Abstract B005: Using EMR-based SOGI data collection to identify disparities in sexual and gender minority patients with cancer
Sara Burke,Constantine Daskalakis,Khaldoun Hamade,Christopher McNair,Nicole Simone
DOI: https://doi.org/10.1158/1538-7755.disp24-b005
2024-09-23
Cancer Epidemiology Biomarkers & Prevention
Abstract:Purpose/Objectives: Sexual and gender minority (SGM) patients have been suggested to face disparities in cancer screening and outcomes. Despite this, a large barrier to improving care for this population is the lack of sexual orientation and gender identity (SOGI) data available in surveillance, research, and clinical settings. This study aimed to create integrated datasets containing SOGI and clinical data from electronic medical records (EMR) to better characterize SGM cancer patients with the goal of narrowing disparities and improving cancer care by tailoring to the needs of this population. Materials/Methods: The cancer patient population was identified using Thomas Jefferson University's Epic-based registry focused on actively managed cancer patients. Additional data elements were linked to this initial dataset: SOGI data (collected using Epic's SOGI module), general demographics, SDOH data from Epic's SDOH module as well as geospatially inferred SDOH information (social vulnerability index), and molecular diagnostic information. A patient was considered to be part of the SGM community if any of the following criteria were met: (i) their gender identity was listed as other than "male" or "female"; (ii) there was a difference between their sex assigned at birth, their legal sex, and/or their gender identity; (iii) their sexual orientation was marked anything other than "straight." Logistic regression was used to compare the data for the SGM community vs. the non-SGM patients within the dataset. Results: 90,506 patients were identified from the active cancer registry with SOGI data available on 11,219 patients (471 SGM and 10,748 non-SGM patients). The SGM population was more likely to be younger and aged 18-34 (5.5% vs. 1.6%) and aged 35-70 (61.6% v 51.2%), single (33.1% vs. 17.9%), Latinx/Hispanic (6.4% vs. 5.0%), and Asian/Pacific Islander (3.2% vs. 2.5). Adjusted for sex, age, race, ethnicity, marital status, and cancer site, SGM and non-SGM patients did show differences in obesity (39.7% vs 34.4%, p=0.347) or smoking (10.8% vs 8.5%, p= 0.629) rates. SGM patients had a higher frequency of alcohol (45.3% vs 39.7%, p=0.05) and illicit drug use (13.4% vs 7.5%, p=0.014). No difference between SGM and non-SGM patients was noted in SDOH or the prevalence of genomic testing rates. Conclusions: This study demonstrates the feasibility of integrating SOGI data with other clinical data within the EMR so that researchers can understand the specific needs of the SGM cancer population. These data demonstrate that risk factors of cancer such as alcohol and drug use were significantly higher in our SGM population highlighting the need for cancer care navigators to assist in connecting patients to resources specific to their needs. Our study was limited by the small number of patients with SOGI and SDOH data. This highlights the need to increase SOGI training for providers so that a comprehensive evaluation of factors leading to health disparities in the SGM community can be discerned and cancer care can be personalized and improved. Citation Format: Sara Burke, Constantine Daskalakis, Khaldoun Hamade, Christopher McNair, Nicole Simone. Using EMR-based SOGI data collection to identify disparities in sexual and gender minority patients with cancer [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl) nr B005.
oncology,public, environmental & occupational health