparisons. Using data from the National Health Interview Survey (NHIS), we tested the hypothesis that Veterans have a higher prevalence of recent abnormal cervical cancer screening results than non-Veterans.

Methods:
We pooled NHIS data from 2010, 2015, and 2018, then restricted the sample to female participants with a cervical cancer screening test in the prior 3 years. The primary outcome was self-reported abnormal result on a Pap and/or HPV test in the prior 3 years. Predictive variables examined included Veteran status, as well as sociodemographic and health factors previously associated with abnormal screening results. Sociodemographics included age, race, geographic region, marital status, income, insurance, VA coverage, healthcare utilization, and sexual orientation; health factors included smoking, hysterectomy, receipt of HPV vaccine, and time since last screening test. We compared sociodemographic and health factors between Veterans and non-Veterans. We then used survey-weighted multivariable logistic regression to estimate the odds of an abnormal screening result in the past 3 years as a function of Veteran status, controlling first for age and survey year, then adding sociodemographic and health factors in subsequent models.

Results:
The pooled sample included 468 Veterans and 28,947 non-Veterans; both groups had a mean age 45 years, and were majority non-Hispanic White (69.7% of Veterans and 64.9% of non-Veterans). Compared to non-Veterans, Veterans were more likely to be unmarried, have public insurance, have VA coverage, attend >10 clinic visits per year, be a current or former smoker, have had a hysterectomy, and have received an HPV vaccine. Overall, 17.9% of Veterans and 13.6% of non-Veterans reported an abnormal cervical cancer screening test in the past 3 years (unadjusted p = 0.053). The odds of an abnormal screening result among Veterans compared to non-Veterans was 1.37 (CI 0.98-1.90). Adding sociodemographics and health factors to the model attenuated the difference between Veterans and non-Veterans (aOR 1.1, 0.74-1.64). In the final model, survey year, being unmarried, having any public insurance, >10 clinic visits per year, ever smoking, hysterectomy, and HPV vaccination were all significantly associated with higher odds of abnormal screening results; while age >45, income >400%FPL, and last Pap greater than 1 year ago were associated with lower odds of an abnormal result.

Implications:
Nearly 1 in 5 Veterans screened for cervical cancer in the past 3 years received an abnormal result. Abnormal screening result prevalence was not statistically higher for Veterans than non-Veterans, but Veterans more often reported both modifiable and non-modifiable risk factors for abnormal screens.

Impacts:
Given the relatively high prevalence of abnormal screening results among Veterans, cervical cancer prevention programs must prioritize results communication and timely, evidence-based follow-up for patients with an abnormal test. Attention to modifiable risk factors, such as smoking cessation, can also help reduce Veterans’ cervical cancer risk.