Abstract B083: Patterns of HIV testing among New Jersey Medicaid enrollees diagnosed with invasive cervical cancer

Abstract

Abstract Purpose: HIV infection and cervical cancer disproportionately impact low-income and racial/ethnic minorities in urban areas. Few studies have examined factors associated with HIV testing during cancer diagnosis in vulnerable populations. Current National Comprehensive Cancer Network (NCCN) guidelines recommend an HIV test during initial invasive cervical cancer (ICC) workup. We examine factors associated with patterns of HIV testing among Medicaid enrollees diagnosed with ICC in New Jersey. Methods: Using linked data from the New Jersey State Cancer Registry and New Jersey Medicaid claims and enrollment files, we examined patterns of HIV and other STI testing (chlamydia, gonorrhea and syphilis) among nonelderly (ages 21-64) ICC cases diagnosed between 2012 and 2014. We evaluated two HIV testing time periods: at any point during our study period (2011-2014; pre- or post-cancer diagnosis) and during the cancer workup (6 months pre/post ICC diagnosis). Bivariate and multivariable logistic regression models were used to identify sociodemographic, clinical tumor, and area-level factors associated with patterns of HIV testing. Results: A total of 248 ICC Medicaid enrollees were included in the analytic sample, of whom 83 (33%) received an HIV test at any time. A little over a quarter (26.6%) received STI testing at any time, including 21% for chlamydia and gonorrhea testing. Of those who received any HIV testing, almost half (46%) received their HIV testing during the cancer workup. In the adjusted model, women who lacked any STI testing had higher odds of also not receiving an HIV test during initial cancer workup compared with at least one STI test pre/post cancer diagnosis (OR: 4.2; 95% CI: 1.98-8.98). Similarly, women enrolled for less than a full year prediagnosis also had higher odds of not receiving an HIV test compared to those with full-year enrollment (OR: 2.6; 95% CI: 1.02-6.94). The odds of nonreceipt of HIV testing during the cancer workup were lower among Hispanic/NH-API/Other women compared with White women (OR: 0.38; 95% CI: 0.16-0.88) and higher for those with no primary care visits post-diagnosis compared to ≥ 3 PCP visits (OR: 2.6; 95% CI: 1.07-6.53) Area-level factors (median household income and population density) were not associated with nonreceipt of HIV testing. Conclusions: Although ICC is considered an AIDS-defining cancer, more than two-thirds of women diagnosed with ICC in our study population did not receive any HIV test during the study period. Strategies to address missed opportunities for HIV testing at ICC diagnosis for vulnerable populations warrant further exploration. Additional validation of claims and patterns of testing should also be explored. Citation Format: Jennifer K. McGee-Avila, Michelle Doose, Jose Nova, Rizie Kumar, Antoinette M. Stroup, Jennifer Tsui. Patterns of HIV testing among New Jersey Medicaid enrollees diagnosed with invasive cervical cancer [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr B083.